[{"page": 1, "text": "CPPTRAJ\nDaniel R. Roe\nJanuary 27, 2026\nhttps://github.com/Amber-MD/cpptraj\nContents\n1 Introduction 8\n1.1 Manual Syntax Format. . . . . . . . . . . . . . . . . . . . . . . . 9\n1.2 Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9\n1.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9\n2 Running Cpptraj 9\n2.1 Command Line Syntax . . . . . . . . . . . . . . . . . . . . . . . . 9\n2.2 Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11\n2.3 Getting Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12\n2.4 Batch mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12\n2.5 Interactive mode . . . . . . . . . . . . . . . . . . . . . . . . . . . 12\n2.6 Trajectory Processing “Run” . . . . . . . . . . . . . . . . . . . . . 13\n2.6.1 Actions and multiple topologies . . . . . . . . . . . . . . . 13\n2.7 Parallelization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13\n2.7.1 MPI Trajectory Parallelization . . . . . . . . . . . . . . . 13\n2.7.2 OpenMP Parallelization . . . . . . . . . . . . . . . . . . . 14\n2.7.3 CUDA Parallelization . . . . . . . . . . . . . . . . . . . . 15\n3 General Concepts 15\n3.1 Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15\n3.2 Atom Mask Selection Syntax . . . . . . . . . . . . . . . . . . . . 16\n3.3 Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18\n3.4 Parameter/Reference Tagging . . . . . . . . . . . . . . . . . . . . 18\n4 Variables and Control Structures 19\n4.1 for . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20\n4.2 set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21\n4.3 show . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22\n1"}, {"page": 2, "text": "5 Data Sets and Data Files 23\n5.1 Data Set Selection Syntax . . . . . . . . . . . . . . . . . . . . . . 25\n5.2 Data Set Math . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25\n6 Data File Options 27\n6.1 Standard Data File Options . . . . . . . . . . . . . . . . . . . . . 28\n6.2 Grace Data File Options . . . . . . . . . . . . . . . . . . . . . . . 31\n6.3 Gnuplot Data File Options . . . . . . . . . . . . . . . . . . . . . 31\n6.4 Amber REM Log Options . . . . . . . . . . . . . . . . . . . . . . 32\n6.5 Amber MDOUT Options . . . . . . . . . . . . . . . . . . . . . . 32\n6.6 Evecs File Options . . . . . . . . . . . . . . . . . . . . . . . . . . 33\n6.7 Vector psuedo-traj Options . . . . . . . . . . . . . . . . . . . . . 33\n6.8 OpenDX file options . . . . . . . . . . . . . . . . . . . . . . . . . 33\n6.9 CCP4 file options . . . . . . . . . . . . . . . . . . . . . . . . . . . 34\n6.10 Charmm REPD log options . . . . . . . . . . . . . . . . . . . . . 34\n6.11 Amber Constant pH Out options . . . . . . . . . . . . . . . . . . 34\n7 Coordinates (COORDS) Data Set Commands 34\n7.1 catcrd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36\n7.2 combinecrd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36\n7.3 crdaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37\n7.4 crdout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37\n7.5 crdtransform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37\n7.6 createcrd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38\n7.7 emin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38\n7.8 extendedcomp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40\n7.9 graft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40\n7.10 loadcrd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41\n7.11 loadtraj . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42\n7.12 permutedihedrals . . . . . . . . . . . . . . . . . . . . . . . . . . . 42\n7.13 prepareforleap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44\n7.14 reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48\n7.15 rotatedihedral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48\n7.16 sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49\n7.17 splitcoords. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50\n7.18 zmatrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50\n8 General Commands 51\n8.1 activeref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52\n8.2 calc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53\n8.3 clear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53\n8.4 create . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53\n8.5 createset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53\n8.6 datafile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54\n8.7 datafilter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54\n8.8 dataset. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55\n2"}, {"page": 3, "text": "8.9 debug | prnlev. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59\n8.10 ensextension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59\n8.11 exit | quit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59\n8.12 flatten . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59\n8.13 go | run . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60\n8.14 help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60\n8.15 list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61\n8.16 noexitonerror . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61\n8.17 noprogress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61\n8.18 parallelanalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61\n8.19 parsedata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62\n8.20 precision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63\n8.21 random . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63\n8.22 readdata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64\n8.23 readensembledata . . . . . . . . . . . . . . . . . . . . . . . . . . . 65\n8.24 readinput . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65\n8.25 removedata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65\n8.26 rst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65\n8.27 runanalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66\n8.28 select. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67\n8.29 selectds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67\n8.30 sortensembledata . . . . . . . . . . . . . . . . . . . . . . . . . . . 67\n8.31 usediskcache. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67\n8.32 write | writedata . . . . . . . . . . . . . . . . . . . . . . . . . . . 67\n8.33 System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . 68\n9 Topology File Commands 68\n9.1 angleinfo | angles | printangles. . . . . . . . . . . . . . . . . . . . 69\n9.2 atominfo | atoms | printatoms . . . . . . . . . . . . . . . . . . . . 70\n9.3 bondinfo | bonds | printbonds . . . . . . . . . . . . . . . . . . . . 70\n9.4 bondparminfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71\n9.5 change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71\n9.6 charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73\n9.7 comparetop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73\n9.8 dihedralinfo | dihedrals | printdihedrals . . . . . . . . . . . . . . . 74\n9.9 hmassrepartition . . . . . . . . . . . . . . . . . . . . . . . . . . . 74\n9.10 improperinfo | impropers | printimpropers . . . . . . . . . . . . . 75\n9.11 mass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75\n9.12 molinfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76\n9.13 parm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76\n9.13.1 PDB format: . . . . . . . . . . . . . . . . . . . . . . . . . 77\n9.13.2 Charmm PSF: . . . . . . . . . . . . . . . . . . . . . . . . 78\n9.13.3 Gromacs Top . . . . . . . . . . . . . . . . . . . . . . . . . 78\n9.14 parmbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78\n9.15 parminfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79\n9.16 parmstrip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79\n3"}, {"page": 4, "text": "9.17 parmwrite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79\n9.17.1 Amber Topology . . . . . . . . . . . . . . . . . . . . . . . 80\n9.17.2 Charmm PSF . . . . . . . . . . . . . . . . . . . . . . . . . 80\n9.18 printub | ubinfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80\n9.19 resinfo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81\n9.20 scaledihedralk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81\n9.21 solvent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82\n9.22 updateparameters . . . . . . . . . . . . . . . . . . . . . . . . . . 82\n10 Trajectory File Commands 82\n10.1 ensemble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84\n10.2 ensemblesize. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85\n10.3 reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86\n10.4 trajin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87\n10.4.1 Options for Amber NetCDF, Amber NC Restart, Amber\nRestart: . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89\n10.4.2 Options for CHARMM DCD: . . . . . . . . . . . . . . . . 89\n10.4.3 Options for PDB files: . . . . . . . . . . . . . . . . . . . . 90\n10.5 trajout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90\n10.5.1 Options for pdb format . . . . . . . . . . . . . . . . . . . 92\n10.5.2 Options for Amber ASCII format: . . . . . . . . . . . . . 93\n10.5.3 Options for Amber NetCDF format: . . . . . . . . . . . . 94\n10.5.4 Options for Amber Restart/NetCDF Restart format: . . . 94\n10.5.5 Options for CHARMM COORdinates: . . . . . . . . . . . 94\n10.5.6 Options for CHARMM DCD: . . . . . . . . . . . . . . . . 95\n10.5.7 Options for GROMACS TRX/XTC format: . . . . . . . . 95\n10.5.8 Options for mol2 format: . . . . . . . . . . . . . . . . . . 95\n10.5.9 Options for SQM input format: . . . . . . . . . . . . . . . 96\n10.5.10Options for XYZ format: . . . . . . . . . . . . . . . . . . 96\n11 Action Commands 96\n11.1 addatom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101\n11.2 align . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101\n11.3 angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102\n11.4 areapermol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103\n11.5 atomiccorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103\n11.6 atomicfluct | rmsf . . . . . . . . . . . . . . . . . . . . . . . . . . . 104\n11.7 atommap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106\n11.8 autoimage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107\n11.9 average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109\n11.10avgbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110\n11.11avgcoord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110\n11.12bounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110\n11.13box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111\n11.14center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112\n11.15check | checkoverlap | checkstructure . . . . . . . . . . . . . . . . 113\n4"}, {"page": 5, "text": "11.16checkchirality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114\n11.17closest | closestwaters. . . . . . . . . . . . . . . . . . . . . . . . . 115\n11.18cluster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116\n11.19clusterdihedral . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116\n11.20contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117\n11.21createcrd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118\n11.22createreservoir . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118\n11.23density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118\n11.24diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120\n11.25dihedral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122\n11.26dihedralrms | dihrms . . . . . . . . . . . . . . . . . . . . . . . . . 123\n11.27dihedralscan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124\n11.28dipole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124\n11.29distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125\n11.30drms | drmsd (distance RMSD) . . . . . . . . . . . . . . . . . . . 126\n11.31dssp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127\n11.32enedecomp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127\n11.33energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129\n11.34esander . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132\n11.35filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133\n11.36fixatomorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134\n11.37fiximagedbonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135\n11.38gist (Grid Inhomogeneous Solvation Theory) . . . . . . . . . . . 136\n11.39grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149\n11.40hbond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152\n11.41image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157\n11.42jcoupling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158\n11.43keep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159\n11.44lessplit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161\n11.45lie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161\n11.46lipidorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161\n11.47lipidscd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163\n11.48makestructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163\n11.49mask . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166\n11.50matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167\n11.51mindist/maxdist . . . . . . . . . . . . . . . . . . . . . . . . . . . 169\n11.52minimage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169\n11.53molsurf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170\n11.54multidihedral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171\n11.55multipucker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172\n11.56multivector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173\n11.57nastruct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174\n11.58nativecontacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180\n11.59outtraj . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184\n11.60pairdist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184\n11.61pairwise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185\n5"}, {"page": 6, "text": "11.62principal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186\n11.63projection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187\n11.64pucker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188\n11.65radgyr | rog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189\n11.66radial | rdf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190\n11.67randomizeions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192\n11.68remap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192\n11.69replicatecell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193\n11.70rms | rmsd. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194\n11.71rms2d | 2drms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197\n11.72rmsavgcorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197\n11.73rmsf | atomicfluct . . . . . . . . . . . . . . . . . . . . . . . . . . . 197\n11.74rotate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197\n11.75rotdif. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199\n11.76runavg | runningaverage . . . . . . . . . . . . . . . . . . . . . . . 199\n11.77scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199\n11.78secstruct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199\n11.79setvelocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203\n11.80spam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204\n11.81stfcdiffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205\n11.82strip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206\n11.83surf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207\n11.84symmrmsd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208\n11.85temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209\n11.86time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210\n11.87tordiff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210\n11.88trans | translate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211\n11.89unstrip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212\n11.90unwrap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212\n11.91vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214\n11.92velocityautocorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216\n11.93volmap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217\n11.94volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218\n11.95watershell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219\n11.96xtalsymm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219\n12 Analysis Commands 221\n12.1 autocorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223\n12.2 avg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224\n12.3 calcdiffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225\n12.4 calcstate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226\n12.5 cluster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229\n12.6 cphstats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242\n12.7 corr | correlationcoe . . . . . . . . . . . . . . . . . . . . . . . . . 243\n12.8 crank | crankshaft . . . . . . . . . . . . . . . . . . . . . . . . . . 244\n12.9 crdfluct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244\n6"}, {"page": 7, "text": "12.10crosscorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245\n12.11curvefit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245\n12.12diagmatrix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246\n12.13divergence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249\n12.14evalplateau . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249\n12.15fft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250\n12.16hausdorff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250\n12.17hist | histogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252\n12.18integrate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254\n12.19ired . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254\n12.20kde . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256\n12.21lifetime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257\n12.22lowestcurve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260\n12.23meltcurve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260\n12.24modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260\n12.25multicurve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263\n12.26multihist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264\n12.27phipsi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265\n12.28projectdata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266\n12.29regress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266\n12.30remlog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267\n12.31rms2d | 2drms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268\n12.32rmsavgcorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269\n12.33rotdif. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271\n12.34runningavg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274\n12.35slope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275\n12.36spline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275\n12.37statistics | stat . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276\n12.37.1Torsion Analysis . . . . . . . . . . . . . . . . . . . . . . . 277\n12.37.2Distance Analysis . . . . . . . . . . . . . . . . . . . . . . 277\n12.37.3Pucker Analysis. . . . . . . . . . . . . . . . . . . . . . . . 277\n12.38ti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278\n12.39tica. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279\n12.40timecorr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280\n12.41vectormath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281\n12.42wavelet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282\n13 Analysis Examples 286\n13.1 Cartesian covariance matrix calculation and projection (PCA). . 286\n13.2 Dihedral covariance matrix calculation and projection for back-\nbone phi/psi (PCA) . . . . . . . . . . . . . . . . . . . . . . . . . 286\n7"}, {"page": 8, "text": "1 Introduction\nCpptraj[1](thesuccessortoptraj)isthemainprograminAmberforprocessing\ncoordinatetrajectoriesanddatafiles. Cpptraj hasawiderangeoffunctionality,\nand makes use of OpenMP/MPI to speed up many calculations, including pro-\ncessingensemblesoftrajectoriesand/orconductingmultipleanalysesinparallel\nwith MPI.[2]\nHere are several notable features of cpptraj:\n1. Trajectories with different topologies can be processed in the same run.\n2. Several actions/analyses in cpptraj are OpenMP parallelized; see section\n2.7.2 for more details.\n3. Trajectory and ensemble reads can be MPI parallelized.\n4. Almostanyfilereadorwrittenbycpptraj canbecompressed(withtheex-\nceptionoftheNetCDFtrajectoryformat). Soforexamplegzipped/bzipped\ntopology files can be read, and data files can be written out as gzip/bzip2\nfiles. Compression is detected automatically when reading, and is deter-\nmined by the filename extension (.gz and .bz2 respectively) on writing.\n5. Theformatofoutputdatafilescanbespecifiedbyextension. Forexample,\ndata files can be written in xmgrace format if the filename given has a\n’.agr’ extension. A trajectory can be written in DCD format if the ’.dcd’\nextension is used.\n6. Multiple output trajectories can be specified, and can be written during\naction processing (as opposed to only after) via the outtraj command.\nInaddition, outputfilescanbedirectedtowriteonlyspecificframesfrom\nthe input trajectories.\n7. Multiple reference structures can be specified. Specific frames from tra-\njectories may be used as a reference structure.\n8. The rmsd action allows specification of a separate mask for the reference\nstructure. In addition, per-residue RMSD can be calculated easily.\n9. Actions that modify coordinates and topology such as the strip/closest\nactionscanoftenwriteanaccompanyingfully-functionalstrippedtopology\nfile.\n10. Usersusuallyareabletofine-tunetheoutputformatofdatafilesdeclared\nin actions using the “out” keyword (for example, the precision of the\nnumbers can be changed). In addition, users can control which data sets\nare written to which files (e.g. if two actions specify the same data file\nwith the ’out’ keyword, data from both actions will be written to that\ndata file).\n8"}, {"page": 9, "text": "11. Userscanmanipulatedatasetsusingmathematicalexpressions(withsome\nlimitations), see 5.2 on page 25 for details.\n12. There is some support for creating internal loops over e.g. mask expres-\nsions and setting internal variables (see for, set, and show commands).\nSeetheREADME.mdfileinthecpptraj homedirectoryforinformationonhow\nto build, authors, and so on.\n1.1 Manual Syntax Format\nThe syntax presented in this manual uses the following conventions:\n<> Denotes a variable.\n[] Denotes something is optional.\n{|} Denotes several choices separated by the ’|’ character; one of the choices\nmust be specified.\n... Denotes the preceding option can be repeated.\nEverything else is as printed.\n1.2 Installation\nSee instructions in the CPPTRAJ GitHub repository README.md file under\n’Installation & Testing’: https://github.com/Amber-MD/cpptraj\n1.3 Examples\nSome examples of running CPPTRAJ are available in the examples subdirec-\ntory. There are also many tests in the test subdirectory which can serve as\nsimple examples.\n2 Running Cpptraj\nCpptraj can be run in either “interactive mode” or in “batch mode”.\n2.1 Command Line Syntax\ncpptraj [-p <Top0>] [-i <Input0>] [-y <trajin>] [-x <trajout>]\n[-ya <args>] [-xa <args>] [<file>]\n[-c <reference>] [-d <datain>] [-w <dataout>] [-o <output>]\n[-h | --help] [-V | --version] [--defines] [-debug <#>]\n[--interactive] [--log <logfile>] [-tl]\n[-ms <mask>] [-mr <mask>] [--mask <mask>] [--resmask <mask>]\n9"}, {"page": 10, "text": "[--rng {marsaglia|stdlib|mt|pcg32|xo128}] [--charge <mask>]\n* denotes a flag may be specified multiple times.\n-p <Top0>* Load <Top0> as a topology file.\n-i <Input0>* Read input from <Input0>.\n-y <trajin>* Read from trajectory file <trajin>; same\nas input ’trajin <trajin>’.\n-x <trajout>* Write trajectory file <trajout>; same as\ninput ’trajout <trajout>’.\n-ya <args>* Input trajectory file arguments.\n-xa <args>* Output trajectory file arguments.\n<file>* A topology, input trajectory, or file\ncontaining cpptraj input.\n-c <reference>* Read <reference> as reference\ncoordinates; same as input ’reference <reference>’.\n-d <datain>* Read data in from file <datain> (’readdata\n<datain>’).\n-w <dataout> Write data from <datain> as file\n<dataout> (’writedata <dataout>).\n-o <output> Write CPPTRAJ STDOUT output to file\n<output>.\n-h | –help Print command line help and exit.\n-V | –version Print version and exit.\n–defines Print compiler defines and exit.\n-debug <#> Set global debug level to <#>; same as\ninput ’debug <#>’.\n–interactive Force interactive mode.\n–log <logfile> Record commands to <logfile> (interactive\nmode only). Default is ’cpptraj.log’.\n-tl Print length of trajectories specified with ’-y’ to\nSTDOUT. The total number of frames is written out as\n’Frames: <X>’\n-ms <mask> Print selected atom numbers to STDOUT.\nSelected atoms are written out as ’Selected= 1 2 3\n...’\n-mr <mask> : Print selected residue numbers to\nSTDOUT. Selected residues are written out as\n’Selected= 1 2 3 ...’\n–mask <mask> Print detailed atom selection to STDOUT.\n10"}, {"page": 11, "text": "–resmask <mask> Print detailed residue selection to\nSTDOUT.\n--rng <type> Change default random number generator.\n--charge <mask> Print total charge (in e-) of atoms\nselected by <mask> to STDOUT.\nNote that unlike ptraj, in cpptraj it is not required that a topology file be\nspecified on the command line as long as one is specified in the input file with\nthe ’parm’ keyword. Multiple topology/input files can be specified by use of\nmultiple ’-p’ and ’-i’ flags. All topology and coordinate flags will be processed\nbefore any input flags.\n2.2 Commands\nInput to cpptraj is in the form of commands, which can be categorized in to 2\ntypes: immediate and queued. Immediate commands are executed as soon as\nthey are encountered. Queued commands are initialized when they are encoun-\ntered, but are not executed until a Run is executed via a run or go command.\nActions, Analyses, and Trajectory commands (except reference) are queued\ncommands; however, they can also be run immediately via commands such as\ncrdaction, runanalysis, loadcrd, etc. See 7 on page 34 for more details.\nCommands fall into seven categories:\nGeneral (Immediate) These commands are executed immediately when en-\ntered.\nSystem (Immediate) These are unix system commands (e.g. ’ls’, ’pwd’, etc).\nCoords (Immediate) These commands are used to manipulate COORDS data\nsets; see 7 on page 34 for more details.\nTrajectory (Queued) These commands prepare cpptraj for reading or writing\ntrajectories during a Run.\nTopology (Immediate) These commands are used to read, write, and modify\ntopology information.\nAction (Queued) These commands specify actions that will be performed on\ncoordinate frames read in from trajectories during a Run.\nAnalysis (Queued) These commands specify analyses that will be performed\non data that has been either generated from a Run or read in from an\nexternal source.\nControl (Immediate) These commands set up control blocks that can be used\nto e.g. loop over a set of commands.\n11"}, {"page": 12, "text": "Inadditiontonormalcommands,cpptraj nowhastheabilitytoperformcertain\nbasic math operations, even on data sets. See 5.2 on page 25 for more details.\nCommandsincpptraj canbereadinfromaninputfileorfromtheinteractive\ncommandprompt. A’#’anywhereonalinedenotesacomment;anythingafter\n’#’ will be ignored no matter where it occurs. A ’\\’ allows the continuation of\none line to another. For example, the input:\n# Sample input\ntrajin mdcrd # This is a trajectory\nrms first out rmsd.dat \\\n:1-10\nTranslates to:\ntrajin mdcrd\nrms first out rmsd.dat :1-10\n2.3 Getting Help\nIf in interactive mode, the ’help’ command can be used to list recognized com-\nmands and topics; topics (such as mask syntax) start with uppercase letters.\n’help <command>’ can be used to get the associated keywords as well as an\nabbreviated description of the command. Most commands have a correspond-\ning test which also serves as an example of how to use the command. See\n$AMBERHOME/AmberTools/test/cpptraj/README for more details.\n2.4 Batch mode\nIn “batch” mode, cpptraj is executed from the command line with one or more\ninput files containing commands to be processed or STDIN. The syntax of <in-\nput file> is similar to that of ptraj. Keywords specifying different commands\naregivenoneperline. Linesbeginningwith’#’areignoredascomments. Lines\ncan also be continued through use of the ’\\’ character. This is the only allowed\nmode for cpptraj.MPI.\n2.5 Interactive mode\nIn“interactivemode” userscanentercommandsinaUNIX-likeshell. Interactive\nmodeisusefulforrunningshortandsimpleanalysesorfortryingoutnewkinds\nof analyses. If cpptraj is run with ’–interactive’, no arguments, or no specified\ninput file:\ncpptraj\ncpptraj --interactive\ncpptraj <parm file>\ncpptraj -p <parm file>\n12"}, {"page": 13, "text": "this brings up the interactive interface. This interface supports command his-\ntory (via the up and down arrows) and tab completion for commands and file\nnames. Ifnologfilenamehasbeengiven(with’–log<logfile>’), allcommands\nused in interactive mode will be logged to a file named ’cpptraj.log’, which can\nsubsequently be used as input if desired. When starting cpptraj, command\nhistories will be read from any existing logs.\n2.6 Trajectory Processing “Run”\nLike ptraj, a trajectory processing “Run” is one of the main ways to run cpp-\ntraj. First the Run is set up via commands read in from an input file or the\ninteractive prompt. Trajectories are then read in one frame at a time (or in\nthe case of ensemble processing all frames from a given step are read). Actions\nare performed on the coordinates stored in the frame, after which any output\ncoordinates are written. At the end of the run, any data sets generated are\nwritten, and any queued Analyses are performed.\n2.6.1 Actions and multiple topologies\nSince cpptraj supports multiple topology files, during a Run actions are set up\nevery time the topology changes in order to recalculate things like what atoms\nareinamasketc. Actionsthatarenotvalidforthecurrenttopologyareskipped\nfor that topology. So for example given two topology files with 100 residues, if\nthe first topology file processed includes a ligand named MOL and the second\none does not, the action:\ndistance :80 :MOL out D_80-to-MOL.dat\nwill be valid for the first topology but not for the second, so it will be skipped\nas long as the second topology is active.\n2.7 Parallelization\nCpptraj has many levels of parallelization that can be enabled via the ’-mpi’,\n’-openmp’, and/or ’-cuda’ configure flags for MPI, OpenMP, and CUDA paral-\nlelization respectively. At the highest level, trajectory and ensemble reads are\nparallelized with MPI. In addition, certain time consuming actions have been\nparallelized with OpenMP and/or CUDA.\nNote that any combination of the ’-openmp’, ’-cuda’, and ’-mpi’ flags may\nbe used to generate a hybrid MPI/OpenMP/CUDA binary; however this may\nrequire additional runtime setup (e.g. setting OMP_NUM_THREADS for\nOpenMP) to work properly and not oversubscribe cores.\n2.7.1 MPI Trajectory Parallelization\nCpptraj has two levels of MPI parallelization for reading input trajectories.\nThefirstisfor’trajin’trajectoryinput, wherethetrajectoryreadisdividedas\n13"}, {"page": 14, "text": "evenly as possible among all input frames (across-trajectory parallelism). For\nexample, if given two trajectories of 1000 frames each and 4 MPI processes,\nprocess 0 reads frames 1-500 of trajectory 1, process 1 reads frames 501-1000\nof trajectory 1, process 2 reads frames 1-500 of trajectory 2, and process 3\nreads frames 501-1000 of trajectory 2. Most Actions will work with across-\ntrajectory parallelization with the exception of the following: ’clusterdihe-\ndral’, ’contacts’, ’createreservoir’, ’lipidorder’, ’pairwise’, ’stfcdiffu-\nsion’, ’tordiff’, ’unwrap’, and ’xtalsymm’. The ’diffusion’ Action will\nonly work with across-trajectory parallelism if no imaging is to be performed.\nIn addition to across-trajectory parallelism, the ’gist’ command will also\nMPI-parallelize the entropy calculation that occurs after trajectory processing.\nThesecondisfor’ensemble’ trajectoryinput,wherethereading/processing/writing\nof each member of the ensemble is divided up among MPI processes. The num-\nber of MPI processes must be a multiple of the ensemble size. If the number of\nprocessesisgreaterthantheensemblesizethentheprocessingofeachensemble\nmemberwillbedividedamongMPIprocesses(i.e. across-trajectoryparallelism\nwill be used). For example, given an ensemble of 4 trajectories and 8 processes,\nprocesses0and1areassignedtothefirstensembletrajectory,processes2and3\nareassignedtothesecondensembletrajectory,andsoon. Whenusingensemble\nmode in parallel it is recommended that the ensemblesize command be used\nprior to any ensemble command as this will make set up far more efficient.\nNotethatmostAnalysesarenotMPI-parallelized,withtheexceptionofthe\ncalcdiffusion Analysis (12.3 on page 225).\nIn order to use the MPI version, Amber/cpptraj should be configured with\nthe ’-mpi’ flag. You can tell if cpptraj has been compiled with MPI as it will\nprint ’MPI’ in the title, and/or by calling ’cpptraj —defines’ and looking for\n’-DMPI’.\n2.7.2 OpenMP Parallelization\nSome of the more time-consuming actions/analyses in cpptraj have been paral-\nlelized with OpenMP to take advantage of machines with multiple cores. In or-\ndertouseOpenMPparallelizationAmber/cpptraj shouldbeconfiguredwiththe\n’-openmp’ flag. You can easily tell if cpptraj has been compiled with OpenMP\nas it will print ’OpenMP’ in the title, and/or by calling ’cpptraj —defines’ and\nlookingfor’-D_OPENMP’.Thefollowingactions/analyseshavebeenOpenMP\nparallelized:\n2drms/rms2d\natomiccorr\ncalcdiffusion\ncheckstructure\nclosest\ncluster (pair-wise distance calculation and sieved frame restore only)\ndiffusion\ndssp/secstruct\n14"}, {"page": 15, "text": "energy\ngist (non-bonded calculation)\nhbond\nkde\nlipidscd\nmask (distance-based masks only)\nmatrix (coordinate covariance matrices only)\nminimage\nradial\nreplicatecell\nrotdif\nrmsavgcorr\nspam\nsurf\ntordiff\nunwrap\nvelocityautocorr\nvolmap\nwatershell\nwavelet\nBydefaultOpenMPcpptraj willuseallavailablecores. ThenumberofOpenMP\nthreadscanbecontrolledbysettingtheOMP_NUM_THREADSenvironment\nvariable.\n2.7.3 CUDA Parallelization\nSome time-consuming actions in cpptraj have been parallelized with CUDA\nto take advantage of machines with NVIDIA GPUs. In order to use CUDA\nparallelization Amber/cpptraj should be configured with the ’-cuda’ flag. You\ncan easily tell if cpptraj has been compiled with CUDA as it will print ’CUDA’\nanddetailsonthecurrentgraphicsdeviceinthetitle,and/orbycalling’cpptraj\n—defines’ and looking for ’-DCUDA’. The following actions have been CUDA\nparallelized:\nclosest\nwatershell\ngist\nradial\n3 General Concepts\n3.1 Units\nCpptraj uses the AKMA system of units. The execption is time, which is\ntypically expressed in ps (except where noted).\n15"}, {"page": 16, "text": "Variable Unit\nLength Angstrom\nEnergy kcal/mol\nMass AMU\nCharge electron\nTime ps (typically)\nForce kcal/mol*Angstrom\n3.2 Atom Mask Selection Syntax\nThe mask syntax is similar to ptraj. Note that the characters ’:’, ’@’, and ’*’\nare reserved for masks and should not be used in output file or data set names.\nAll masks are case-sensitive. Either names or numbers can be used. Masks\ncan contain ranges (denoted with ’-’) and comma separated lists. The logical\noperands ’&’ (and), ’|’ (or), and ’!’ (not) are also supported.\nThe syntax for elementary selections is the following:\n@{atom numlist} e.g. ’@12,17’, ’@54-85’, ’@12,54-85,90’\n@{atom namelist} e.g. ’@CA’, ’@CA,C,O,N,H’\n@%{atom type name} e.g. ’@%CT’\n@/{atom_element_name} e.g. ’@/N’\n:{residue numlist} e.g. ’:1-10’, ’:1,3,5’, ’:1-3,5,7-9’\n:{residue namelist} e.g. ’:LYS’, ’:ARG,ALA,GLY’\n::{chain id} e.g. ’::B’, ’::A,D’. Requires chain ID information be present in\nthe topology.\n:;{pdb residue number} e.g. ’:;2-4,8’. Requires a PDB loaded as topology,\nor Amber topology with embedded PDB information (see ?? on page ??).\n^{molecule numlist} e.g. ’^1-10’, ’:23,84,111’\n<mask><distance operator><distance> Selection by distance, see be-\nlow.\nSeveral wildcard characters are supported:\n’*’ Zero or more characters.\n’=’ Same as ’*’\n’?’ One character.\nThe wildcards can also be used with numbers or other mask characters, e.g.\n’:?0’ means “:10,20,30,40,50,60,70,80,90”, ’:*’ means all residues and ’@*’ means\nall atoms. If the atom name (or type name) contains a wildcard character like\nanasterisk,itcanbeexplictlyselectedbyescaping(i.e. preceding)thewildcard\ncharacter with a backslash ’\\’. So for example:\n16"}, {"page": 17, "text": "atoms @C?*\nwould select atoms named C5, C4*, C422, etc., but:\natoms @C?\\*\nwould only select C4* out of the above 3 atoms.\nCompound expressions of the following type are allowed:\n:{residue numlist | namelist}@{atom namelist | numlist}\nand are processed as:\n:{residue numlist | namelist} & @{atom namelist | numlist}\ne.g. ’:1-10@CA’ is equivalent to “:1-10 & @CA”.\nMore examples:\n:ALA,TRP All alanine and tryptophan residues.\n:5,10@CA CA carbon in residues 5 and 10.\n:*&!@H= All non-hydrogen atoms (equivalent to \"!@H=\").\n@CA,C,O,N,H All backbone atoms.\n!@CA,C,O,N,H All non-backbone atoms (=sidechains for proteins only).\n:1-500@O&!(:WAT|:LYS,ARG) Allbackboneoxygensinresidues1-500but\nnot in water, lysine or arginine residues.\n^1-2:ASP All residues named ’ASP’ in the first two molecules.\n::A,D@CA All atoms named ’CA’ in chains A and D.\nDistance-based Masks\n<mask><distance operator><distance>\n<mask> Atoms to consider.\n<distance operator> Distance operator. {<|>}{@|:|;|^}\n< Distances less than <distance> will be selected.\n> Distances greater than or equal to <distance> will be selected.\n@ Any atom.\n: Any atom within a residue.\n; Residue geometric center.\n^ Any atom within a molecule.\n<distance> The distance criteria in Angstroms.\n17"}, {"page": 18, "text": "Therearetwoveryimportantthingstokeepinmindwhenusingdistancebased\nmasks:\n1. Distance-basedmasksthatupdateeachframearecurrentlyonlysupported\nby the mask action.\n2. Selectionbydistanceforeverythingbutthemask actionrequiresdefining\nareference framewith reference; distancesare thencalculatedusing the\nspecified reference frame only. This reference frame can be changed using\nthe activeref command.\nThe syntax for selection by distance is a <mask> expression followed by a\n<distance operator> followed by a <distance> (which is in Angstroms).\nThe<distance operator>consistsof2characters: ’<’(within)or’>’(with-\nout) followed by either ’^’ (molecules), ’:’ (residues), ’;’ (residue centers), or\n’@’ (atoms). For example, ’<:3.0’ means “residues within 3.0 Angstroms” etc.\nFor ’:’ residue- and ’^’ molecule-based distance selection, if any atom in that\nresidue/moleculemeetsthegivendistancecriterion,theentireresidue/molecule\nis selected. For ’;’ residue center, the geometric center of the residue must meet\nthe given distance criterion in order to be selected.\nInplainlanguage,theentiredistancemaskcanbereadas“Select<distance\noperator> <distance> of <mask>”. So for example, the mask expression:\n:11-17<@2.4\nMeans“Selectatomswithin2.4Ådistanceofatomsselectedby’:11-17’(residues\nnumbered 11 through 17)”.\nTo strip everything outside 3.0 Å (i.e. without 3.0 Å) from residue 4 using\nspecified reference coordinates:\nreference mol.rst7\ntrajin mol.rst7\nstrip !(:4<:3.0)\n3.3 Ranges\nFor several commands some arguments are ranges (e.g. ’trajout onlyframes\n<range>’,’nastructresrange<range>’,’rmsdperresrange<range>’);THESE\nARENOTATOMMASKS.Theyaresimplenumberrangesusing’-’tospec-\nify a range and ’,’ to separate different ranges. For example 1-2,4-6,9 specifies\n1 to 2, 4 to 6, and 9, i.e. ’1 2 4 5 6 9’.\n3.4 Parameter/Reference Tagging\nParameterandreferencefilesmaybe’tagged’(i.e. givenanickname);thesetags\ncan then be used in place of the file name itself. A tag in cpptraj is recognized\nbybeingboundedbybrackets(’[’and’]’). Thiscanbeparticularlyusefulwhen\nreading in many parameter or reference files. For example, when reading in\nmultiple reference structures:\n18"}, {"page": 19, "text": "trajin Test1.crd\nreference 1LE1.NoWater.Xray.rst7 [xray]\nreference Test1.crd lastframe [last]\nreference Test2.crd 225 [open]\nrms Xray ref [xray] :2-12@CA out rmsd.dat\nrms Last ref [last] :2-12@CA out rmsd.dat\nrms Open ref [open] :2-12@CA out rmsd.dat\nThis defines three reference structures and gives them tags [xray], [last], and\n[open]. These reference structures can then be referred to by their tags instead\nof their filenames by any action that uses reference structures (in this case the\nRMSD action).\nSimilarly, this can be useful when reading in multiple parameter files:\nparm tz2.ff99sb.tip3p.truncoct.parm7 [tz2-water]\nparm tz2.ff99sb.mbondi2.parm7 [tz2-nowater]\ntrajin tz2.run1.explicit.nc parm [tz2-water]\nreference tz2.dry.rst7 parm [tz2-nowater] [tz2]\nrms ref [tz2] !(:WAT) out rmsd.dat\nThis defines two parm files and gives them tags [tz2-water] and [tz2-nowater],\nthen reads in a trajectory associated with one, and a reference structure associ-\nated with the other. Note that in the ’reference’ command there are two tags;\nthe first goes along with the ’parm’ keyword and specifies what parameter file\nthe reference should use, the second is the tag given to the reference itself (as\nin the previous example) and is referred to in the subsequent RMSD action.\n4 Variables and Control Structures\nAs of version 18, CPPTRAJ has limited support for “script” variables and ’for’\nloops. Script variables are referred to by a dollar sign (’$’) prefix and are\nreplaced when they are processed. These are stored in the master data set list\nlike other data and are assigned the type “string variable”. Note that to use\nscript variables in CPPTRAJ input that is inside another script (e.g.\na BASH script), they must be escaped with the ’\\’ character, e.g.\n#!/bin/bash\nTOP=MyTop.parm7\ncpptraj <‌<EOF\nset topname=$TOP # TOP is a BASH script variable\nparm \\$topname # topname is a CPPTRAJ script variable\nEOF\nNote that regular CPPTRAJ 1D Data Sets that contain a single value can be\nused as script variables (if the Data Set contains more than 1 value only the\nfirst value will be used).\n19"}, {"page": 20, "text": "Command Description\nfor Create a ’for’ loop.\nset Set or update a script variable.\nshow Show all current script variables and their values.\n4.1 for\nfor { {atoms|residues|molecules|molfirstres|mollastres}\n<var> inmask <mask> [parm <name> | parmindex <#> | <#>] |\n<var> in <list> |\n<var> indata <data set name> |\n<var> oversets <list> |\n<var> datasetblocks <set> blocksize <#> [blockoffset <#>]\n[cumulative [firstblock <#>]] |\n<var>=<start>;[<var><end OP><end>;]<var><increment OP>[<value>] ... }\nEND KEYWORD: ’done’\nAvailable ’end OP’ : ’<’ ’>’ ’<=’ ’>=’\nAvailable ’increment OP’ : ’++’, ’--’, ’+=’, ’-=’\natoms|residues|molecules|molfirstres|mollastres <var> inmask <mask>\nLoop over atoms/residues/molecules/first residue in\nmolecules/last residue in molecules selected by the\ngiven mask expression, set as script variable <var>.\nparm <name> | parmindex <#> <#> Select\ntopology that <mask> should be based on (default\nfirst topology).\n<var> in <list> Loop over a comma-separated list of\nstrings. File name wildcards can be used.\n<var> in <data set name> Loop over elements of\nspecified data set. Currently only 1D scalar sets\nand string sets can be specified.\n<var> oversets <list> Loop over sets selected by\ncomma-separated list of names. Data set wildcards\ncan be used.\n<var> datasetblocks <set> Loop over blocks in\nspecified DataSet.\nblocksize <#> Size of blocks to use.\n[blockoffset <#>] Offset between blocks.\n[cumulative] Instead of blocks of fixed size, use\nblocks of increasing size incremented by\nblocksize.\n[firstblock <#>] When cumulative, the size of\nthe first block (default is first data set\nelement).\n20"}, {"page": 21, "text": "<var>=<start>;[<var><end OP><end>;]<var><increment OP>[<value>]\nLoop over integer script variable <var> starting\nfrom <start>, optionally ending at <end>, increment\nby <value>.\nData Sets Created (datasetblocks loops):\n<var>[block]:<start idx> (Data set blocks only) Data\nset block of blocksize starting at <start idx>.\n<var>[cumul]:<end idx> (Cumulative data set blocks\nonly) Data set block starting at firstblock and\nending at <end idx>.\nCreate a for loop using one or more mask expressions, integers, etc. Loops\ncan be nested inside each other. Integer loops may be used without an end\ncondition, but in that case at least one descriptor in the loop should have an\nend condition or refer to a mask. Loops are ended by the done keyword.\nNote that non-integer variables (e.g. ’inmask’ loops) are NOT incremented\nafter the final loop iteration, i.e. these loop variables always retain their final\nvalue.\nFor example:\nfor atoms A0 inmask :1-3@CA i=1;i++\ndistance d$i :TCS $A0 out $i.dat\ndone\nThis loops over all atoms in the mask expression ’:1-3@CA’ (all atoms named\nCAinresidues1to3)andcreatesavariablenamed’i’thatstartsfrom1andis\nincrementedby1eachiteration. Insidetheloop,themaskselectionisreferredto\nby $A0andtheintegerby $i. Thisisequivalenttodoing3distancecommands\nlike so:\ndistance d1 :TCS :1@CA out 1.dat\ndistance d2 :TCS :2@CA out 2.dat\ndistance d3 :TCS :3@CA out 3.dat\nTo loop over files named trajA*.nc and trajB*.nc:\nfor TRAJ in trajA*.nc,trajB*.nc\ntrajin $TRAJ 1 last 10\ndone\n4.2 set\nset { <variable> <OP> <value> |\n<variable> <OP> {atoms|residues|molecules|atomnums|\nresnums|oresnums|molnums|\ncharge|mass} inmask <mask>\n21"}, {"page": 22, "text": "[parm <name> | crdset <set> | parmindex <#> | <#>]\n<variable> <OP> trajinframes }\nAvailable <OP> : ’=’, ’+=’\n<variable> <OP> <value> Set or append a script\nvariable.\n<variable> <OP> {atoms|residues|molecules|atomnums|resnums\n|oresnums|molnums} inmask <mask> Set/append a\nscript variable to/by the total number of\natoms/residues/molecules in, a range expression of\nselected atom #s/residue #s/original residue\n#s/molecule #s in, or the total charge/mass of atoms\nselected by the given mask expression.\nparm <name> | parmindex <#> | <#> Topology to\nwhich mask should correspond (default first).\n<variable> <OP> trajinframes Set/append a script\nvariable to/by the total number of frames in\ntrajectories currently loaded by trajin commands.\nSet(<OP>=’=’)orappend(<OP>=’+=’)ascriptvariable. Scriptvariables\nare character strings, and are referred to in CPPTRAJ input by using a dollar\nsign ’$’ prefix.\nFor example, the following input will load files my.parm7 and my.rst7:\nset PREFIX = my\ntrajin $PREFIX.parm7\ntrajin $PREFIX.rst7\nForexample,thefollowinginputwillprintinfoforthelast10atomsinatopology\nto ’last10.dat’:\nset Natom = atoms inmask *\nlast10 = $Natom - 10\nshow\natoms \"@$last10 - $Natom\" out last10.dat\nThe following input will put a range of residues selected by :LYS:\n> set SELECTED1 = resnums inmask :1-183&:LYS\nUsing topology: FtuFabI.NAD.TCL.parm7\nVariable ’SELECTED1’ set to ’7-8,18,26,44,49,71,79,128,135,151,163,183’\n4.3 show\nshow [<var1> ...]\nIfnovariablenamesspecified,showallcurrentscriptvariablesandtheirvalues.\nOtherwise, show the values of the specified script variables.\n22"}, {"page": 23, "text": "5 Data Sets and Data Files\nIn cpptraj, Actions and Analyses can generate one or more data sets which are\navailable for further processing. For example, the distance command creates\na data set containing distances vs time. The data set can be named by the\nuser simply by specifying a non-keyword string as an additional argument. If\nno name is given, a default one will be generated based on the action name and\ndata set number. For example:\ndistance d1-2 :1 :2 out d1-2.dat\nwill create a data set named “d1-2”. If a name is not specified, e.g.:\ndistance :1 :2 out d1-2.dat\nthe data set will be named “Dis_00000”.\nData files are created automatically by most commands, usually via the\n“out” keyword. Datafilescanalsobeexplicitlycreatedwiththewrite/writedata\nand create commands. Data can also be read in from files via the readdata\ncommand. Cpptrajcurrentlyrecognizestheformatslistedin1,althoughitcan-\nnot write in all formats. In addition, a data set must be valid for the data file\nformat. For example, 3D data (such as a grid) can be written to an OpenDX\nformat file but not a Grace format file.\nThedefaultfileformatiscalled’Standard’,whichsimplyhasdataincolumns,\nlike ptraj, although multiple data sets can be directed to the same output file.\nThe format of a file can be changed either by specifying a recognized keyword\n(eitheronthecommandlineitselforlaterviaa’datafile’command)orbygiving\nthe file an extension corresponding to te format, so ’filename.agr’ will output\nin Grace format, and ’filename.gnu’ will output in Gnuplot contour, and so on.\nThe xmgrace/gnuplot output is particularly nice for the secstruct sumout and\nrmsd perresout files. Additional options for data files can be found in 6 on\npage 27.\nAny action using the “out” keyword will allow data sets from separate com-\nmands to be written into the same file. For example, the commands:\ndihedral phi :1@C :2@N :2@CA :2@C out phipsi.dat\ndihedral psi :2@N :2@CA :2@C :3@N out phipsi.dat\nwill assign the “phi” and “psi” data sets generated from each action to the stan-\ndard data output file “phipsi.dat”:\n#Frame phi psi\nNote that when reading the Amber Prep and Amber OFF Library formats, a\nCOORDS data set will be created for each unit present in these files.\n23"}, {"page": 24, "text": "Format Filename Keyword Valid Notes\nExtensions Dimensions\nStandard .dat dat 1D, 2D, 3D\nGrace .agr, .xmgr grace 1D\nGnuplot .gnu gnu 1D, 2D\nXplor .xplor, .grid xplor 3D\nOpenDX .dx opendx 3D\nAmber REM .log remlog - Read Only\nlog\nAmber .mdout mdout - Energy\nMDOUT information,\nRead Only\nAmber Energy .ene amberene 1D Read Only\nFile\nAmber Evecs .evecs evecs Modes data set\nonly\nAmber .cpout cpout pH data only\nConstant pH\noutput\nDensity Peaks .peaks peaks 3D density\npeaks\n(spam/volmap)\nVector .vectraj vectraj Vector data set Write Only\npseudo-traj only.\nGromacs XVG .xvg xvg - Read Only\nCCP4 .ccp4 ccp4 3D\nCharmm .exch charmmrepd - Read Only\nREPD log\nCharmm .charmmout charmmout - Energy\nOutput information,\nRead Only\nPairwise Cache .cmatrix cmatrix pairwise Used for cluster\n(binary) distances analysis.\nPairwise Cache .nccmatrix nccmatrix pairwise Used for cluster\n(NetCDF) distances analysis.\nNetCDF Data .nc netcdf All data Only state info\nsaved for pH\ndata.\nAmber Prep .prepin prepin COORDS Read Only\nFile\nAmber OFF .off, .lib off,lib COORDS Read Only\nLibrary File\nTable1: DataFileformatsrecognizedbycpptraj. ’ValidDimensions’showswhat\ndimensions the format is valid for (e.g. you cannot write a 1D data set with\nOpenDX format).\n24"}, {"page": 25, "text": "5.1 Data Set Selection Syntax\nManyanalysiscommandscanbeusedtoanalyzemultipledatasets. Thegeneral\nformat for selecting data sets is:\n<name>[<aspect>]:<index>\nThe ’*’ character can be used as a wild-card for entire names (no partial\nmatches).\n• <name>: The data set name, usually specified in the action (e.g. in\n’distance d0 @1 @2’ the data set name is “d0”).\n• <aspect>: Optional; this is set for certain data sets internally in or-\nder to easily select subsets of data. The brackets are required. For\nexample, when using ’hbond series’, both solute-solute and solute-solvent\nhydrogen bond time series may be generated. To select all solute-solute\nhydrogen bonds one would use the aspect “[solutehb]”; to select solute-\nsolvent hydrogen bonds the aspect “[solventhb]” would be used. Aspects\nare hard-coded and are listed in the commands that use them.\n• <index>: Optional; for actions that generate many data sets (such as\n’rmsd perres’) an index is used. Depending on the action, the index may\ncorrespond to atom #s, residue #s, etc. A number range (comma and/or\ndash separated) may be used.\nFor example: to select all data sets with aspect “[shear]” named NA_00000:\nNA_00000[shear]\nTo select all data sets with aspect “[stagger]” with any name, indices 1 and 3:\n*[stagger]:1,3\nIn ensemble mode, data set selection has additional syntax:\n<name>[<aspect>]:<index>%<member>\nWhere <member> is the ensemble member number starting from 0.\n5.2 Data Set Math\nAs of version 15, cpptraj can perform basic math operations, even on data sets\n(with some limitations). Currently recognized operations are:\nOperation Symbol\nMinus -\nPlus +\nDivide /\nMultiply *\nPower ^\nNegate -\nAssign =\n25"}, {"page": 26, "text": "Several functions are also supported:\nFunction Form\nSquare Root sqrt()\nExponential exp()\nNatural Logarithm ln()\nAbsolute Value abs()\nSine sin()\nCosine cos()\nTangent tan()\nSummation sum()\nAverage avg()\nStandard Deviation stdev()\nMinimum min()\nMaximum max()\nNumberscanbeexpressedinscientificnotationusing“E” notation,e.g. 1E-5\n=0.00001. TheparseralsorecognizesPIasthenumberpi. Expressionscanalso\nbe enclosed in parentheses. So for example, the following expression is valid:\n> 1 - ln(sin(PI/4) * 2)^2\nResult: 0.879887\nResults of numerical calculations like the above can be assigned to a variable\n(essentially a data set of size 1) for use in subsequent calculations, e.g.\n> R = 1 - ln(sin(PI/4) * 2)^2\nResult stored in ’R’\n> R + 1 Result: 1.879887\nData sets can be specified in expressions as well. Currently data sets in an\nexpression must be of the same type and only 1D, 2D, and 3D data sets are\nsupported. Functions are applied to each member of the data set. So for exam-\nple, given two 1D data sets of the same size named D0 and D1, the following\nexpression:\n> D2 = sqrt( D0 ) + D1\nwould take the square root of each member of D0, add it to the corresponding\nmember of D1, and assign the result to D2. The following table lists which\noperationsarevalidfordatasettypes. Ifatypeisnotlisteditisnotsupported:\n26"}, {"page": 27, "text": "Data Set Type Supported Ops Supported Funcs Notes\n1D (integer, All All\ndouble, float)\n1D (vector) +, -, *, /, = None ’*’ is dot product\n2D (matrices) +, -, /, *, = sum, avg, stdev,\nmin, max\n3D (grids) +, -, /, *, = sum, avg, stdev,\nmin, max\n6 Data File Options\nDatafileoutputcanbehandledmultiplewaysincpptraj. Outputdatafilescan\nbe created by Actions/Analyses/Commands, or can be explicitly created with\nwritedata (8.32 on page 67) or create (8.4 on page 53) commands. Reading\ndata from files is only done via the readdata command (8.22 on page 64).\nIn general, data files which have been declared with an ’out’ keyword will\nrecognize data file write keywords on the same command line. For example,\nthe ’time’ argument can be passed directly to the output from a distance\ncommand:\ndistance d0 :1 :2 out d0.agr time 0.001\nThe data file format can be changed from standard implicitly by using specific\nfilenameextensionsorkeywords. Iftheextensionisnotrecognizedornokeyword\nis give the default format is ’Standard’. Keywords and extensions for data file\nformats recognized by cpptraj are shown in 1. Note that the use of certain\noptions may be restricted for certain data file formats. These options can also\nbe passed to data files via the datafile command (8.6 on page 54).\n[<format keyword>]\n[{xlabel|ylabel|zlabel} <label>] [{xmin|ymin|zmin} <min>] [sort]\n[{xstep|ystep|zstep} <step>] [time <dt>] [prec <width>[.<precision>]]\n[xprec <width>[.<precision>]] [xfmt {double|scientific|general}]\n[noensextension]\n{xlabel | ylabel | zlabel} <label> Set the x-axis label\nfor the specified datafile to <label>. For regular\ndata files this is the header for the first column\nof data. If the data is at least 2-dimensional\n’datafile ylabel <label>’ will likewise set the\ny-axis label.\n{xmin | ymin | zmin} <min> Set the starting X\ncoordinate value to <min>. If the data is at least\n2-dimensional ’datafile ymin <min>’ will likewise\nset the starting Y coordinate value.\nsort Sort data sets prior to write. Ordering is by\nname, aspect, then index (all descending).\n27"}, {"page": 28, "text": "{xstep | ystep | zstep} <step> Multiply each frame\nnumber by <step> (x coordinates). If the data is at\nleast 2-dimensional ’datafile ystep <step>’ will\nlikewise multiply y coordinates by <step>.\ntime <dt> Equivalent to the ptraj argument ’time’ that\ncould be specified with many actions. Multiplies\nframe numbers (x-axis) by <dt>.\nprec <width>[.<precision>] Change the output format\nwidth (and optionally precision) of all sets\nsubsequently added to the data file (i.e. does not\nchange the precision of any data sets currently in\nthe file). For example,\nprec 12.4\nprec 10\nxprec <width>[.<precision>] Change output ordinate\nwidth and precision.\nxfmt {double|scientific|general} Change output ordinate\nformat.\n[noensextension] Omit ensemble extension in ensemble\nprocessing mode. NOTE: THIS OPTION HAS NOT BEEN\nFULLY TESTED IN PARALLEL.\n6.1 Standard Data File Options\nWrite\n[invert] [noxcol] [groupby <type>] [noheader] [square2d|nosquare2d]\n[nosparse|sparse [cut <cutoff>]]\ninvert Normally, data is written out with X-values\npertaining to frames (i.e. data over all\ntrajectories is printed in columns). This command\nflips that behavior so that X-values pertain to data\nsets (i.e. data over all trajectories is printed in\nrows).\ngroupby <type> (1D) group data sets by <type>:\nname Group by name.\naspect Group by aspect.\nidx Group by index.\nens Group by ensemble number.\ndim Group by dimension.\nxcol Write indices for the specified datafile. This is\nusually the default behavior.\n28"}, {"page": 29, "text": "noxcol Prevent printing of indices (i.e. the #Frame\ncolumn in most datafiles) for the specified\ndatafile. Useful e.g. if one would like a 2D plot\nsuch as phi vs psi. For example, given the input:\ndihedral phi :1@C :2@N :2@CA :2@C out phipsi.dat\ndihedral psi :2@N :2@CA :2@C :3@N out phipsi.dat\ndatafile phipsi.dat noxcol\nCpptraj will write a 2 column datafile containing\nonly phi and psi, no frame numbers will be written.\nheader Write header line at beginning of data file.\nThis is usually the default behavior.\nnoheader Prevent printing of header line (e.g.\n’#Frame D1’) at the beginning of data file.\nsquare2d Write 2D data as a square matrix, e.g.:\n<1,1> <2,1> <3,1>\n<1,2> <2,2> <3,2>\nnosquare2d Write 2D data in 3 columns as:\n<X> <Y> <Value>\nsparse Only write 3D grid voxels with value > cutoff\n(default 0).\ncut <cut> Cutoff for ’sparse’; default 0.\nnosparse Write all 3D voxels (default).\nRead\n[prec {flt|dbl}]\n{[read1d [index <col>] [onlycols <range>] [floatcols <range>]\n[intcols <range>] [stringcols <range>]] |\n[read2d [{square2d|nosquare2d}]] |\n[vector [magnitude]] |\n[mat3x3] |\n[read3d [dims <nx>,<ny>,<nz>] [origin <ox>,<oy>,<oz>]\n[delta <dx>,<dy>,dz>] [prec {dbl|flt}] [bin {center|corner} ]\n}\nprec {flt|dbl} Read 2d/3d data as single (flt) or double\n(dbl, default) precision.\nread1d Read data as 1D data sets (default).\nindex <col> Use column <col> (starting from 1) as\nindex column (1D data only).\nonlycols <range> Only read columns in range.\n29"}, {"page": 30, "text": "floatcols <range> Force specified columns to be\nread as single-precision floats.\nintcols <range> Force specified columns to be read\nas integers.\nstringcols <range> Force specified columns to be\nread as strings.\nread2d Read data as 2D matrix.\nsquare2d Read data as square matrix (default).\nnosquare2d Read data as XYZ matrix (i.e. each line\ncontains ’<column> <row> <data>’).\nvector Read data as vector. If indices are present they\nwill be skipped. Assume first 3 columns after the\nindex column are vector X, Y, and Z, and (if\npresent) the next 3 columns contain vector origin X,\nY, and Z.\nmagnitude If specified, assume vector file last\ncolumn contains vector magnitudes; read these\ninto a set with Aspect ’Mag’.\nmat3x3 Read data as 3x3 matrix. If indices are present\nthey will be skipped. Assume matrices are in row\nmajor order on each line, i.e. M(1,1) M(1,2) ...\nM(3,2) M(3,3).\nread3d Read data as 3D grid. If no dimension data in\nfile must also speify ’dims’.\ndims <dx>,<dy>,<dz> Grid dimensions.\norigin <ox>,<oy>,<oz> Grid origins (default\n0,0,0).\ndelta <dx>,<dy>,dz> Grid spacings (default\n1,1,1).\nprec {dbl|flt} Grid precision, double or float\n(default float).\nbin {center|corner} Coords specify bin centers or\ncorners (default corners).\nBydefault,standarddatafilesareassumedtocontain1Ddataincolumns. Data\nsetlegendswillbereadinifthefilehasaheaderline(denotedby’#’). Columns\nlabeled ’#Frame’ are automatically considered the ’index’ column and skipped.\nData sets are stored as <name>:<idx> where <name> is the given data set\nname (the file name if not specified) and <idx> corresponds to the column\nthe data was read from starting from 1. Cpptraj assumes the data increases\nmonotonically and will automatically attempt to determine the dimensions of\nthe data set(s); a warning will be printed if this is not successful.\nIf a file contains the header:\n30"}, {"page": 31, "text": "#F1 F2 <name>\nCPPTRAJwillassumethefilecontainspairwisedistancesforclustering, where\nthe F1 and F2 columns contain the frame numbers, and the <name> column\ncontains the distance.\n6.2 Grace Data File Options\nFor more information on Grace see http://plasma-gate.weizmann.ac.il/Grace/.\nWrite\n[{invert|noinvert}] [{xydy|noxydy}] [<label set>]\ninvert Normally, data is written out with X-values\npertaining to frames (i.e. data over all\ntrajectories is printed in columns). This command\nflips that behavior so that X-values pertain to data\nsets.\nnoinvert Do not flip X-Y axes (default).\nxydy Combine consecutive pairs of sets into XYDY sets.\nnoxydy Do not combine consecutuve pairs of sets into\nXYDY sets (default).\n<label set> If a string dataset is specified, assume it\nhas data point labels.\nIf a single string data set is specified when writing Grace format, it is assumed\nthey are data point labels.\nRead\nCpptraj will read set legends from grace files, and data sets are stored as\n<name>:<idx> where <name> is the given data set name (the file name if\nnotspecified)and<idx>correspondstothesetnumberthedatawasreadfrom\nstarting from 0.\n6.3 Gnuplot Data File Options\nFor more information on these options it helps to look at the PM3D options in\nthe Gnuplot manual (see http://www.gnuplot.info/).\nWrite\n[{nolabels|labels}] [{usemap|pm3d|nopm3d}] [title <title>]\n[jpeg] [noheader] [{xlabels|ylabels|zlabels} <labellist>]\n31"}, {"page": 32, "text": "nolabels Do not print axis labels.\nlabels Print axis labels.\nusemap pm3d output with 1 extra empty row/col (may\nimprove look).\npm3d Normal pm3d map output.\nnopm3d Turn off pm3d\njpeg Plot will write to a JPEG file when used with\ngnuplot.\ntitle <title> Set plot title (default is file name).\nbinary Plot will be written in binary format.\nheader Format the plot so it can be directly processed\nby gnuplot. This is usually the default behavior.\nnoheader Do not format plot; data output only.\npalette <arg> Change gnuplot pm3d palette to <arg>:\n’rgb’ Red, yellow, green, cyan, blue, magenta, red.\n’kbvyw’ Black, blue, violet, yellow, white.\n’bgyr’ Blue, green, yellow, red.\n’gray’ Grayscale.\nxlabels|ylabels|zlabels <labellist> Set x, y, or z axis\nlabels with comma-separated list, e.g. ’xlabels\nX1,X2,X3’.\n6.4 Amber REM Log Options\nNote that multiple REM logs can be specified in a single readdata command.\nSee 12.30 on page 267 for more on replica log analysis.\nRead\n[nosearch] [dimfile <file>] [crdidx <crd indices>]\n[nosearch] If specified do not automatically search for\nMREMD dimension logs.\n[dimfile <file>] remd.dim file for processing MREMD logs.\n[crdidx <crd indices>] Use comma-separated list of\nindices as the initial coordinate indices (H-REMD\nonly). For example (4 replicas):\ncrdidx 4,2,3,1\n6.5 Amber MDOUT Options\nNote that multiple MDOUT files can be specified in a single readdata com-\nmand.\n32"}, {"page": 33, "text": "6.6 Evecs File Options\nRead\n[ibeg <firstmode>] [iend <lastmode>]\nibeg <firstmode> Number of the first mode (or principal\ncomponent) to read from evecs file. Default 1.\niend <lastmode> Number of the last mode (or principal\ncomponent) to read from evecs file. Default is to\nread all for newer evecs files (generated by cpptraj\nversion > 12), 50 for older evecs files.\n6.7 Vector psuedo-traj Options\nThis can be used to write out a representation of a vector data set which can\nthen be visualized. See 11.91 on page 214 for more on generating vector data\nsets.\nWrite\n[trajfmt <format>] [parmout <file>] [noorigin]\ntrajfmt <format> Output pseudo-trajectory format. See\n10 on page 83 for trajectory format keywords.\nparmout <file> File to write pseudo-trajectory topology\nto.\n[noorigin] Do not write vector origin coordinates.\n6.8 OpenDX file options\nRead\n[type {float|double}]\ntype {float|double} Precision to read in 3D grid\n(default float).\nWrite\n[bincenter] [gridwrap] [gridext]\nbincenter Center grid points on bin centers instead of\ncorners.\ngridwrap Like ’bincenter’, but also wrap grid density.\nUseful when grid encompasses unit cell.\ngridext Like ’bincenter’, but also print extra layer of\nempty bins.\n33"}, {"page": 34, "text": "6.9 CCP4 file options\nWrite\n[title <title>]\n[title <title>] Set CCP4 output title.\n6.10 Charmm REPD log options\nRead\n[nrep <#>] [crdidx <crd indices>]\nnrep <#> Total number of replicas.\ncrdidx <crd indices> Comma-separated list of indices to\nuse as initial coordinate indices.\n6.11 Amber Constant pH Out options\nRead\ncpin <file>\ncpin <file> Constant pH input (CPIN) file name.\nNote that when reading in constant pH data the data set aspect will be set to\ntheresiduenameandtheindexwillbesettotheresiduenumber. Whenreading\nin constant pH REMD data the data is unsorted, and sortensembledata should\nbe used to create sorted constant pH data sets (see 8.30 on page 67).\n7 Coordinates (COORDS) Data Set Commands\nCoordinate I/O tends to be the most time-consuming part of trajectory anal-\nysis. In addition, many types of analyses (for example two-dimensional RMSD\nand cluster analysis) require using coordinate frames multiple times. To sim-\nplify this, trajectory coordinates may be saved as a separate data set via the\nloadcrd command or createcrd action. Any action can then be performed on\nthe COORDS data set with the crdaction command. The crdout command\ncan be used to write coordinates to an output trajectory (similar to trajout).\nAlthoughCOORDSdatasetsstoreeverythinginternallywithsingle-precision,\ntheycanstillusealargeamountofmemory. Becauseofthisthereisaspecialized\ntypeofCOORDSdatasetcalledaTRAJdataset(trajectory),whichfunctions\nexactly like a COORDS data set except all data is stored on disk. TRAJ data\nsets can be created with the loadtraj command. TRAJ data sets cannot be\nmodified.\nThereareseveralanalysesthatcanbeperformedusingCOORDSdatasets,\neither as part of the normal analysis list or via the runanalysis command.\nNote that while these analyses can be run on specified COORDS data sets, if\n34"}, {"page": 35, "text": "one is not specified a default COORDS data set will be created, made up of\nframes from trajin commands.\nAs an example of where this might be useful is in the calculation of atomic\npositional fluctuations. Previously this required two steps: one to generate an\naverage structure, then a second to rms-fit to that average structure prior to\ncalculatingthefluctuations. Thiscannowbedoneinonepasswiththefollowing\ninput:\nparm topology.parm7\nloadcrd mdcrd.nc\n# Generate average structure PDB, @CA only\ncrdaction mdcrd.nc average avg.pdb @CA\n# Load average structure PDB as reference\nparm avg.pdb\nreference avg.pdb parm avg.pdb\n# RMS-fit to average structure PDB\ncrdaction mdcrd.nc rms reference @CA\n# Calculate atomic fluctuations for @CA only\ncrdaction mdcrd.nc atomicfluct out fluct.dat bfactor @CA\nThe following COORDS data set commands are available:\n35"}, {"page": 36, "text": "Command Description\ncatcrd Concatenate two or more COORDS sets.\ncombinecrd Combine two or more COORDS sets.\ncrdaction Run a single Action on a COORDS set.\ncrdout Write a COORDS set to a file.\ncrdtransform Transform a COORDS set in one of several ways.\ncreatecrd (Action) Create a COORDS set during a Run.\nemin Run simple energy minimization on a frame of a COORDS\nset.\nextendedcomp Calculate extended comparison similarity values for each\nframe in COORDS set.\ngraft Graft partof one COORDS setontoanother COORDS set.\nloadcrd Create or append to a COORDS set from a file.\nloadtraj Create special COORDS set where frames remain on disk.\npermutedihedrals Rotate specified dihedral(s) in given COORDS set by\nspecific interval or to random values.\nprepareforleap Prepare a structure (usually loaded from a PDB) for\nprocessing with LEaP from Amber.\nreference Load a single trajectory frame as a reference.\nrotatedihedral Rotate specified dihedral to specified value or by given\nincrement.\nsequence Create a new molecule from a sequence of COORDS sets.\nsplitcoords Split molecules in a COORDS set into a trajectory.\nzmatrix Apply Z-matrix to a COORDS set or calculate Z-matrix\nfor a molecule/frame in a COORDS set.\n7.1 catcrd\ncatcrd <crd1> <crd2> ... name <name>\n<crdX> COORDS data sets to concatentate, specify 2 or\nmore.\nname <name> New COORDS set name\nConcatentate two or more COORDS data sets into a single COORDS data set.\nThe topologies must have the same number of atoms for this to work. If the\ntopologies differ in other ways, the topology of the first COORDS set takes\npriority.\n7.2 combinecrd\ncombinecrd <crd1> <crd2> ... [parmname <topname>] [crdname <crdname>]\n<crdX> COORDS data sets to combine, specify 2 or more.\n[parmname <topname>] Name of combined Topology.\n36"}, {"page": 37, "text": "[crdname <crdname>] Name of combined COORDS data set.\nCombined two or more COORDS data sets into a single COORDS data set.\nNote that the resulting topology will most likely not be usable for MD simu-\nlations. Box information will be retained - the largest box dimensions will be\nused.\nFor example, to load two MOL2 files as COORDS data sets, combine them,\nand write them out as a single MOL2:\nloadcrd Tyr.mol2 CRD1\nloadcrd Pry.mol2 CRD2\ncombinedcrd CRD1 CRD2 parmname Parm-1-2 crdname CRD-1-2\ncrdout CRD-1-2 Tyr.Pry.mol2\n7.3 crdaction\ncrdaction <crd set> <actioncmd> [<action args>] [crdframes <start>,<stop>,<offset>]\nPerform action <actioncmd> on COORDS data set <crd set>. A subset of\nframes in the COORDS data set can be specified with ’crdframes’.\nForexample,tocalculateRMSDforapreviouslycreatedCOORDSdataset\nnamed crd1 using frames 1 to the last, skipping every 10:\ncrdaction crd1 rmsd first @CA out rmsd-ca.agr crdframes 1,last,10\n7.4 crdout\ncrdout <crd set> <filename> [<trajout args>] [crdframes <start>,<stop>,<offset>]\nWriteCOORDSdataset<crdset>totrajectorynamed<filename>. Asubset\nof frames in the COORDS data set can be specified with ’crdframes’.\nFor example, to write frames 1 to 10 from a previously created COORDS\ndata set named “crd1” to separate PDB files:\ncrdout crd1 crd1.pdb multi crdframes 1,10\n7.5 crdtransform\ncrdtransform <input crd set> [name <output crd set>]\n{ rmsrefine [mask <mask>] [mass] [rmstol <tolerance>] |\nnormcoords |\ntrim [metric <metric>] [{ntrimmed <#>|cutoff <val>}]\n[criterion {comp|medoid}]]\n}\n37"}, {"page": 38, "text": "<input crd set> COORDS set to transform.\n[name <output crd set>] COORDS set to create; if not\nspecified <input crd set> will be modified.\nrmsrefine Do iterative RMS refinement.\n[mask <mask>] Mask of atoms to fit during\nrefinement.\n[mass] Mass-weight the refinement.\n[rmstol <tolerance>] Tolerance (in Ang.) below\nwhich RMS-refinement will stop.\nnormcoords Normalize coordinates between 0.0 and 1.0\nusing the minimum and maximum coordinate values.\ntrim Remove trajectory frames using extended similarity\nmetrics.\n[metric <metric>] Metric to use; default MSD.\n[{ntrimmed <#>|cutoff <val>} # of frames or\nfraction of trajectory to trim.\n[criterion {comp|medoid}] Trim frames by comparitive\nsimilarity (i.e. trim most dissimilar frames)\nor comparison to medoid (i.e. trim most\ndissimilar to medoid frame).\nTransformaCOORDSsetinoneofseveralways. DoesnotyetworkwithTRAJ\ndatasets. TheiterativeRMSrefinementissimilartotheprocedureoutlinedby\nKlem et al. (J. Chem. Theory Comput. 2022, 18, 3218−3230). The extended\nsimilaritymetricsarethosedefinedbyRaczetal. (J.Comp.-Aid. Mol. Design,\n2022, 36, 157-173).\n7.6 createcrd\nThis command is actually an Action that can be used to create COORDS data\nsets during trajectory processing, see 11.21 on page 118.\n7.7 emin\nemin crdset <name> [trajoutname <name>] [rmstol <tol>] [nsteps <#>]\n[<mask>] [frame <#>] [dx0 <step0>] [out <file>] [name <setname>]\n[{nonbond|openmm}] [<potential options>]\ncrdset <name> COORDS set to use.\n[trajoutname <name>] Optional output trajectory for\nminimization steps.\n[rmstol <tol>] Minimum RMS tolerance (default 1E-4).\n[nsteps <#>] Number of minimization steps (default 1).\n38"}, {"page": 39, "text": "[<mask>] Atoms to minimize (default all).\n[frame <#>] Frame from COORDS set to minimize (default\n1).\n[dx0 <step0>] Size of initial minimization step\n(default 0.01).\n[out <file>] File to write energies to.\n[name <setname>] If specified, create an energy per\nstep data set.\n[nonbond] If specified, use simple nonbonded potential\nterm in additon to bonded terms.\n[openmm] If specified and if CPPTRAJ was compiled with\nOpenMM support, use OpenMM to calculate the forces.\n<potential options>\ncut <cutoff> Set nonbonded interaction cutoff in Ang.\n(electrostatics and vdW). Default 8.0.\ncutee <cutoff> Set electrostatics interaction cutoff in\nAng.\ncutnb <cutoff> Set vdW interaction cutoff in Ang.\nscaleee <factor> Scaling factor to multiply 1-4\nelectrostatics interactions by.\nscalenb <factor> Scaling factor to multiply 1-4 vdW\ninteractions by.\nshake <type> Use SHAKE constraints of <type>:\n’hydrogen’ Constrain bonds to hydrogen.\n’all’ Constrain all bonds.\nnexclude <#> Number of bonded atoms within which\nnonbonded interactions are excluded. Default 4.\nqfac <factor> Factor to use in electrostatic\ncalculation.\nDataSets Created\n<setname>[Energy] Total energy at each minimization\nstep.\nTHIS COMMAND IS STILL IN DEVELOPMENT AS OF VERSION 5.0.2.\nPerformsteepestdescentminimizationonaframeinaCOORDSsetusinga\nvery basic force field (bonds, angles, dihedrals). A simple nonbonded term can\nbe added as well if desired.\n39"}, {"page": 40, "text": "7.8 extendedcomp\nextendedcomp <input crd set> [name <output data set>]\n[metric <metric>] [out <file>]\n<metric> = msd bub fai gle ja jt rt rr sm ss1 ss2\n<input crd set> Input COORDS set.\n[name <output data set>] Output data set name\ncontaining values.\n[metric <metric>] Metric to use.\n[out ,file>] File to write values to.\nDataSets generated:\n<output set name> Set containing similarity values for\neach COORDS frame.\nCalculate extended comparison similarity values for each frame in a COORDS\nset. TheextendedsimilaritymetricsarethosedefinedbyRaczetal. (J.Comp.-\nAid. Mol. Design, 2022, 36, 157-173). The metrics are as follows:\nKeyword Metric\nmsd Mean-squared deviation.\nbub Bhattacharyya’s U coefficient\nfai Faiman’s coefficient\ngle Gleason’s coefficient\nja Jaccard’s coefficient\njt Jaccard-Tanimoto coefficient\nrt Rogers-Tanimoto coefficient\nrr Russell-Rao coefficient\nsm Simpson’s coefficient\nss1 Sokal-Sneath 1 coefficient\nss2 Sokal-Sneath 2 coefficient\n7.9 graft\ngraft src <source COORDS> [srcframe <#>] [srcmask <srcmask> [srccharge <srccharge>]]\ntgt <target COORDS> [tgtframe <#>] [tgtmask <tgtmask> [tgtcharge <tgtcharge>]]\n{ic | [srcfitmask <srcmask>] [tgtfitmask <tgtmask>]}\nname <output COORDS> [bond <tgt>,<src> ...]\nsrc <source COORDS> Source coordinates.\n[srcframe <#>] Frame # from source coordinates to use\n(default 1).\n[srcmask <mask>] Atoms to keep from source (default\nall).\n40"}, {"page": 41, "text": "[srccharge <charge>] If trimming atoms from source,\nensure sum of charges on remaining atoms equals\n<charge> via scaling.\ntgt <target COORDS> Target coordinates that will be\ngrafted onto.\n[tgtframe <#> Frame # from target coordinates to use\n(default 1).\n[tgtmask <mask>] Atoms to keep from target (default\nall).\n[tgtcharge <charge>] If trimming atoms from target,\nensure sum of charges on remaining atoms equals\n<charge> via scaling.\n[ic] Connect source and target using internal\ncoordinates.\n[srcfitmask <mask>] Atoms from source to use if\nRMS-fitting source onto target.\n[tgtfitmask <mask>] Atoms from target to use if\nRMS-fitting source onto target.\nname <output COORDS> Name of output COORDS set\ncontaining source grafted onto target.\n[bond <tgt>,<src>] Create a bond between target atom\nselected by <tgt> and source atoms selected by <src>\nin the final structure. Must be specified only once\nif connecting via internal coordinates, otherwise\nmay be specified multiple times.\nGraft one COORDS set onto another. If srcfitmask and/or tgtfitmask is\nspecified, the source coordinates will be RMS best-fit onto target using the\nspecified atoms. Only the atoms specified by srcmask and tgtmask will be\nkept. Thebondkeywordcanbeusedtocreatebondsbetweentargetandsource\ninthefinalstructure. Ifusinginternalcoordinatestoconnecttheunits, exactly\none bond must be specified.\n7.10 loadcrd\nloadcrd <filename> [parm <parm> | parmindex<#>] [<trajin args>] [name <name>]\n[prec {single|double}\n<filename> Trajectory file to load.\n[parm <parmfile/tag>] Topology filename/tag to\nassociate with trajectory (default first topology).\n[parmindex <#>] Index of Topology to associate with\ntrajectory (default 0, first topology).\n41"}, {"page": 42, "text": "[<trajin args>] Additional ’trajin’ args; see 10.4 on\npage 87.\n[name <name>] Name of the COORDS set.\n[prec {single|double}] Load as either a single-precision\nCOORDS set (the default) or a double-precision\nFRAMES set (which will use much more memory).\nImmediatelyloadtrajectory<filename>asaCOORDSdatasetnamed<name>\n(default base name of <filename>). If <name> is already present the coordi-\nnates will be appended to the existing data set.\nForexample,toloadframesfromtrajectoriesnamed’traj1.nc’and’traj2.nc’\ninto a COORDS data set named Crd1:\nloadcrd traj1.nc name Crd1\nloadcrd traj2.nc name Crd2\n7.11 loadtraj\nloadtraj name <setname> [<filename>]\nname <setname> Name of the TRAJ set.\n[<filename>] If specified, trajectory to add to the\nTRAJ set.\nThis command functions in two ways. If <filename> is not provided, all cur-\nrently loaded input trajectories (from trajin commands) are added to TRAJ\ndata set named <setname>. Note that if the input trajectory list is\ncleared (via ’clear trajin’) this will invalidate the TRAJ data set. In\naddition, currently all trajectories must have the same number of atoms. Oth-\nerwise add trajectory <filename> to TRAJ data set <setname>.\nTRAJ data sets cannot be modified.\n7.12 permutedihedrals\npermutedihedrals crdset <COORDS set> resrange <range> [{interval | random}]\n[outtraj <filename> [<outfmt>]] [crdout <output COORDS>]\n[<dihedral types>]\nOptions for ’random’:\n[rseed <rseed>] [out <# problems file> [<set name>]]\n[ check [cutoff <cutoff>] [rescutoff <rescutoff>] [maxfactor <max_factor>]\n[backtrack <backtrack> [checkallresidues] [increment <increment>]] ]\nOptions for ’interval’:\n<interval deg>\n<dihedral types> = alpha beta gamma delta epsilon zeta nu1 nu2 h1p c2p chin\nphi psi chip omega\n42"}, {"page": 43, "text": "crdset <COORDS set> COORDS data set to operate on.\nresrange <range> Residue range to search for\ndihedrals.\ninterval Rotate found dihedrals by <interval>. This is\ndone in an ordered fashion so that every combination\nof dihedral rotations is sampled at least once.\nrandom Rotate each found dihedral randomly.\n[outtraj <filename>] Trajectory file to write\ncoordinates to.\n[<outfmt>] Trajectory file format.\n[crdout <output COORDS>] COORDS data set to write\ncoordinates to.\n<dihedral type> One or more dihedral types to search\nfor.\nOptions for ’interval:\n<interval deg> Amount to rotate dihedral by each step.\nOptions for ’random’:\n[rseed <rseed>] Random number seed.\n[out <# problems file>] File to write number of\nproblems (clashes) each frame to.\n[<set name>] Number of problems data set name.\n[check] Check randomly rotated structure for clashes.\n[cutoff <cutoff>] Atom cutoff for checking for\nclashes (default 0.8 Å).\n[rescutoff <cutoff>] Residue cutoff for checking for\nclashes (defualt 10.0 Å).\n[maxfactor <max_factor>] The maximum number of\ntotal attempted rotations will be <max_factor> *\n<total # of dihedrals> (default 2).\n[backtrack <backtrack>] (No longer recommended as\nof version 5.1.0). If a clash is encountered at\ndihedral N and cannot be resolved, go to\ndihedral N-<backtrack> to try and resolve the\nclash (default is no backtracking).\n[checkallresidues] If specified all residues\nchecked for clashes, otherwise only residues\nup to the currently rotated dihedral check.\n[increment <increment>] If a clash is\nencountered, first attempt to rotate dihedral\nby increment to resolve it; if it cannot be\nresolved by a full rotation the calculation\nwill backtrack (default 1).\n43"}, {"page": 44, "text": "Create a trajectory by rotating specified dihedrals in a structure by regular in-\ntervals(interval),orcreate1structurebyrandomlyrotatingspecifieddihedrals\n(random). When randomly rotating dihedrals steric clashes will be checked if\ncheck is specified; in such cases the algorithm will attempt to resolve the clash\nas best it can. If clashes are not being resolved you can increase the number of\nrotation attempts cpptraj will make by increasing maxfactor.\nFor example, to rotate all backbone dihedrals in a protein with coordinates\ninafilenamedtz2.rst7in-120degreeintervalsandwritetheresultingtrajectory\nin Amber format to rotations.mdcrd:\nreference tz2.rst7 [TZ2]\npermutedihedrals crdset [TZ2] interval -120 outtraj rotations.mdcrd phi psi\nTo randomly rotate backbone dihedrals for the same structure and write to file\nrandom.mol2 in MOL2 format:\nreference tz2.rst7 [TZ2]\npermutedihedrals crdset [TZ2] random rseed 1 check maxfactor 10 phi psi \\\nouttraj random.mol2 multi\n7.13 prepareforleap\nprepareforleap crdset <coords set> [frame <#>] name <out coords set>\n[pdbout <pdbfile> [terbymol]]\n[leapunitname <unit>] [out <leap input file> [runleap <ff file>]]\n[skiperrors]\n[nowat [watermask <watermask>] [noh]\n[keepaltloc {<alt loc ID>|highestocc}]\n[stripmask <stripmask>] [solventresname <solventresname>]\n[molmask <molmask> ...] [determinemolmask <mask>]\n[{nohisdetect |\n[nd1 <nd1>] [ne2 <ne2] [hisname <his>] [hiename <hie>]\n[hidname <hid>] [hipname <hip]}]\n[{nodisulfides |\nexistingdisulfides |\n[cysmask <cysmask>] [disulfidecut <cut>] [newcysname <name>]}]\n[{nosugars |\nsugarmask <sugarmask> [noc1search] [nosplitres]\n[rescut <residue cutoff>] [bondoffset <offset>]\n[resmapfile <file>]\n[hasglycam] [determinesugarsby {geometry|name}]\n}]\ncrdset <coords set> COORDS data set containing\ncoordinates and topology to prepare.\n[frame <#>] Frame to use from COORDS set (default\nfirst).\n44"}, {"page": 45, "text": "name <out coords set> Output COORDS set containing\nprepared topology/coordinates.\n[pdbout <pdbfile>] Output PDB name.\n[terbymol] If specified, base TER cards on molecules\ninstead of PDB chains.\n[leapunitname <unit>] LEaP unit name to use when\nwriting to <leap input file> (i.e. the LEaP input\nfile will contain ’<unit> = loadpdb <pdbfile>’).\n[out <leap input file>] File containing LEaP input needed\nto read in the prepared system (loadpdb, bond\ncommands for disulfides, etc).\n[runleap <ff file>] If specified, CPPTRAJ will attempt to\nrun LEaP directly to generate a topology and\ncoordinates; <ff file> should contain the\nappropriate ’source’ commands for loading the\ndesired force field parameters. Will attempt to\nproduce topology <unit>.parm7 and coordinates\n<unit>.rst7.\n[skiperrors] If specified, the command will try to ignore\nany errors encountered. Can be useful for\ndebugging.\n[nowat] If specified, remove waters from the system.\n[watermask <watermask>] Mask selecting waters to\nremove (default ’:<solventresname>’).\n[noh] If specified, strip all hydrogen atoms from the\nsystem (recommended).\n[keepaltloc {<alt loc ID>|highestocc}] LEaP cannot handle\nalternate atom locations, so the command will choose\nlocation ’A’ by default. This can be changed to\neither <alt loc id> or the location with the highest\noccupancy if ’highestocc’ is specified.\n[stripmask <stripmask>] Mask of atoms to remove from\nthe system.\n[solventresname <solventresname>] Solvent residue name\n(default ’HOH’).\n[molmask <mask>] If specified, atoms in <mask> will\nbe considered all part of one molecule. May be\nspecified multiple times.\n[determinemolmask <mask>] If specified, determine if\natoms selected in <mask> are in the same molecule\nvia bonds.\n45"}, {"page": 46, "text": "Histidine Detection:\n[nohisdetect] Disable renaming of histidine residues\nbased on existing hydrogens.\n[nd1 <nd1>] Delta nitrogen atom name (default ’ND1’).\n[ne2 <ne2<] Epsilon nitrogen atom name (default ’NE2’).\n[hisname <his>] Histidine residue name (default ’HIS’).\n[hiename <hie>] Epsilon-protonated histidine name\n(default ’HIE’).\n[hidname <hid>] Delta-protonated histidine name\n(default ’HID’).\n[hipname <hip>] Doubly-protonated histidine name\n(default ’HIP’).\nDisulfide Handling:\n[nodisulfides] Disable handling of disulfides.\n[existingdisulfides] Only handle disulfides already\npresent; do not search for additional disulfides.\n[cysmask <cysmask>] Mask for selecting cysteine\nresidues (default ’CYS’).\n[disulfidecut <cut>] Sulfur to sulfur atom distance\ncutoff for forming a disulfide (default 2.5 Ang).\n[newcysname <name>] Name to change cysteine residues\nthat participate in a disulfide bond to (default\n’CYX’).\nSugar Handling:\n[nosugars] Disable handling of sugars.\n[sugarmask <sugarmask>] Mask selecting sugars to be\nhandled. If not specified the default is all\nresidues defined in resmapfile.\n[noc1search] If specified, disable search for missing\nlinkages to sugar C1 atom bonds.\n[nosplitres] If specified, do not attempt to split off\nfunctional groups from sugars into separate\nresidues.\n[rescut <residue cutoff>] Initial distance cutoff\n(default 8 Ang.) for residue center to residue\ncenter distance when looking for missing sugar\nlinkages.\n[bondoffset <offset>] Offset (default 0.2 Ang.) to add\nto “ideal” bond distances when looking for missing\nsugar linkages. Can be increased to accommodate\ndistorted structures.\n46"}, {"page": 47, "text": "[resmapfile <file>] File containing sugar residue/atom\nname mapping. Default is\n’$CPPTRAJHOME/dat/Carbohydrate_PDB_Glycam_Names.txt’.\n[hasglycam] If specified, assume sugars already have\nGLYCAM residue names; just check sugar anomer\ntype/configuration/linkage.\n[determinesugarsby {geometry|name}] Determine whether\nsugar anomer type/configuration should be chosen\nbased on sugar geometry (default) or the residue\nname. CPPTRAJ will report when a mismatch is\ndetected between the sugar anomer type/configuration\nbased on geometry and anomer type/configuration\nbased on the residue name.\nThiscommandwillprepareastructure(usuallyfromaPDB)forprocessingwith\nthe Amber program LEaP to generate topology and coordinates files for MD\nsimulations.[3] It will handle things like choosing alternate atom locations, re-\nmoving waters/hydrogen atoms from the structure, renaming residues and gen-\nerating ’bond’ commands for disulfide bonds, change histidine names based on\nany existing protonation, and renaming residues/atoms and generating ’bond’\ncommands for carbohydrates. The command can also call LEaP directly to\ngenerate the parameters once the structure is prepared.\nIfhydrogenatomsarepresentinthestructure, thecommandwillattempta\nsimple and straightforward determination of the protonation state of any histi-\ndineresiduesbasedonwherehydrogensarebonded, andassigntheappropriate\nresidue name. The command will also identify any existing disulfide bonds as\nwell as potential disulfide bonds and generate the corresponding LEaP ‘bond’\ncommandswhichcanbeappliedafterthestructureisloadedinLEaP.Potential\ndisulfidebondingatomscanbeidentifiedviaauser-specifiablemaskexpression.\nBydefault,sugarswillhavetheirresiduenameschangedtothosecompatible\nwith the GLYCAM force field based on their anomer type (alpha/beta), config-\nuration (D/L), and linkages (glycosidic and covalent sugar to non-sugar). Any\nrecognized functional groups that are part of sugar residues (hydroxyl, acetyl,\nsulfate, etc) will be split into separate residues as required by GLYCAM. If\nthis happens and ’runleap’ has not been specified, CPPTRAJ will warn about\nany residues/atoms that require charge to be adjusted. If ’runleap’ has not\nbeen specified the command will warn about any atoms that need to have their\ncharges adjusted after LEaP is run.\nThe command will try to report any potential problems that LEaP might\nencounter. These include residue names that may be unrecognized (and there-\nfore may not have parameters), mismatches between detected sugar anomer\ntype/configuration and anomer type/configuration based on the sugar residue\nname, unrecognized sugar linkages, and so on.\nFor example, the following input prepares PDB 4zzw for processing with\nPDB, putting the proper leap commands in leap.4zzw.in, writing the prepared\n47"}, {"page": 48, "text": "PDB to 4zzw.cpptraj.pdb, removing waters and hydrogen atoms, and keeping\nalternate atom locations with the highest occupancy:\nparm 4zzw.pdb\nloadcrd 4zzw.pdb name MyCrd\nprepareforleap crdset MyCrd name Final out leap.4zzw.in leapunitname m \\\npdbout 4zzw.cpptraj.pdb nowat noh keepaltloc highestocc\nSugar Residue/Atom Name Mapping File\nThisfilecontrolshowCPPTRAJwillnamesugarsbasedonsugarform/chirality\nlinkage. It consists of three sections separated by a blank line. The first section\ndefinessugarPDBresiduenamesandhowtheyaremappedtoGLYCAMresidue\ncharacters:\nFormat: <ResName> <GlycamCode> <Anomer> <Config> <RingType> \"<Name>\"\nAnomer: A=alpha, B=beta\nConfig: D/L\nRingType: P=pyranose, F=furanose\nExample: 64K A A D P \"alpha-D-arabinopyranose\"\nThe second section contains PDB to GLYCAM atom name maps for residues:\nFormat: <GLYCAM residue codes> <PDB atom name>,<GLYCAM atom name>[,<anomer>] ...\nIf <anomer> (A=alpha, B=beta) is specified, the atom name map is only valid for that specific form.\nExample: V,W,Y C7,C2N O7,O2N C8,CME\nThe third section contains PDB to GLYCAM linkage residue (i.e. non-sugar\nresidues bonded to sugars) name maps:\nFormat: <PDB residue name> <GLYCAM residue name>\nExample: SER OLS\n7.14 reference\nReference coordinates can now be used and manipulated like COORDS data\nsets. See 10.3 on page 86 for command syntax.\n7.15 rotatedihedral\nrotatedihedral crdset <COORDS set> [frame <#>] [name <output set name>]\n{value <value> | increment <increment>}\n{ <mask1> <mask2> <mask3> <mask4> |\nres <#> type <dih type> }\n<dih type> = alpha beta gamma delta epsilon zeta nu1 nu2 h1p c2p chin\nphi psi chip omega\n48"}, {"page": 49, "text": "crdset <COORDS set> Coordinates data set to work on.\nIf a TRAJ data set is specified, name must also be\nspecified.\n[frame <#>] Frame of the COORDS set to work on.\n[name <output set name>] Output COORDS set. If not\nspecified the input COORDS set will be modified.\nvalue <value> Set specified dihedral to given value in\ndegrees.\nincrement <increment> Increment specified dihedral by\nincrement in degrees.\n<mask1> <mask2> <mask3> <mask4> Define dihedral\nby atom masks. Each mask should only select one\natom.\nres <#> Rotate dihedral specified by type in residue\nnumber <#>.\ntype <dih type> Dihedral type to rotate in specified\nresidue.\nRotatethespecifieddihedralingivenCOORDSsettoatargetvalueorbygiven\nincrement. Forexample,tosettheproteinchidihedralinresidue8to35degrees\nand write out to a mol2 file:\nparm ../tz2.parm7\nloadcrd ../tz2.nc 1 1 name TZ2\nrotatedihedral crdset TZ2 value 35 res 8 type chip\ncrdout TZ2 tz2.rotate.1.mol2\n7.16 sequence\nsequence name <output set name> <unit0> <unit1> ...\n[{libset <libsetname>} ...]\nname <output set name> Name of final molecule.\n<unit0> <unit1> Name of COORDS set (with connection\ninfo).\n[{libset <libsetname>} ...} One or more set name prefixes\nof data sets (libraries) containing units.\nData sets created:\n<output set name> COORDS set containing final\nmolecule.\nConnect units in different COORDS sets together to form a single molecule.\nInternal coordinates are used to try to determine the correct geometry around\nconnection sites. The COORDS sets must have connection information set,\n49"}, {"page": 50, "text": "either from reading in an Amber OFF library file or set manually via dataset\nconnect (see 8.8 on page 55).\nFor example, the following reads in two Mol2 files as COORDS sets, sets up\nconnection atoms, then creates a molecule via sequence:\nparm MOC.mol2\nloadcrd MOC.mol2 parm MOC.mol2 name MOC\ndataset connect MOC tailmask @O5\nparm CNALA.mol2\nloadcrd CNALA.mol2 parm CNALA.mol2 name CNALA\ndataset connect CNALA headmask @N\nsequence MOC CNALA name Mol\ncrdout Mol Mol.mol2\n7.17 splitcoords\nsplitcoords <crd set> name <output set name>\n<crd set> COORDS set to split.\nname <output set name> Name of new set to create.\nSplit trajectory specified by <crd set> by molecule into a new COORDS set.\nAll molecules in <crd set> must be the same size. For example, if there are\n10 molecules and 10 frames in COORDS set “Set0”, the following would create\na new COORDS set with 100 frames (original molecules 1-10 frame 1, original\nmolecules 1-10 frame 2, etc):\nsplitcoords Set0 name Set0Split\n7.18 zmatrix\nzmatrix <COORDS set name> [name <output set name>]\n{ zset <input zmatrix set> [parm <top>|parmindex <#>] |\n[molnum <mol#>] [frame <frame#>] [out <zmatrix file>] }\n<COORDS set name> COORDS set to calculate Z-matrix\nfrom or use as topology for applied Z-matrix.\n[name <output set name>] Name of output COORDS set (if\n’zset’) or output Z-matrix set.\nzset <input zmatrix set> Name of Z-matrix set to use to\ngenerate coordinates.\nparm <top> Use specified topology name as\ntopology for generated coordinates.\nparmindex <#> Use topology index as topology for\ngenerated coordinates.\n[molnum <mol#>] Calculate Z-matrix from specified\nmolecule (default first molecule).\n50"}, {"page": 51, "text": "[frame <frame#>] Calculate Z-matrix from specified\nmolecule in specified frame (default first frame).\n[out <zmatrix file>] File to write calculated Z-matrix\nto.\nData sets created:\n<output set name> If ’zset’, COORDS set containing\nfinal coordinates. Otherwise contains Z-matrix\ndata.\nCommand for working with Z-matrices. If ’zset’ is specified, generate coordi-\nnates from the specified Z-matrix data set and topology. Otherwise, calculate a\nZ-matrix for a single molecule from the specified frame of given COORDS data\nset.\n8 General Commands\nThe following general commands are available:\n51"}, {"page": 52, "text": "Command Description\nactiveref Select the reference for distance-based masks.\ncalc Evaluate the given mathematical expression.\nclear Clear various objects from the cpptraj state.\ncreate Create (but do not yet write) a data file.\ncreateset Create a dataset from a simple mathematical expression.\ndatafile Used to manipulate data files.\ndatafilter Filter data sets based on given criteria.\ndataset Use to manipulate data sets.\ndebug | prnlev Set debug level. Higher levels give more info.\nensextension Enable/disable ensemble number extension for files in ensemble mode.\nexit | quit Quit cpptraj.\nflatten Distribute elements of 2d matrix across 1d array.\ngo | run Start a trajectory processing Run.\nhelp Provide help for commands.\nlist List various objects in the cpptraj state.\nnoexitonerror Attempt to continue even if errors are encountered.\nnoprogress Do not print a progress bar during a Run.\nparallelanalysis (MPI only) Divide current Analyses among MPI processes.\nparsedata Parse timing data from CPPTRAJ output.\nprecision Change the output precision of data sets.\nprintdata Print data set to screen.\nrandom Change default random number generator, create random sets.\nreaddata Read data sets from files.\nreadensembledata Read data files in ensemble mode.\nreadinput Read cpptraj input from a file.\nremovedata Remove specified data set(s).\nrst Generate Amber-style distance/angle/torsion restraints.\nrunanalysis Run an analysis immediately or run all queued analyses.\nselect Print the results of an atom mask expression.\nselectds Print the results of a data set selection expression.\nsilenceactions Prevent Actions from writing information to STDOUT.\nsortensembledata Sort data sets using replica information (currently constant pH only).\nusediskcache Turn caching of data sets to disk on or off.\nwrite | writedata Immediately write data to a file or write to all current data files.\n8.1 activeref\nactiveref <#>\nSet which reference structure should be used when setting up distance-based\nmasks for everything but the ’mask’ action. Numbering starts from 0, so ’ac-\ntiveref 0’ selects the first reference structure read in, ’activeref 1’ selects the\nsecond, and so on.\n52"}, {"page": 53, "text": "8.2 calc\ncalc <expression>\n[prec <width>.<precision>] [format {double|general|scientific}]\n<expression> Mathematical expression to evaluate. See\n5.2 on page 25 for details.\nprec <width>.<precision> Set the width and precision\nof the result.\nformat {double|general|scientific} Set the format of the\nresult.\nEvaluate the given mathematical expression. This version gives more control\nover the format of the output.\n8.3 clear\nclear [{all | <type>}]\n(<type> = actions,trajin,trajout,ref,parm,analysis,datafile,dataset)\nClear list of indicated type, or all lists if ’all’ specified. Note that when clearing\nactions or analyses, associated data sets and data files are not cleared and vice\nversa.\n8.4 create\ncreate <filename> <datasetname0> [<datasetname1> ...] [<DataFile Options>]\nAdd specified data sets to the data file named <filename>; if the file does not\nexist,itwillbeaddedtotheDataFileList. Datafilescreatedinthiswayareonly\nwritten at the end of coordinate processing, analyses, or via the ’writedata’\ncommand. See 6 on page 27 for more data file format options.\n8.5 createset\ncreateset <expression> [xmin <min>] xstep <step> nx <nxvals>\nexpression Simple mathematical expression, must contain\nequals sign, can contain X (e.g. Y=2*X). If not\nenclosed in quotes must not contain whitespace.\nxmin <min> Minimum X value.\nxstep <step> X step.\nnx <nxvals> Number of X values.\nGenerate a data set from a simple mathematical expression.\n53"}, {"page": 54, "text": "8.6 datafile\ndatafile <filename> <datafile arg>\nPass <datafile arg> to datafile <filename>. See 6 on page27 for more details.\n8.7 datafilter\ndatafilter {<dataset arg> min <min> max <max> ...} [out <file>] [name <setname>]\n{[multi] | [filterset <set> [newset <newname>]] [countout <countfile>]}\n<dataset arg> min <min> max <max> Data set name\nand min/max cutoffs to use; can specify more than\none.\n[out <file>] Write out to file named <file>.\n[name <setname>] Name of filter data set containing 1\nwhen cutoffs satisfied, 0 otherwise.\n[multi] Filter each set separately instead of all\ntogether (creates filter set for each input set).\nCannot be used with ’filterset’.\n[filterset <set>] If specified, <set> will be filtered to\nonly contain data that satisfies cutoffs. Cannot be\nused with ’multi’.\n[newset <newname>] If specified a new set will be\ncreated from ’filterset’ instead of replacing\n’filterset’.\n[countout <count>] If specified, write number of\nelements passed and filtered to <countfile>. Cannot\nbe used with ’multi’.\nSets Created (not ’multi’)\n<setname> For each input element contains 1 for\nelements that “passed”, 0 otherwise.\n<setname>[npassed] Number of elements that passed.\n<setname>[nfiltered] Number of elements filtered out.\nSets Created (’multi’)\n<setname>:<idx> For each input set (number with\n<idx>, starting from 0) contains 1 for elements that\n“passed”, 0 otherwise.\nCreate a data set (optionally named <setname>) containing 1 for data within\ngiven <min> and <max> criteria for each specified data set. There must be\nat least one <min> and <max> argument, and can be as many as there are\nspecifieddatasets. If ’multi’isspecifiedthenonlyfilterdatasetswillbecreated\n54"}, {"page": 55, "text": "for each data set instead. If ’filterset’ is specified, the specified <set> will be\nmodified to only contain ’1’ frames; cannot be used with ’multi’. If ’newset’\nis also specified, a new set will be created containing the ’1’ frames instead.\nThe ’filterset’ functionality only works for 1D scalar sets. If ’countout’ is\nspecified, the final number of elements passed and filtered out will be written\nto <countfile>.\nForexample,toreadindatafromtwoseparatefiles(d1.datanda1.dat)and\ngenerate a filter data set named FILTER having 1 when d1 is between 0.0 and\n3.0 and a1 is between 135.0 and 180.0:\nreaddata a1.dat name a1\nreaddata d1.dat name d1\ndatafilter d1 min 0.0 max 3.0 a1 min 135.0 max 180.0 out filter.dat name FILTER\nNote that a similar command that can be used with data generated by Actions\nduring trajectory processing is filter (see page 133).\n8.8 dataset\ndataset { legend <legend> <set> |\nmakexy <Xset> <Yset> [name <name>] |\nvectorcoord {X|Y|Z} <set> [name <name>] |\ncat <set0> <set1> ... [name <name>] [nooffset] |\nmake2d <1D set> cols <ncols> rows <nrows> [name <name>] |\n{drop|keep}points {range <range arg> | [start <#>] [stop <#>] [offset <#>]}\n[name <output set>] <set arg1> ... |\nremove <criterion> <select> <value> [and <value2>] [<set selection>] |\nconnect {[head <head atom>] [tail <tail atom>]|[headmask <headmask>] [tailmask <tailmask>]}\ndim {xdim|ydim|zdim|ndim <#>} [label <label>] [min <min>] [step <step>] |\noutformat {double|scientific|general} <set arg1> [<set arg 2> ...] |\ninvert <set arg0> ... name <new name> [legendset <set>] |\nshift [above <value> by <offset>] [below <value> by <offset>] <set arg0> ...\n[mode <mode>] [type <type>] <set arg1> [<set arg 2> ...]\n}\n<mode>: ’distance’ ’angle’ ’torsion’ ’pucker’ ’rms’ ’matrix’ ’vector’\n<type>: ’alpha’ ’beta’ ’gamma’ ’delta’ ’epsilon’ ’zeta’ ’nu0’ ’nu1’ ’nu2’ ’nu3’\n’nu4’ ’h1p’ ’c2p’ ’chin’ ’phi’ ’psi’ ’chip’ ’omega’ ’chi2’ ’chi3’ ’chi4’\n’chi5’ ’pucker’ ’noe’ ’distance’ ’covariance’ ’mass-weighted covariance’\n’correlation’ ’distance covariance’ ’IDEA’ ’IRED’ ’dihedral covariance’\nOptions for ’type noe’:\n[bound <lower> bound <upper>] [rexp <expected>] [noe_strong] [noe_medium]\n[noe_weak]\n[name <name>] New data set name for\nmakexy/vectorcoord/cat/make2d/droppoints/keeppoints.\nlegend <legend> <set> Set the legend for data set\n<set> to <legend>.\n55"}, {"page": 56, "text": "makexy <Xset> <Yset> Create a new data set\n(optionally named <name>) with X values from <Xset>\nand Y values from <Yset>.\nvectorcoord {X|Y|Z} <set> Extract X/Y/Z coordinates\nfrom vector data set into a new 1D data set.\ncat <set0> <set1> ... Concatenate two or more data sets\ninto a new data set (optionally named <name>). Only\nworks for scalar 1D and string sets.\nmake2d <1D set> cols <ncols> rows <nrows> Convert\n1D data set into row-major 2D data set with\nspecified number of rows and columns.\n{drop|keep}points <set arg1> ... Drop or keep specified\npoints from data set(s), optionally creating a new\ndata set.\nrange <range arg> Range of points to drop/keep.\n[start <#>] [stop <#>] [offset <#>]\nStart/stop/offset values of points to drop/keep.\nremove <criterion> <select> <value> [and <value2>] [<set selection>]\nRemove data sets from <set selection> according to\nspecified criterion and selection.\n<criterion>: ’ifaverage’ ’ifsize’ ’ifmode’ ’iftype’\n<select> : ’equal’ ’==’ ’notequal’ ’!=’ ’lessthan’ ’<’\n’greaterthan’ ’>’ ’between’ ’outside’\nconnect <set args> Add/change connect atom information\n(used by the sequence command, 7.16 on page 49) to\nCOORDS set(s). Head atoms connect to previous\nresidue, tail atoms connect to next residue. Can\nuse either absolute atom numbers or atom mask\nexpressions.\n[head <head atom>] [tail <tail atom>] Specify the\nhead/tail atoms by atom number.\n[headmask <headmask>] [tailmask <tailmask>]\nSpecify the head/tail atoms by atom mask.\ndim {xdim|ydim|zdim|ndim <#>} Change specified\ndimension in set(s).\nlabel <label> Change dimension label to <label>\nmin <min> Change dimension minimum to <min>.\nstep <step> Change dimension step to <step>.\ninvert <set arg0> ... name <new name> [legendset <set>]\n56"}, {"page": 57, "text": "<set arg0> ... Specify sets to invert.\nname <new name> Inverted output set name.\n[legendset <set>] String data set containing legends\nshift\n[above <value> by <offset>] Values in set(s) above\n<value> will be shifted by <offset>.\n[below <value> by <offset>] Values in set(s) below\n<value> will be shifted by <offset>.\n<set arg0> ... Set(s) to shift.\n[mode <mode>] Set data set(s) mode to <mode>.\n[type <type>] Set data set(s) type to ’type’, useful\nfor e.g. analysis with statistics. Note this can\nalso be done with ’type <type>’ for certain commands\n(distance, dihedral, pucker etc). Note that not\nevery <type> is compatible with a given <mode>.\nOptions for ’type noe’ only:\n[bound <lower> bound <upper>] Lower and upper bounds\nfor NOE (in Angstroms); must specify both.\n[rexp <expected>] Expected value for NOE (in\nAngstroms); if not given ’(<lower> + <upper>)’ / 2.0\nis used.\n[noe_strong] Set lower and upper bounds to 1.8 and 2.9\nÅ respectively.\n[noe_medium] Set lower and upper bounds to 2.9 and 3.5\nÅ respectively.\n[noe_weak] Set lower and upper bounds to 3.5 and 5.0 Å\nrespectively.\nEither set the legend for a single data set, create a new set with X values from\none set and Y values from another, concatenate 2 or more sets, make a 2D\nset from 1D set, remove sets according to a certain criterion, or change the\nmode/type for one or more data sets.\nSetting the mode/type can be useful for cases where the data set is being\nread in from a file; for example when reading in a dihedral data set the type\ncan be set to ’dihedral’ so that various Analysis routines like statistics know\nto treat it as periodic. A brief description of possible modes and types follows:\n57"}, {"page": 58, "text": "Mode Type Description\ndistance noe NOE distance.\nangle Angle.\ntorsion alpha Nucleic acid alpha.\nbeta Nucleic acid beta.\ngamma Nucleic acid gamma.\ndelta Nucleic acid delta.\nepsilon Nucleic acid epsilon.\nzeta Nucleic acid zeta.\nnu1 Nucleic pucker (O4’).\nnu2 Nucleic pucker (C4’).\nh1p Nucleic acid H1’.\nc2p Nucleic acid C2’.\nchin Nucleic acid chi.\nphi Protein Phi.\npsi Protein psi.\nchip Protein chi.\nomega Protein omega.\npucker pucker Sugar pucker.\nrms RMSD.\nmatrix distance Distance matrix.\ncovariance Cartesian covariance matrix.\n’mass-weighted covariance’ Mass weighted Cartesian covariance matrix.\ncorrelation Dynamic cross correlation matrix.\n’distance covariance’ Distance covariance matrix.\nIDEA IDEA matrix.\nIRED IRED matrix.\n’dihedral covariance’ Dihedral covariance matrix.\nvector IRED IRED vector.\nThe invert mode takes a group of M 1D data sets of size N and create N\nnew\"inverted\"datasetsofsizeM.Thisissimilartotheinvertkeywordalready\navailableforstandardandGracedatawrites,butoperatesdirectlyondatasets.\nFor example, given the following two data sets:\nD0 D1\n1 4\n2 5\n3 6\nThe new data sets will be laid out like so:\nN0 N1 N2\n1 2 3\n4 5 6\n58"}, {"page": 59, "text": "The dataset invert command can be useful if you want to easily view output\nfrom multiple analysis commands in a single graph. For example, to view state\ncounts from two different simulations side by side:\ncalcstate name Sim1 state bound1,dist1,0.0,2.0\ncalcstate name Sim2 state bound1,dist1,0.0,2.0\nrunanalysis dataset invert Sim*[Count] name Inverted legendset Sim1[Name]\ndataset dim xdim label Simulation min 1 step 1 Inverted*\nwritedata statecount.agr Inverted*\nThe dataset shift command can be used for wrapping circular values, such as\ntorsions. For example, to ensure a pucker has a range from 0 to 360 instead of\n-180 to 180:\npucker Furanoid @C2 @C3 @C4 @C5 @O2 cremer out CremerF.dat amplitude\nrun\ndataset shift Furanoid below 0 by 360\n8.9 debug | prnlev\ndebug [<type>] <#>\n(<type> = actions,trajin,trajout,ref,parm,analysis,datafile,dataset)\nSet the level of debug information to print. In general the higher the <#> the\nmoreinformationthatisprinted. If<type>isspecifiedonlysetthedebuglevel\nfor a specific area of cpptraj.\n8.10 ensextension\nensextension {on|off}\nTurn printing of ensemble member number filename extensions on or off. By\ndefault ensemble extensions are printed in parallel and not in serial.\nNOTE:THE’ensextensionoff’OPTIONHASNOTBEENFULLY\nTESTED IN PARALLEL AND IS NOT CURRENTLY RECOM-\nMENDED.\n8.11 exit | quit\nExit normally.\n8.12 flatten\nflatten name <output set name> [mode {sum|avg}] <input set args>\n59"}, {"page": 60, "text": "name <output set name> Name of “flattened” 1D output\nset(s).\nmode {sum|avg} If sum, matrix elements will be summed.\nIf avg, matrix elements will be averaged.\n<input set args> Specify matrices to “flatten”.\nDataSets Created\n<output set name> Flattened 1D set if only one input\nmatrix.\n<output set name>:<idx> Flattened 1D sets when more\nthan one input matrix; index starts from 1.\nFlatten1ormorematricesinto1Darray(s)bysummingoraveragingelements.\nFor example, given a matrix with values like this:\nX Y Value\n1 3 5.0\n1 4 4.0\n2 3 2.0\nThe “flattened” 1D array with mode SUM would be determined as follows:\nElement 1 = (5.0/2) + (4.0/2) = 4.5\nElement 2 = (2.0/2) = 1.0\nElement 3 = (5.0/2) + (2.0/2) = 3.5\nElement 4 = (4.0/2) = 2.0\nAnd the final 1D array would look like so:\nIndex Value\n1 4.5\n2 1.0\n3 3.5\n4 2.0\n8.13 go | run\nBegin trajectory processing, followed by analysis and datafile write.\n8.14 help\nhelp [ { All |\n<cmd> |\n<command category> |\nForm[ats] [{read|write}] |\nForm[ats] [{trajin|trajout|readdata|writedata|parm|parmwrite} [<fmt key>]] |\nMask } ]\n60"}, {"page": 61, "text": "Command Categories: Gen[eral] Sys[tem] Coor[ds] Traj[ectory] Top[ology]\nAct[ion] Ana[lysis] Con[trol]\nAll : Print all known commands.\n<cmd> : Print help for command <cmd>.\n<command category> : Print all commands in specified category.\nForm[ats] : Help for file formats.\nMask : Help for mask syntax.\nIf’All’isspecifed,listallcommandsknowntocpptraj. Ifgivenwithacommand,\nprinthelpforthatcommand. Otherwise,listallcommandsofacertaincategory\n(General, System, Coords, Trajectory, Topology, Action, Analysis, or Control),\nhelp for various file formats, or help with atom mask syntax.\n8.15 list\nlist <type>\n(<type> = actions,trajin,trajout,ref,parm,analysis,datafile,dataset)\nList the currently loaded objects of <type>. If no type is given then list all\nloaded objects.\n8.16 noexitonerror\nnoexitonerror\nNormallycpptraj willexitifactionsfailtoinitializeproperly. If noexitonerror\nisspecified,cpptraj willattempttocontinuepastsucherrors. Thisisthedefault\nif in interactive mode.\n8.17 noprogress\nnoprogress\nDo not display read progress during trajectory processing.\n8.18 parallelanalysis\nparallelanalysis [sync]\nMPI only. Divide all currently set up analyses as evenly as possible among\navailable MPI processes and execute. Each analysis will get a single MPI pro-\ncess. If sync is specified all data will be synced back to the master process (for\ne.g. subsequent analysis). For an example of how to use the parallelanalysis\ncommand, see 12.16 on page 250.\n61"}, {"page": 62, "text": "8.19 parsedata\nparsetiming <filename args> ... [out <file>] [name <setname>]\n[sortby {time|cores|filename}] [includebad] [showdetails]\n[type {trajproc|trajread|actframe}] [reverse]\n[groupout <file> [grouptype {prefix|name|kind}]]\n<filename args> Files containing CPPTRAJ output to get\ntiming data from.\n[out <file>] Write total sorted timing sets to <file>.\n[name <setname>] Set name for timing data sets.\n[sortby {time|cores|filename}] Sort timing data sets by\neither time, number of cores (MPI processes * OpenMP\nthreads), or file name.\n[includebad] If specified, include run output for which\ntiming data cannot be extracted (e.g. an\nincomplete/failed run).\n[showdetails] If specified, details about each run will\nbe printed to STDOUT.\n[type {trajproc|trajread|actframe} If specified, report\ntime other than the total time (which is the\ndefault):\ntrajproc Total trajectory processing time\n(trajectory I/O plus action frame time).\ntrajread Trajectory read time (requires compiling\nwith ’-timer’ configure option/-DTIMER compiler\ndefine).\nactframe Action frame time (requires compiling with\n’-timer’ configure option/-DTIMER compiler\ndefine).\n[reverse] Instead of longest time to shortest time, sort\nshortest time to longest time.\n[groupout <file>] Group run output by a property and\nwrite to a file. Additional details will be written\nlike speedup and efficiency (relative to the slowest\nrun).\n[grouptype {prefix|name|kind}] Property to group\nruns by in group output file.\nprefix Group by directory prefix.\nname Group by run type name (see\n<setname>[name] below).\nkind Group by run type name; OpenMP runs are\nseparated by number of OpenMP threads.\n62"}, {"page": 63, "text": "DataSets Created:\n<setname> Set containing sorted run times.\n<setname>[name] Set containing shortedhand run names.\nConsists of a prefix (S for serial, O for OpenMP, M\nfor MPI, H for hybrid MPI/OpenMP) followed by\nnumbers indicating number of MPI processes ’x’\nOpenMP threads; e.g., H16x4 means a hybrid run\nconsisting of 16 MPI processes and 4 OpenMP threads\nper process. If CUDA is active, the name will be\nwrapped in ’G()’, e.g. G(H16x4).\n<setname>[dir] Set containing output directory\nprefixes. If no directory prefix, just contains\noutput file name.\nThe parsedata command can be used to extract timing data from CPPTRAJ\noutput.\n8.20 precision\nprecision {<filename> | <dataset arg>} [<width>] [<precision>]\nSettheprecisionforalldatasetsindatafile<filename>ordataset(s)specified\nby<datasetarg>towidth.precision, wherewidthisthecolumnwidthandpre-\ncisionisthenumberofdigitsafterthedecimalpoint. Notethatthe<precision>\nargument only applies to floating-point data sets.\nFor example, if one wanted to set the precision of the output of an Rmsd\ncalculation to 8.3, the input could be:\ntrajin ../run0.nc\nrms first :10-260 out prec.dat\nprecision prec.dat 8 3\nand the output would look like:\n#Frame RMSD_00000\n1 0.000\n2 0.630\n8.21 random\nrandom [setdefault {marsaglia|stdlib|mt|pcg32|xo128}]\n[createset <name> count <#> [seed <#>]\nsettype {int|float01|gauss [mean <mean>] [sd <SD>]}]\nsetdefault If specified, change the default random\nnumber generator (RNG).\n63"}, {"page": 64, "text": "marsaglia Use the Marsaglia RNG that is used in the\nAmber MD programs sander/pmemd.\nstdlib Use the C standard library RNG.\nmt Use the C++11 implementation of the Mersenne\ntwister (mt19937); only available with C++11\nsupport.\npcg32 Use the 32 bit version of the Permuted\nCongruential Generator.[4]\nxo128 Use the Xoshiro128++ RNG.[5]\ncreateset If specified, create a 1D data set filled with\nrandom numbers of the specified type.\n<name> Name of created set.\ncount <#> The number of elements to put into the\nset.\nsettype {int|float01|gauss} Type of numbers to use;\ninteger, floating point between 0 and 1,\nGaussian distribution.\nmean <mean> Mean of distribution for ’gauss’.\nsd <SD> Standard deviation of distribution for\n’gauss’.\nseed <#> Optional seed for the RNG.\nThis command can be used to set the default random number generator used\nin CPPTRAJ, and/or create a 1D data set filled with random values.\n8.22 readdata\nreaddata <filename> [name <dsname>] [as <fmt>] [separate] [<format options>]\nname <dsname> Name for read-in data set(s). Default\nis <filename>.\nas <fmt> Force <filename> to be read as a specific\nformat using given format keyword.\nseparate Read each file specified into separate data\nsets indexed from 0.\nRead data from file <filename> and store as data sets. For more information\nonformatscurrentlyrecognizedbycpptrajsee1onpage24. Forformat-specific\noptions see 6. For example, given the file calc.dat:\n#Frame R0 D1\n1 1.7 2.22\nThecommand’readdata calc.dat’wouldreaddataintotwodatasets,calc.dat:2\n(legend set to “R0”) and calc.dat:3 (legend set to “D1”).\n64"}, {"page": 65, "text": "8.23 readensembledata\nreadensembledata <filename> [filenames <additional files>] [<readdata args>]\n<filename> Lowest replica file name.\nfilenames <additional files> Specified additional members\nof the ensemble. If not specified ensemble members\nwill be search for using numerical extensions.\n<readdata args> Additional data file arguments.\nReaddatasetsasanensemble,i.e. eachfileisadifferentmemberofanensemble.\nThis command is MPI-aware.\nIf one filename is given, it is assumed it is the \"lowest\" member of an en-\nsemble with a numerical extension, e.g. ’file.001’ and the remaining files are\nsearched for automatically. Otherwise all other members of the ensemble can\nbespecifiedwith’filenames’andacomma-separatedliste.g. ’file.001filenames\nfile.002,file.003,file.004. Foradditional’readdata’argumentsthatcanbepassed\nin see 6 on page 27.\nFor example, to read in data files named cpout.001 to cpout.006 automati-\ncally:\nreadensembledata cpout.001 cpin cpin name PH\nOr specified:\nreadensembledata cpout.001 \\\nfilenames cpout.002,cpout.003,cpout.004,cpout.005,cpout.006 \\\ncpin cpin name PH\n8.24 readinput\nreadinput <filename>\nRead cpptraj commands from file <filename>.\n8.25 removedata\nremovedata <arg>\nRemove data set corresponding to <arg>.\n8.26 rst\nrst <mask1> <mask2> [<mask3>] [<mask4>]\nr1 <r1> r2 <r2> r3 <r3> r4 <r4> rk2 <rk2> rk3 <rk3>\n{[parm <parmfile / tag> | parmindex <#>]}\n[{ref <refname> | refindex <#> | reference} [offset <off>] [width <width>]]\n[out <outfile>]\n65"}, {"page": 66, "text": "<mask1> (Required) First atom mask.\n<mask2> (Required) Second atom mask. If only two\nmasks assume distance restraint.\n[<mask3>] (Optional) Third atom mask. If 3 atom masks\nassume angle restraint.\n[<mask4>] (Optional) Fourth atom mask. If 4 atom\nmasks assume dihedral restraint.\nrX <rX> Value of RX (X=1-4, default 0.0)\nrk2 <rk2> Value of RK2 (force constant to be applied\nwhen R is R1 <= R < R2)\nrk3 <rk3> Value of RK3 (force constant to be applied\nwhen R is R3 <= R < R4)\n[parm <parmfile / tag> | parmindex <#>] Topology to\nbe used for atom masks.\n{ref <refname> | refindex <#> | reference} Use\ndistance/angle/dihedral in reference structure to\ndetermine values for r1, r2, r3, and r4. The value\nof r2 is set to <r2> + <off>, r3 = r2, r1 = r2 -\n<width>, r4 = r3 + <width>.\n[offset <off>] (Reference only) Value to offset\ndistance/angle/torsion in reference by (default\n0.0).\n[width <width>] (Reference only) Width between r1 and\nr2, r3 and r4 (default 0.5).\n[out <outfile>] Write restraints to outfile. If not\nspecified, write to STDOUT.\nGenerateAmber-styledistancerestraintsforusewithnmropt=1. Thisispartic-\nularlyusefulforgeneratingdistancerestraintsbasedoffofreferencecoordinates.\nFor example to generate a distance restraint between two C5’ atoms using the\ncurrent distance between them in a reference structure, offsetting the distance\nby 1.0 Ang.:\nparm 30bp-longbox-tip3p-na.parm7\nreference 30bp-longbox.rst7\nrst :1@C5’ :31@C5’ reference offset 1.0 rk2 10.0 rk3 10.0 out output\n8.27 runanalysis\nrunanalysis [<analysiscmd> [<analysis args>]]\nRun given analysis command immediately and write any data generated. If no\ncommandisgivenrunanyanalysiscurrentlysetup. NOTE:When’runanalysis’\nisspecifiedalone,dataisnotautomaticallywritten;towritedatageneratedwith\n’runanalysis’ use the ’writedata’ command (this allows multiple analysis runs\nbetween output if desired).\n66"}, {"page": 67, "text": "8.28 select\nselect <mask>\nPrints the number of selected atoms corresponding to the given mask, as well\nas the atom numbers with format:\nSelected= <#atom1> <#atom2> ...\nThis does not affect the state in any way, but is intended for use in scripts etc.\nfor testing the results of a mask expression.\n8.29 selectds\nselectds <dataset arg>\nShow the results of a data set selection. Data set selection has the format:\n<name>[<aspect>]:<index>\nEitherthe[<aspect>]orthe<index>argumentsmaybeomitted. A’*’canbe\nused in place of <name> or [<aspect>] as a wildcard. The <index> argument\ncan be a single number or a range separated by ’-’ and ’,’.\nThiscommanddoesnotaffectthestateinanyway,butisparticularlyuseful\nin interactive mode for determining the results of a dataset argument.\n8.30 sortensembledata\nsortensembledata <dset arg0> [<dset arg1> ...]\n<dset arg0> [<dset arg1> ...] Data set(s) to sort.\nSort unsorted data sets. Currently only works for constant pH REMD data.\n8.31 usediskcache\nusediskcache {on|off}\nIf on, CPPTRAJ will attempt to cache data sets to disk if possible. This\ncurrently only works for integer data sets (e.g. hbond series data sets, etc).\n8.32 write | writedata\nwrite [<filename> <datasetname0> [<datasetname1> ...]] [<DataFile Options>]\nWithnoarguments,writeallfilescurrentlyinthedatafilelist. Otherwise,write\nspecified data set(s) to <filename>. This is like the ’create’ command except a\ndata file is not added to the data file list; it is written immediately. See 6 on\npage 27 for more data file format options.\n67"}, {"page": 68, "text": "8.33 System Commands\nThese commands call the equivalent external system commands.\ngnuplot <args> Call gnuplot (if it is installed on your system) with the\ngiven arguments.\nhead <args> Call head, which lists the first few lines of a file.\nless <args> Call less, which can be used to view the contents of a file.\nls <args> List the contents of a directory.\npwd <args> Print the current working directory.\nxmgrace <args> Call xmgrace (if it is installed on your system) with the\ngiven arguments.\n9 Topology File Commands\nThese commands control the reading and writing of topology files. Cpptraj\nsupports the following topology file formats:\nFormat Keyword Extension Notes\nAmber Topology amber .parm7 Only fully-supported format for write.\nPDB pdb .pdb Read Only\nMol2 mol2 .mol2 Read Only\nCIF cif .cif Read Only\nCharmm PSF psf .psf Limited Write\nGromacs Topology gromacs .top Read only\nSDF sdf .sdf Read Only\nTinker ARC arc .arc Read Only\nFor most commands that require a topology one can be specified via two\nkeywords:\nparm [<name>] Select topology corresponding to given file name, tag, or\nname.\nparmindex [<#>] Select topology by order in which it was loaded, starting\nfrom 0.\nThe following topology related commands are available:\n68"}, {"page": 69, "text": "Command Description\nangleinfo, angles, printangles Print angle info for selected atoms.\natominfo, atoms, printatoms Print details for selected atoms.\nbondinfo, bonds, printbonds Print bond info for selected atoms.\nbondparminfo Print the bond parameter table.\nchange Change specified parts of a topology.\ncharge Print total charge for selected atoms.\ncomparetop Compare two topologies and report differences.\ndihedralinfo, dihedrals, printdihedrals Print dihedral info for selected atoms.\nhmassrepartition Perform hydrogen mass repartitioning.\nimproperinfo, impropers, printimpropers Print improper info for selected atoms.\nmass Print total mass for selected atoms.\nmolinfo Print molecule info for selected atoms.\nparm Load a topology file.\nparmbox Modify box info for a loaded topology.\nparminfo Print details for selected topology.\nparmstrip Remove selected atoms from topology.\nparmwrite Write selected topology to file.\nprintub, ubinfo Print Urey-Bradley info for selected atoms.\nresinfo Print residue info for selected atoms.\nscaledihedralk Scale selected dihedral force constants.\nsolvent Change which molecules are considered solvent.\nupdateparameters Update/add parameters in/to a topology.\n9.1 angleinfo | angles | printangles\nangleinfo [parm <name> | parmindex <#> | <#>] [<mask1>] [<mask2> <mask3>]\n[out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n[<mask1>] Mask to print angle info for.\n[<mask2> <mask3>] If specified, angles must match\nall masks.\n[out <file>] File to print to (default STDOUT).\nPrint angle information of atoms in <mask> for selected topology (first loaded\ntopology by default) with format:\n# Angle Kthet degrees atom names (numbers)\nWhere Angle is the internal angle index, Kthet is the angle force constant,\ndegrees is the angle equilibrium value, atom names shows the atoms involved\n69"}, {"page": 70, "text": "intheanglewithformat:<residue num>@<atom name>,and(numbers)shows\nthe atom indices involved ina comma-separated list. Atomtypes will be shown\nin the last column.\nIf 3 masks are given instead of 1, print info for angles with first atom in\n<mask1>, second atom in <mask2>, and third atom in <mask3>.\n9.2 atominfo | atoms | printatoms\natominfo [parm <name> | parmindex <#> | <#>] <mask> [out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n<mask> Mask selecting atoms to print info for.\n[out <file>] File to print to (default STDOUT).\nPrintinformationonatomsin<mask>forselectedtopology(firstloadedtopol-\nogy by default) with format:\n#Atom Name #Res Name #Mol Type Charge Mass GBradius El [rVDW] [eVDW]\nwhere#Atomistheinternalatomindex,thefirstNamecolumnistheatomname,\n#Res is the atom’s residue number, the second Name column is residue name,\n#Molistheatom’smoleculenumber,Typeistheatom’stype(certaintopologies\nonly), Charge is the atom charge (in units of electron charge), Mass is the\natom’s mass (in amu), GBradius is the generalized Born radius of the atom\n(Amber topologies only), and El is the 2 character element string. The final\ntwo columns are only shown if the topology contains non-bonded parameters:\nrVDW is the atom’s Lennard-Jones radius and eVDW is the atom’s Lennard-Jones\nepsilon.\n9.3 bondinfo | bonds | printbonds\nbondinfo [parm <name> | parmindex <#> | <#>]\n[<mask1>] [<mask2>] [out <file>] [nointrares]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n[<mask1>] Mask to print bond info for.\n[<mask2>] If specified, bonds must match both masks.\n[out <file>] File to print to (default STDOUT).\n[nointrares] Do not print intra-residue bonds.\n70"}, {"page": 71, "text": "Printbondinformationforatomsin<mask>forselectedtopology(firstloaded\ntopology by default) with format:\n# Bond Kb Req atom names (numbers)\nwhere Bond is the internal bond index, Kb is the bond force constant, Req is the\nbond equilibrium value (in Angstroms), atom names shows both atom names\nwith format :<residue num>@<atom name>, and (numbers) shows both atom\nnumbers in a comma-separated list. Atom types will be shown in the last\ncolumn.\nIf 2 masks are given instead of 1, print info for bonds with first atom in\n<mask1> and second atom in <mask2>.\n9.4 bondparminfo\nbondparminfo [parm <name> | crdset <set> | parmindex <#> | <#>] [out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n[out <file>] File to print to (default STDOUT).\nPrint the bond parameter table with format:\n#Idx Rk Req\nWhere Idx is the internal bond parameter index, Rk is the bond force constant\n(in kcal/mol*Ang^2), and Req is the bond equilibrium value (in Ang.).\n9.5 change\nchange [parm <name> | parmindex <#> | <#> |\ncrdset <COORDS set> ]\n{ resname from <mask> to <value> |\nchainid of <mask> to <value> |\noresnums of <mask> min <range min> max <range max> |\nicodes of <mask> min <char min> max <char max> resnum <#> |\natomname from <mask> to <value> |\naddbond <mask1> <mask2> [req <length> <rk> <force constant>]\nremovebonds <mask1> [<mask2>] [out <file>] |\nbondparm <mask1> [<mask2>] {setrk|scalerk|setreq|scalereq} <value> |\n{mass|charge} [of <mask>] {to <value> |by <offset> |\nbyfac <factor> |fromset <data set>} |\nmergeres firstres <start res#> lastres <stop res#>\n}\nparm <name> | parmindex <#> | <#> | crdset <COORDS set>\nTopology to change.\n71"}, {"page": 72, "text": "resname from <mask> to <value> Change residue names\nfor residues in <mask> to the given <value>.\nchainid of <mask> to <value> Change the chain ID of\nresidues in <mask> to given <value>.\noresnums of <mask> min <range min> max <range max>\nChange original residue numbers (to e.g. original\nPDB numbers) of residues in <mask> to a range\nstarting from <min> and ending with <max>.\nicode of <mask> min <char min> max <char max> <resnum> <#>\nChange residue insertion codes of residues in <mask>\nto a range of characters starting from <min> and\nending with <max>; set the original residue number\nto <resnum>.\natomname from <mask> to <value> Change atom names\nfor atoms in <mask> to the given <value>.\naddbond <mask1> <mask2> Add bond between atom\nspecified by <mask1> and atom specified by <mask2>.\n[req <length>] The equilibrium bond length in\nAngstroms.\n[rk <force constant>] The bond force constant in\nkcal/mol*Angstrom.\nremovebonds <mask1> [<mask2>] Remove bonds from\natoms in <mask1>. If <mask2> also given, remove\nbonds between atoms in <mask1> and atoms in <mask2>.\n[out <file>] If specified, write removed bonds to\n<file> with format ’<residue name> <residue num>\n<atom name> <atom num>’.\nbondparm <mask1> [<mask2>] {setrk|scalerk|setreq|scalereq} <value>\nModify bond parameters in bonds selected by <mask1>\n(and <mask2> if specified) by specified <value>.\nsetrk Set bond force constants to <value>.\nscalerk Scale bond force constants by <value>.\nsetreq Set bond equilibrium lengths to <value>.\nscalereq Scale bond equilibirum lengths by <value>.\nmass|charge Change mass or charge in specified\ntopology.\nof <mask> Atoms to change mass/charge of.\nto <value> Value to change mass/charge to.\nby <offset> Value to offset masses/charges by.\nbyfac <factor> Value to multiply masses/charges by.\n72"}, {"page": 73, "text": "fromset <data set> Use values in <data set> for\nmass/charge; must have the same number of values\nas atoms selected by <mask>.\nmergeres Merge consecutive residues.\nfirsres <start res#> Index (starting from 1) of\nfirst residue to merge.\nlastres <stop res#> Index (starting from 1) of last\nresidue to merge. Should be greater than\nfirstres.\nChange specified parts of the specified topology. For example, to change atoms\nnamed ’HN’ to ’H’ in topology 0:\nchange parmindex 0 atomname from @HN to H\n9.6 charge\ncharge [parm <name> | parmindex <#> | <#>] <mask> [out <file>] [name <set>]\nparm <name> | parmindex <#> Topology to calculate\ncharge from.\n<mask> Atom(s) to calculate total charge for (default\nall).\n[out <file>] File to write total charge to.\n[name <set>] If specified, a data set named <set> will\nbe cretaed containing total charge.\nPrint the total charge of atoms in <mask> (in units of electron charge) for\nselected topology (first loaded topology by default).\n9.7 comparetop\ncomparetop {parm <name> | parmindex <#>} {parm <name> | parmindex <#>} [out <file>]\n[atype] [lj] [bnd] [ang] [dih] [atoms]\nparm <name> | parmindex <#> Topologies to compare.\nout <file> Print results to file instead of screen.\n[atype] Only report atom type differences.\n[lj] Only report differences in Lennard-Jones parameters.\n[bnd] Only report differences in bond parameters.\n[ang] Only report differences in angle parameters.\n[dih] Only report differences in dihedral parmeters.\n73"}, {"page": 74, "text": "[atoms] Only report differences in atom properties.\nCompare and report differences in atoms/parameters between two topologies.\nDifferences are reported in standard ’diff’ format, with ’<’ prefix indicating the\nparameter is from the first topology and ’>’ prefix indicating the parameter is\nfrom the second topology.\n9.8 dihedralinfo | dihedrals | printdihedrals\ndihedralinfo [parm <name> | parmindex <#> | <#>] [<mask1>] [<mask2> <mask3> <mask4>]\n[out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n[<mask1>] Mask to print dihedral info for.\n[<mask2> <mask3> <mask4>] If specified, dihedrals\nmust match all masks.\n[out <file>] File to print to (default STDOUT).\nPrint dihedral information of atoms in <mask> for selected topology (first\nloaded topology by default) with format:\n#Dihedral pk phase pn atoms\nwhere#Dihedralistheinternaldihedralindex,pkisthedihedralforceconstant,\nphaseisthedihedralphase, pnisthedihedralperiodicity, andatomsshowsthe\nnames of the atoms involved in the angle with format :<residue num>@<atom\nname>, followed by the atom indices involved in a comma-separated list. In\naddition if the dihedral is an end dihedral, improper dihedral, or both it will be\nprefaced with an E, I, or B respectively. Atom types will be shown in the last\ncolumn.\nIf 4 masks are given instead of 1, print info for dihedrals with first atom in\n<mask1>,secondatomin<mask2>,thirdatomin<mask3>,andfourthatom\nin <mask4>.\n9.9 hmassrepartition\nhmassrepartition [parm <name> | crdset <set> | parmindex <#> | <#>]\n[<mask>] [hmass <hydrogen new mass>] [dowater]\nparm <name> Modify topology selected by name.\ncrdset <set> Modify topology of COORDS set.\nparmindex <#> | <#> Modify topology selected by\nindex <#> (starting from 0).\n74"}, {"page": 75, "text": "<mask> Atoms to modify (all solute atoms by default).\nhmass <hydrogen new mass> Mass to change hydrogens\nto (3.024 u by default).\ndowater If specified, modify water hydrogen mass as\nwell.\nPerformhydrogenmassrepartitioningonthespecifiedtopology. Hydrogenmass\nrepartitioning means that for a given heavy atom, the mass of all bonded hy-\ndrogens are increased (to 3.024 u by default) and the mass of that heavy atom\nis decreased so as to maintain the same overall mass. The main use case is\nto allow longer time steps for molecular dynamics integration due to reduced\nfrequency of vibration of bonds to hydrogen atoms.\n9.10 improperinfo | impropers | printimpropers\nimproperinfo [parm <name> | parmindex <#> | <#>] [<mask1>] [<mask2> <mask3> <mask4>]\n[out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n[<mask1>] Mask to print improper info for.\n[<mask2> <mask3> <mask4>] If specified, impropers\nmust match all masks.\n[out <file>] File to print to (default STDOUT).\nForspecifiedtopology(firstbydefault)eitherprintCHARMMimproperinfofor\nall atoms in <mask1>, or print info for dihedrals with first atom in <mask1>,\nsecondatomin<mask2>,thirdatomin<mask3>,andfourthatomin<mask4>.\n9.11 mass\n[<parmindex>] [parm <name> | parmindex <#> | <#>] <mask> [out <file>] [name <set>]\nparm <name> | parmindex <#> Topology to calculate\nmass from.\n<mask> Atom(s) to calculate total mass for (default\nall).\n[out <file>] File to write total mass to.\n[name <set>] If specified, a data set named <set> will\nbe cretaed containing total mass.\nPrint the total mass of atoms in <mask> (in amu) for selected topology (first\nloaded topology by default).\n75"}, {"page": 76, "text": "9.12 molinfo\nmolinfo [parm <name> | parmindex <#> | <#>] <mask> [out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n<mask> Mask selecting molecules to print info for.\n[out <file>] File to print to (default STDOUT).\nPrint molecule information for atoms in <mask> for selected topology (first\nloaded topology by default) with format:\n#Mol Natom #Res Name C [SOLVENT]\nwhere #Mol is the molecule number, Natom is the number of atoms in the\nmolecule, #Res and Name are the residue number and residue name of the first\nresidue in the molecule respectively, and C is the chain ID of the first residue.\nIf the molecule is composed on non-consecutive fragments, #Res, Name, and C\nwill be printed for each fragment. SOLVENT will be printed if the molecule is\ncurrently considered a solvent molecule.\n9.13 parm\nparm <filename> [{[TAG] | name <setname>}]\n[{ nobondsearch |\n[bondsearch <offset>] [searchtype {grid|pairlist}]\n}] [nomolsearch] [renumresidues]\n<filename> Parameter file to read in; format is\nauto-detected.\n’[TAG]’ Optional tag (bounded in brackets) which can be\nreferred to in place of the topology file name in\norder to simplify references to it (see 3.4 on\npage 18 for examples of how to use tags).\n[name <setname>] Optional name that can be used to\nrefer to the topology in place of the file name.\n[bondsearch <offset>] Optional; when searching for\nbonds via geometry search (default for Topologies\nwithout bond information) add <offset> to distances\n(default 0.2 Å). Increase this if your system\nincludes unusually long bonds.\n[searchtype {grid|pairlist}] Change search algorithm from\nthe default search between residues algorithm:\n76"}, {"page": 77, "text": "grid Uses a grid when searching for bonds between\nresidues. This can find bonds between residues\nthat are not sequential (e.g. disulfide bonds).\npairlist Uses a pair list to search for bonds between\natoms. This can potentially find bonds across\nperiodic boundaries, but is the more\nexperimental of the two.\nAdvanced Options - Not recommended for general use\n[nobondsearch] If specified do not search for bonds via\ngeometry if Topology does not include bond\ninformation. May cause some Actions to fail.\n[nomolsearch] If specified do not search for molecule\ninformation. May cause some Actions to fail.\n[renumresidues] If specified, ensure that any residue\ncannot be part of more than 1 molecule (can occur\nwith e.g. alternate sites). Residues will be\nrenumbered according to molecule information in that\ncase.\nRead in parameter file. The file format will be auto-detected. Current formats\nrecognized by cpptraj are listed on page 68. If the file does not contain bond\ninformation, cpptraj will attempt to assign bonds based on a simple distance\nsearch of atoms within and between residues. The distance cutoff for determin-\ningbondsbetweenatomsdependsontheelementsofthetwoatomsinquestion,\naugmented by <offset>. Molecule information is then determined from bond\ninformation.\n9.13.1 PDB format:\n[pqr] [readbox] [conect] [noconect] [link] [nolink] [keepaltloc <char>]\n[pqr] Read charge and radius information from the\noccupancy and B-factor columns.\n[readbox] Read unit cell information from CRYST1 record\nif present.\n[conect] Read CONECT records if present (default).\n[noconect] Do not read in CONECT records from PDB file.\n[link] Read LINK records if present.\n[nolink] Do not read LINK records if present (default).\n[keepaltloc <char>] If specified, only keep alternate\natom location IDs matching the specified character\n<char>.\n77"}, {"page": 78, "text": "IMPORTANT NOTES FOR PDB FILES SometimesPDBfilescancon-\ntain alternate coordinates for the same atom in a residue, e.g.:\nATOM 806 CA ACYS A 105 6.460 -34.012 -21.801 0.49 32.23\nATOM 807 CB ACYS A 105 6.054 -33.502 -20.415 0.49 35.28\nATOM 808 CA BCYS A 105 6.468 -34.015 -21.815 0.51 32.42\nATOM 809 CB BCYS A 105 6.025 -33.499 -20.452 0.51 35.38\nIfthisisthecasecpptraj willprintawarningaboutalternatelocationIDsbeing\npresent but will take no other action. Both residues are considered ’CYS’ and\nthemask’:CYS@CA’wouldselectbothatom806and808. Ifdesired, aspecific\nlocationIDcanbekeptviathekeepaltlockeyword. If keepaltlocisspecified,\nit should also be specified for any trajin commands (see 10.4.3 on page 90).\nResidue insertion codes are read in but also not used by the mask parser.\n9.13.2 Charmm PSF:\n[param <file>]\n[param <file>] Read CHARMM parameters from given file.\nCan do multiple times.\n9.13.3 Gromacs Top\nBydefaultcpptrajwilllookforGromacstopologydata(thatisnotinthesame\ndirectory) in the directory defined by the GMXDATA environment variable;\nspecifically, it expects things to be in the \"$GMXDATA/top\" directory.\n9.14 parmbox\nparmbox [parm <name> | parmindex <#> | <#>] [nobox] [truncoct]\n[x <xval>] [y <yval>] [z <zval>] [alpha <a>] [beta <b>] [gamma <g>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology to modify. Default is first\nloaded topology.\n[nobox] Remove box information.\n[truncoct] Set truncated octahedon angles with lengths\nequal to <xval>.\n[x <xval>] Box X length.\n[y <yval>] Box Y length.\n[z <zval>] Box Z length.\n[alpha <a>] Box alpha angle.\n[beta <b>] Box beta angle.\n78"}, {"page": 79, "text": "[gamma <g>] Box gamma angle.\nModify the box information for specified topology. Overwrites any box infor-\nmation if present with specified values; any that are not specified will remain\nunchanged. Note that unlike the box action this command affect box informa-\ntion immediately. This can be useful for e.g. removing box information from a\nparm when stripping solvent:\nparm mol.water.parm7\nparmstrip :WAT\nparmbox nobox\nparmwrite out strip.mol.nobox.parm7\n9.15 parminfo\nparminfo [parm <name> | parmindex <#> | <#>] [<mask>]\nPrintasummaryofinformationcontainedinthespecifiedtopology(firstloaded\ntopology by default) .\n9.16 parmstrip\nparmstrip <mask> [parm <name> | parmindex <#> | <#>]\nStrip atoms in <mask> from specified topology (by default the first topology\nloaded). Note that unlike the strip Action, this permanently modifies the\ntopology for as long as cpptraj is running, so this should not be used if the\ntopology is being used to read or write a trajectory via trajin/trajout. This\ncommand can be used to quickly created stripped Amber topology files. For\nexample, to strip all residues name WAT from a topology and write a new\ntopology:\nparm mol.water.parm7\nparmstrip :WAT\nparmwrite out strip.mol.parm7\n9.17 parmwrite\nparmwrite out <filename> [{parm <name> | parmindex <#> | <#> | crdset <setname>}]\n[<fmt>] [nochamber]\n<filename> File to write to.\n[parm <name> | parmindex <#> | <#>] Topology to\nwrite out.\n[crdset <setname>] Write topology from specified COORDS\ndata set.\n79"}, {"page": 80, "text": "[<fmt>] Format keyword. If not specified the file name\nextension will be used. Default is Amber Topology.\n[nochamber] (Amber topology only) Remove any CHAMBER\ninformation from the topology.\nWrite out specified topology (first topology loaded by default) to <filename>\nwith format <fmt> (Amber topology if not specified). Note that the Amber\ntopology format is the only fully supported format for topology writes.\n9.17.1 Amber Topology\n[nochamber] [writeempty] [nopdbinfo]\n[nochamber] Do not write CHAMBER information to\ntopology (useful for e.g. using topology for\nvisualization with older versions of VMD).\n[writeempty] Write Amber tree, join, and rotate info\neven if not present.\n[nopdbinfo] Do not write \"PDB\" info (e.g. chain IDs,\noriginal res #s, etc).\n9.17.2 Charmm PSF\n[oldpsf] [ext]\n[oldpsf] Write atom type indices instead of type names\n(not recommended).\n[ext] Use extended format.\n9.18 printub | ubinfo\nprintub [parm <name> | parmindex <#> | <#>] [<mask1>] [<mask2>] [out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n[<mask1>] Atoms to print UB info for.\n[<mask2>] If specified, UB info must match both masks.\n[out <file>] File to print to (default STDOUT).\nFor specified topology (first by default) either print CHARMM Urey-Bradley\ninfo for all atoms in <mask1>, or print info for bonds with first atom in\n<mask1> and second atom in <mask2>.\n80"}, {"page": 81, "text": "9.19 resinfo\nresinfo [parm <name> | parmindex <#> | <#>] <mask>\n[short {nucleic|[maxwidth <#res>] [groupsize <#res>]}]\n[out <file>]\n[parm <name> | parmindex <#> | <#>] Name/tag or\nindex of topology. Default is first loaded\ntopology.\n<mask> Mask selecting residues to print info for.\n[short] Use a short 1 character residue name format\n[maxwidth <#res>] Max # of residues to print in\none line (default 50).\n[groupsize <#res>] Max # of residues to put in a\ngroup (default 10).\n[nucleic] Do a printout suitable for nucleic acid\nresidues (maxwidth 45, groupsize 3).\n[out <file>] File to print to (default STDOUT).\nPrint residue information for atoms in <mask> for selected topology (first\nloaded topology by default) with format:\n#Res Name First Last Natom #Orig #Mol C\nwhere #Res is the residue number, Name is the residue name, First and Last\nare the first and last atom numbers of the residue, Natom is the total number of\natomsintheresidue,#Origistheoriginalresiduenumber(inPDBfiles),#Mol\nis the molecule number, and C is the chain ID.\nIf shortisspecifiedthenresidueswillbeprintedoutinacondensedformat.\nEach residue name will be shortened to 1 character, and residues are printed\nout in groups of 10, 5 groups to a line, with each line beginning with a residue\nnumber, e.g.\n> resinfo short 4\n1 MGFLAGKKIL ITGLLSNKSI AYGIAKAMHR EGAELAFTYV GQFKDRVEKL\n51 CAEFNPAAVL PCDVISDQEI KDLFVELGKV WDGLDAIVHS IAFAPRDQLE\nIf the 1 character name for a residue is unknown it will be shown as the first\nletter of the residue name in lower-case.\n9.20 scaledihedralk\nscaledihedralk [parm <name> | parmindex <#>] <scale factor> [<mask> [useall]]\nScale dihedral force constants for dihderals selected by <mask> for specified\ntopology. If useall is specified all atoms in <mask> must be present to select\na dihedral, otherwise any atom in <mask> will selected a dihedral.\n81"}, {"page": 82, "text": "9.21 solvent\nsolvent [parm <name> | parmindex <#> | <#>] { <mask> | none }\nSet solvent for selected topology (first loaded topology by default) based on\n<mask>, or set nothing as solvent if none is specified.\n9.22 updateparameters\nparm <name> | parmindex <#> setname <parm set>\nparm <name> | parmindex <#> Topology to update.\nsetname <parm set> Topology or parameter data set\ncontaining parameters to use.\nNOTE: This command is provided for convenience only. For editing topology\nfiles, ParmEd is a much better alternative.\nUpdateparametersinspecifiedtopologywiththosefrom<parmset>. <parm\nset> can either be a parameter set or a topology. If a parameter from <parm\nset> does not exist in the topology it will be added.\nFor example, to modify parameters in a topology file named lys.parm7 with\nthose from parameter file kcx.str:\n# Read Topology to modify\nparm lys.parm7\n# Read CHARMM parameters\nreaddata kcx.str as charmmrtfprm name MyParm\n# Update parameters in Topology with those from kcx.str\nupdateparameters parmindex 0 setname MyParm\n# Write out the updated Topology\nparmwrite out lys.kcx.parm7\n10 Trajectory File Commands\nThese commands control the reading and writing of trajectory files. There are\nthree trajectory types in cpptraj: input, output, and reference. In cpptraj,\ntrajectories are always associated with a topology file. If a topology file is not\nspecified, a trajectory file will be associated with the first topology file loaded\nby default (this is true for both input and output trajectories.\nCpptraj currently understands the following trajectory file formats:\n82"}, {"page": 83, "text": "Format Keyword(s) Extension Notes\nAmber Trajectory crd .crd Default format if\nkey-\nwords/extensions\nnot recognized.\nAmber NetCDF cdf, netcdf .nc No compression.\nAmber Restart restart .rst7, .rst\nAmber NetCDF ncrestart, restartnc .ncrst\nRestart\nCharmm “DCD” dcd, charmm .dcd\nTrajectory\nCharmm cor .cor\nCOORdinates\nCharmm Restart charmmres .res Read Only\nPDB pdb .pdb\nMol2 mol2 .mol2\nScripps Binpos binpos .binpos\nGromacs TRR trr .trr\nGromacs GRO gro .gro Read Only\nGromacs XTC xtc .xtc\nGromacs TNG tng .tng Read Only\nCIF cif .cif Read Only\nTinker ARC arc .arc Read Only\nSQM Input sqm .sqm Write Only\nSDF sdf .sdf Read Only\nXYZ xyz .xyz\nDesmond DTR dtr .dtr Read Only\n(Anton)\nLMOD Conflib conflib .conflib Read Only,\nDetection by\nextension\nMDTraj H5 h5 .h5 Read Only\nMDAnalysis H5MD h5md .h5md Read Only\nThe following trajectory-related commands are available:\nCommand Description\nensemble Set up a trajectory ensemble for reading during a run.\nensemblesize (MPI only) specify number of members expected in subsequent ensemble commands.\nreference Read in a reference structure.\ntrajin Set up a trajectory for reading during a Run.\ntrajout Set up an output trajectory or ensemble for writing during a Run.\n83"}, {"page": 84, "text": "10.1 ensemble\nensemble <file0> {[<start>] [<stop> | last] [offset]} | lastframe\n[parm <parmfile / tag> | parmindex <#>]\n[trajnames <file1>,<file2>,...,<fileN>\n[{nosort |\nbycrdidx |\nremlog <remlogfile> [nstlim <nstlim> ntwx <ntwx>]}]\n<file0> Lowest replica filename.\n[<start>] Frame to begin reading ensemble at (default\n1).\n[<stop> | last] Frame to stop reading ensemble at; if not\nspecified or ’last’ specified, end of trajectories.\n[<offset>] Offset for reading in trajectory frames\n(default 1).\n[lastframe] Select only the final frame of the\ntrajectories.\n[parm <parmfile>] Topology filename/tag to associate\nwith trajectories (default first topology).\n[parmindex <#>] Index of Topology to associate with\ntrajectories (default 0, first topology).\n[trajnames <file1>,...,<fileN>] Do not automatically\nsearch for additional replica trajectories; use\ncomma-separated list of trajectory names.\n[nosort] Do not attempt to sort trajectories. Useful for\nH-REMD trajectories which are already sorted by\nreplica/Hamiltonian, or collections of MD\ntrajectories.\n[bycrdidx] For H-REMD trajectories, sort by coordinate\nindices stored in trajectory files. This is\npreferred over sorting via ’remlog’.\n[remlog <remlogfile>] For H-REMD trajectories only, use\nspecified REMD log file to sort trajectories by\ncoordinate index (instead of by\nreplica/Hamiltonian).\n[nstlim <nstlim> ntwx <ntwx>] If trajectory and\nREMD log were not written at the same rate,\nthese are the values for nstlim (steps between\neach exchange) and ntwx (steps between\ntrajectory write) used in the REMD simulation.\n84"}, {"page": 85, "text": "Read in and process trajectories as an ensemble. Similar to ’trajin remdtraj’,\nexcept instead of processing one frame at a target temperature, process all\nframes. This means that action and trajout commands apply to the entire\nensemble; note however that not all actions currently function in ’ensemble’\nmode. For example, to read in a replica ensemble, convert it to temperature\ntrajectories, and calculate a distance at each temperature:\nparm ala2.99sb.mbondi2.parm7\nensemble rem.crd.000 trajnames rem.crd.001,rem.crd.002,rem.crd.003\ntrajout temp.crd\ndistance d1 out d1.ensemble.dat @1 @21\nThiswilloutput4temperaturetrajectoriesnamed’temp.crd.X’,whereXranges\nfrom0to3with0correspondingtothelowesttemperature,and’d1.ensemble.dat’\ncontaining 4 columns, each corresponding to a temperature. If run with MPI,\ndata will be written to separate files named ’d1.ensemble.dat.X’, similar to the\noutput trajectories.\nNote that in parallel (i.e. MPI) users should specify the ensemblesize\ncommand prior to ensemble in order to improve set up efficiency.\nH-REMDtrajectorieswhicharetypicallyalreadysortedbyreplica/Hamiltonian\ncan be sorted by coordinate index instead with ’bycrdidx’ (if the trajectory\ncontains coordinate indices) or by REMD log data specified with the ’remlog’\nkeyword. Forexample,tosortbycoordinateindexusingaREMDlogandwrite\nsorted trajectories:\nparm ../tz2.nhe.parm7\nensemblesize 4\nensemble rem.crd.001 remlog rem.log nstlim 1000 ntwx 1000\ntrajout sorted.remlog.nc\nTo sort by coordinate index when trajectories contain coordinate indices:\nparm ../tz2.nhe.parm7\nensemblesize 4\nensemble rem.crd.001 bycrdidx\ntrajout sorted.crdidx.nc\n10.2 ensemblesize\nensemblesize <#>\nThis command is MPI only. It is used to set the expected number of members\nin any subsequent ensemble command, which dramatically improves set up\nefficiency.\n85"}, {"page": 86, "text": "10.3 reference\nreference <name> [<frame#>] [<mask>] ([tag]) [lastframe] [crdset]\n[parm <parmfile / tag> | parmindex <#>]\n<name> File name (or COORDS set name if ’crdset’\nspecified) to read in as reference; any trajectory\nrecognized by ’trajin’ can be used.\n[<frame#>] Frame number to use (default 1).\n[<mask>] Only load atoms corresponding to <mask> from\nreference.\n([tag]) Tag to give this reference file, e.g. “[MyRef]”;\nBRACKETS MUST BE INCLUDED.\n[lastframe] Use last frame of reference.\n[crdset] Use for COORDS data set named <name> instead of\nfile.\n[parm <parmfile/tag>] Topology filename/tag to\nassociate with reference (default first topology).\n[parmindex <#>] Index of Topology to associate with\nreference (default 0, first topology).\nUse specified trajectory as reference coordinates. For trajectories with multiple\nframes, the first frame is used if a specific frame is not specified. An optional\ntag can be given (bounded in brackets) which can then be used in place of the\nname (see 3.4 on page 18 for examples of how to use tags). If desired, an atom\nmask can be used to read in only specified atoms from a reference.\nReference coordinates are now considered COORDS data sets and can be\nused anywhere a COORDS data set could, which allows reference structures to\nbe manipulated once they are loaded. For example, a reference structure could\nbe centered on the origin like so:\nreference tz2.rst7 [MyRef]\ncrdaction [MyRef] center origin\nNote that the ’average’ keyword has been deprecated for reference. If desired,\nanaveragedreferenceCOORDSdatasetcanbecreatedfromatrajectoryusing\nthe ’average’ command like so:\nparm myparm.parm7\ntrajin mytraj.nc\nrms first :1-12\naverage crdset RefAvg\nrun\nrms ToAvg reference :1-12 out ToAvg.dat\n86"}, {"page": 87, "text": "10.4 trajin\ntrajin <filename> {[<start> [<stop> | last] [<offset>]]} | lastframe\n[parm <parmfile / tag> | parmindex <#>]\n[mdvel <velocities>] [mdfrc <forces>]\n[as <format keyword>] [ <Format Options> ]\n[ remdtraj {remdtrajtemp <Temperature> | remdtrajidx <idx1,idx2,...>\n| remdtrajvalues <value1,value2,...>}\n[trajnames <file1>,<file2>,...,<fileN>] ]\n<filename> Trajectory file to read in.\n[<start>] Frame to begin reading at (default 1). If a\nnegative value is given it means “<start> frames\nbefore <stop>”.\n[<stop> | last] Frame to stop reading at; if not\nspecified or ’last’ specified, end of trajectory.\n[<offset>] Offset for reading in trajectory frames\n(default 1).\n[lastframe] Select only the final frame of the\ntrajectory.\n[parm <parmfile/tag>] Topology filename/tag to\nassociate with trajectory (default first topology).\n[parmindex <#>] Index of Topology to associate with\ntrajectory (default 0, first topology).\n[mdvel <velocities>] Use velocities from specified file.\n[mdfrc <forces>] Use forces from specified file.\n[as <format keyword>] Force file to be read as\nspecified format; overrides file autodetection.\n[<Format Options>] See below.\n[remdtraj] Read <filename> as the first replica in a\ngroup of replica trajectories.\nremdtrajtemp <Temperature> | remdtrajidx <idx1,idx2,...>\nUse frames at <Temperature> (for temperature\nreplica trajectories) or index <idx1,idx2,...>\n(for Hamiltonian replica trajectories); For\nMultidimensional REMD simulations, multiple\nvalues are comma-separated.\nremdtrajvalues <value1,value2,...> Use frames at\n<value1,value2,...> (for Multidimensional REMD\ntrajectories). Each value may correspond to\neither temperature, pH, Redox Potential or\nHamiltonian index. The values need to be\n87"}, {"page": 88, "text": "entered in the same order as the dimensions in\nthe Multidimensional REMD simulation. For\nexample, for T,pH-REMD value1 would correspond\nto a temperature and value2 to a pH. In the\ncommand, the values are comma-separated.\n[trajnames <file1>,...,<fileN>] Do not automatically\nsearch for additional replica trajectories; use\ncomma-separated list of trajectory names.\nRead in trajectory specified by filename. See page 83 for currently recognized\ntrajectory file formats. If just the <start> argument is given, all frames from\n<start>tothelastframeofthetrajectorywillberead. Toreadinatrajectory\nwith offsets where the last frame # is not known, specify the last keyword\ninstead of a <stop> argument, e.g.\ntrajin Test1.crd 10 last 2\nThiswillprocessTest1.crdfromframe10tothelastframe,skippingby2frames.\nTo explicitly select only the last frame, specify the lastframe keyword:\ntrajin Test1.crd lastframe\nHere is an example of loading in multiple trajectories which have difference\ntopology files:\nparm top0.parm7\nparm top1.parm7\nparm top2.parm7 [top2]\nparm top3.parm7\ntrajin Test0.crd\ntrajin Test1.crd parm top1.parm7\ntrajin Test2.crd parm [top2]\ntrajin Test3.crd parmindex 3\nTest0.crdisassociatedwithtop0.parm7;sincenoparmwasspecifieditdefaulted\ntothefirstparmreadin. Test1.crdwasassociatedwithtop1.parm7byfilename,\nTest2.crd was associated with top2.parm7 by its tag, and finally Test3.crd was\nassociated with top3.parm7 by its index (based on the order it was read in).\nReplica Trajectory Processing\nIf the remdtraj keyword is specified the trajectory is treated as belonging to\nthe lowest # replica of a group of REMD trajectories. The remaining repli-\ncas can be either automatically detected by following a naming convention of\n<REMDFILENAME>.X,whereXisthereplicanumber, orexplicitlyspecified\nin a comma-separated list following the trajnames keyword. All trajectories\nwill be processed at the same time, but only frames with a temperature match-\ning the one specified by remdtrajtemp or remdtrajidx will be processed.\n88"}, {"page": 89, "text": "For example, to process replica trajectories rem.001, rem.002, rem.003, and\nrem.004, grabbing only the frames at temperature 300.0 (assuming that this is\na temperature in the ensemble):\ntrajin rem.001 remdtraj remdtrajtemp 300\nor\ntrajin rem.001 remdtraj remdtrajtemp 300 trajnames rem.002,rem.003,rem.004\nNote that the remdout keyword is deprecated. For this functionality see the\nensemble keyword.\n10.4.1 OptionsforAmberNetCDF,AmberNCRestart, AmberRestart:\n[usevelascoords] [usefrcascoords]\nusevelascoords Read in velocities in place of\ncoordinates if present.\nusefrcascoords Read in forces in place of coordinates if\npresent.\n10.4.2 Options for CHARMM DCD:\n[{shape | namdcell | charmmcell}]\nshape Force reading of box info as CHARMM shape matrix\n(XX XY YY XZ YZ ZZ).\nnamdcell Force reading of box info as NAMD unit cell (X\ncos(g) Y cos(b) cos(a) Z).\ncharmmcell Force reading of box info as old CHARMM unit\ncell (X Y Z a b g).\nNote that CHARMM trajectories can have unit cell data stored in one of\ntwo ways. Older versions (<22) of charmm store the 3x lengths (X Y Z, in\nAngstroms) and 3x angles (alpha beta gamma, in degrees). Newer versions\n(>=22) store elements of the symmetric shape matrix (XX, XY, YY, XZ, YZ,\nZZ). CPPTRAJ will attempt to automatically detect which type of parame-\nters are present, but this can be overridden with the shape and charmmcell\nkeywords. The namdcell keyword is provided for compatibility with DCD tra-\njectories produced by VMD/NAMD, but note these trajectories only store a\nversion of the shape matrix that is correct if the unit cell is also X-aligned.\n89"}, {"page": 90, "text": "10.4.3 Options for PDB files:\n[keepaltloc <char>]\n[keepaltloc <char>] If specified, only keep alternate\natom location IDs matching the specified character\n<char>.\nNotethatif keepaltlocisspecified,theassociatedtopologyshouldnothaveal-\nternatelocationIDs,i.e. ifthetopologyisfromaPDBthekeepaltlockeyword\nmay need to be used with the parm command (see 9.13.1 on page 77).\n10.5 trajout\ntrajout <filename> [<format>] [append] [nobox] [novelocity]\n[notemperature] [notime] [noforce] [noreplicadim]\n[parm <parmfile> | parmindex <#>] [onlyframes <range>] [title <title>]\n[onlymembers <memberlist>]\n[start <start>] [stop <stop>] [offset <offset>]\n[ <Format Options> ]\n<filename> Trajectory file to write to.\n[<format>] Keyword specifying output format (see Table\non page 83). If not specified format will be\ndetermined from extension, otherwise default to\nAmber trajectory.\n[append] If <filename> exists, frames will be appended\nto <filename>.\n[nobox] Do not write box coordinates to trajectory.\n[novelocity] Do not write velocities to trajectory.\n[notemperature] Do not write temperature to trajectory.\n[notime] Do not write time to trajectory.\n[noreplicadim] Do not write replica dimensions to\ntrajectory.\n[parm <parmfile>] Topology filename/tag to associate\nwith trajectory (default first topology).\n[parmindex <#>] Index of Topology to associate with\ntrajectory (default 0, first topology).\n[onlyframes <range>] Write only the specified input\nframes to <filename>.\n[title <title>] Output trajectory title.\n[onlymembers <memberlist>] Ensemble processing only;\nonly write from specified members (starting from 0).\n90"}, {"page": 91, "text": "[start <start>] Begin output at frame <start> (1 by\ndefault).\n[stop <stop>] End output at frame <stop> (last frame by\ndefault).\n[offset <offset>] Skip <offset> frames between each\noutput (1 by default).\nDuring a run, write frames to trajectory specified by filename in specified file\nformat (Amber trajectory if none specified) after all Action processing has oc-\ncurred. TowriteouttrajectorieswithintheActionqueueseetheouttrajAction\n(11.59 on page 184). See page 83 for currently recognized output trajectory\nformats and their associated keyword(s). Note that now the file type can be\ndetermined from the output extension if not specified by a keyword. Multiple\noutput trajectories of any format can be specified.\nFrames will be written to the output trajectory when the param-\neter file being processed matches the parameter file the output tra-\njectory was set up with. So given the input:\nparm top0.parm7\nparm top1.parm7 [top1]\ntrajin input0.crd\ntrajin input1.crd parm [top1]\ntrajout output.crd parm [top1]\nonly frames read in from input1.crd (which is associated with top1.parm7)\nwill be written to output.crd. The trajectory input0.crd is associated with\ntop0.parm7;sincenooutputtrajectoryisassociatedwithtop0.parm7noframes\nwill be written when processing top0.parm7/input0.crd.\nIf onlyframes is specified, only input frames matching the specified range\nwill be written out. For example, given the input:\ntrajin input.crd 1 10\ntrajout output.crd onlyframes 2,5-7\nonly frames 2, 5, 6, and 7 from input.crd will be written to output.crd.\nCell not X-aligned Warning\nCertain Actions (e.g. align, rms, principal, etc.) can rotate the unit cell\nvectors (i.e. the box) if they are present. Some trajectory formats do not\nsupport writing out box coordinates if the unit cell is not “X-aligned”; in other\nwords, if the unit cell “A” vector is not aligned with the coordinate X-axis and\nthe“B” vectorisnotintheX-Yplane. Ifthisisthecase,thefollowingwarnings\nmay appear:\nWarning: Unit cell is not X-aligned. Box cannot be properly stored as <format>.\nWarning: Set <#>; unit cell is not X-aligned. Box cannot be properly stored as <format>.\n91"}, {"page": 92, "text": "ThismeansthattheframewillbewrittenwiththeX-alignedunitcellinsteadof\nthe actual unit cell. Imaging will not be possible with a trajectory written this\nway. Currently the only trajectory formats that support writing non-X-aligned\ncells are the Gromacs TRR and XTC formats.\nIf unit cell information is no longer needed, it can be removed (via e.g. the\nbox action, the strip action with the ’nobox’ keyword, etc.) to prevent these\nwarnings from triggering.\n10.5.1 Options for pdb format\n[dumpq | parse | vdw] [pdbres] [pdbatom]\n[pdbv3] [teradvance] [terbyres | pdbter | noter]\n[model | multi] [chainid <ID>] [sg <group>]\n[include_ep] [conect] [conectmode <m>] [keepext] [usecol21]\n[bfacdefault <#>] [occdefault <#>]\n[bfacdata <set>] [occdata <set>] [bfacbyres] [occbyres]\n[bfacscale] [occscale] [bfacmax <max>] [occmax <max>]\n[adpdata <set>]\ndumpq PQR format; write charges (in units of e-) and\nGB radii to occupancy and B-factor columns\nrespectively.\nparse PQR format; write charges and PARSE radii to\noccupancy/B-factor columns.\nvdw PQR format; write charges and vdW radii to\noccupancy/B-factor columns.\npdbres Use PDB V3 residue names. Will write a default\nchain ID (’Z’) for each residue if the corresponding\ntopology does not have chain ID information.\npdbatom Use PDB V3 atom names.\npdbv3 Use PDB V3 residue/atom names. Same as\nspecifying ’pdbres’ and ’pdbatom’.\ntopresnum Use topology residue numbers; otherwise use\noriginal residue numbers.\nteradvance Increment record (atom) number for TER\nrecords (not done by default).\nterbyres Print TER cards based on residue sequence\ninstead of molecules.\npdbter Print TER cards according to original PDB TER\n(if available).\nnoter Do not write TER cards.\nmodel (Default) Frames will be written to a single PDB\nfile separated by MODEL/ENDMDL keywords.\n92"}, {"page": 93, "text": "multi Each frame will be written to a separate file with\nthe frame # appended to <filename>.\nchainid <ID> Write PDB file with chain ID <ID>.\nsg <group> Space group for CRYST1 record; only used if\nbox coordinates written.\ninclude_ep Include extra points.\nconect Write CONECT records for all bonds.\nconectmode <m> Write CONECT records for <m>=’all’\n(all bonds), ’het’ (HETATM only), ’none’ (no\nCONECT).\nkeepext Keep filename extension; write\n’<name>.<num>.<ext>’ instead (implies ’multi’).\nusecol21 Use column 21 for 4-letter residue names.\nbfacdefault <#> Default value to use in B-factor\ncolumn (default 0.0).\noccdefault <#> Default value to use in occupancy\ncolumn (default 1.0).\nbfacdata <set> Use data in <set> for B-factor column.\noccdata <set> Use data in <set> for occupancy column.\nbfacbyres If specified assume X values in B-factor data\nset are residue numbers.\noccbyres If specified assume X values in occupancy data\nset are residue numbers.\nbfacscale If specified scale values in B-factor column\nbetween 0 and <bfacmax>.\noccscale If specified scale values in occupancy column\nbetween 0 and <occmax>.\nbfacmax <max> Max value for bfacscale.\noccmax <max> Max value for occscale.\nadpdata <set> Use data in <set> (e.g. from the\natomicfluct command, on page 104) for anisotropic\nB-factors.\n10.5.2 Options for Amber ASCII format:\n[remdtraj] [highprecision] [mdvel|mdfrc]\nremdtraj Write REMD header to trajectory that includes\ntemperature: ’REMD <Replica> <Step> <Total_Steps>\n<Temperature>’. Since cpptraj has no concept of\nreplica number, 0 is printed for <Replica>. <Step>\nand <Total_Steps> are set to the current frame #.\n93"}, {"page": 94, "text": "highprecision (EXPERT USE ONLY) Write with 8.6 precision\ninstead of 8.3. Note that since the width does not\nchange, the precision of large coords may be lower\nthan 6.\nmdvel Write velocities instead of coordinates.\nmdfrc Write forces instead of coordinates.\n10.5.3 Options for Amber NetCDF format:\n[remdtraj] [mdvel] [mdfrc] [mdcrd]\nremdtraj Write replica temperature to trajectory.\nmdvel Write only velocity information in trajectory.\nmdfrc Write only force information in trajectory.\nmdcrd Write coordinates to trajectory (only required\nwith mdvel/mdfrc).\nhdf5 Create file as NetCDF4/HDF5 instead of NetCDF4\n(classic).\ncompress Use compression in NetCDF4/HDF5 file.\nicompress Use lossy compression in NetCDF4/HDF5 file\nvia conversion to integers.[6]\n10.5.4 Options for Amber Restart/NetCDF Restart format:\n[remdtraj] [novelocity] [notime] [time0 <initial time>] [dt <timestep>] [keepext]\nremdtraj Write replica temperature to restart. Note\nthat this will automatically include time in the\nrestart file (see the time0 keyword).\ntime0 <initial time> Time for first frame (default 1.0).\ndt <timestep> Time step between frames (default 1.0).\nTime is calculated as t=(time0+frame)*dt.\nkeepext Keep filename extension; write\n’<name>.<num>.<ext>’ instead.\n10.5.5 Options for CHARMM COORdinates:\n[keepext] [ext] [segid <segid>] [segmask <mask> <segid> ...]\nkeepext Keep filename extension; write\n’<name>.<num>.<ext>’\next Use ’extended’ format (default when > 99999 atoms).\n94"}, {"page": 95, "text": "segid <segid> Use <segid> as segment ID for all atoms.\nsegmask <mask> <segid> Use <segid> as segment ID for\natoms selected by <mask>. Can be specified more\nthan once.\n10.5.6 Options for CHARMM DCD:\n[x64] [ucell] [veltraj] [{shape | namdcell | charmmcell}]\nx64 Use 8 byte block size (default 4 bytes).\nveltraj Write velocity trajectory instead of\ncoordinates.\ndt Set trajectory time step in ps.\nnstep # steps between frames.\nstep0 Initial step.\nshape Force writing box info as CHARMM shape matrix (XX\nXY YY XZ YZ ZZ).\nnamdcell Force writing box info as NAMD unit cell (X\ncos(g) Y cos(b) cos(a) Z).\ncharmmcell Force writing box info as old CHARMM unit\ncell (X Y Z a b g).\nNote that by default CPPTRAJ will try to write the symmetric shape matrix\nif box information is present. If this is not possible, CPPTRAJ will fall back to\nwritingunitcellparameters(lengthsandangles)aslongasthecellisX-aligned.\nFor more information see 10.4.2 on page 89.\n10.5.7 Options for GROMACS TRX/XTC format:\n[dt <time step>]\ndt Time step tp multiply set numbers by (default 1.0).\nIgnored if time already present.\nNote: these formats can write rotated (i.e. non-X-aligned) unit cells.\n10.5.8 Options for mol2 format:\n[single | multi] [sybyltype] [sybylatom <file>] [sybylbond <file>] [keepext]\nsingle (Default) Frames will be written to a single Mol2\nfile separated by MOLECULE keywords.\nmulti Each frame will be written to a separate file with\nthe frame # appended to <filename>.\n95"}, {"page": 96, "text": "sybyltype Convert Amber atom types (if present) to SYBYL\ntypes. Requires $AMBERHOME is set.\nsybylatom File containing Amber to SYBYL atom type\ncorrespondance (optional).\nsybylbond File containing Amber to SYBYL bond type\ncorrespondance (optional).\nkeepext Keep filename extension; write\n’<name>.<num>.<ext>’ instead (implies ’multi’).\n10.5.9 Options for SQM input format:\n[charge <c>]\ncharge <c> Set total integer charge. If not specified\nit will be calculated from atomic charges.\n10.5.10 Options for XYZ format:\n[ftype {namexyz|atomxyz|xyz}] [titletype {none|single|perframe}] [width <#>] [prec <#>]\nftype {atomxyz|xyz} Choose either ’NAME X Y Z’\n(default), ’ATOM X Y Z’, or ’X Y Z’ output format.\n’namexyz’ format is the standard XYZ format, where\neach frame is preceded by the number of atoms and a\ncomment. The comment written by CPPTRAJ will\ninclude the set number and box information (if\npresent).\ntitletype {none|single|perframe} No title, one title\n(default), or title before every frame. Only\napplies if not ’namexyz’.\nwidth <#> Output format width.\nprec <#> Output format precision.\n11 Action Commands\nActions in cpptraj operate on frames read in by the trajin or ensemble com-\nmandsoneatatimeandextractderiveddata,modifythecoordinates/topology\ninsomeway,orboth. MostActionsincpptraj functionexactlythewaytheydo\ninptraj andarebackwards-compatible. SomeActioncommandsincpptraj have\nextra functionality compared to ptraj (such as the per-residue RMSD function\nofthermsd Action,ortheabilitytowriteoutstrippedtopologiesforvisualiza-\ntion in the strip Action), while other Actions produce slightly different output\n(like the hbond/secstruct Actions).\n96"}, {"page": 97, "text": "Unlike some other command types, when an Action command is issued it\nis by default added to the Action queue and is not executed until trajectory\nprocessing is started (e.g. by a run or go command). However, Actions can be\nexecuted immediately on COORDS data sets via the crdaction command (7.3\non page 37).\nWhen a frame is modified by an Action, it is modified for every Action that\nfollowsthemduringtrajectoryprocessing. Forexample,givenasolvatedsystem\nwith water residues named WAT and the following Action commands:\nrmsd R1 first :WAT out water-rmsd.dat\nstrip :WAT\nrmsd R2 first :WAT out water-rmsd-2.dat\nthe first rms command will be valid, but the second rms command will not\nsince all residues named WAT are removed from the state by the strip com-\nmand.\nNote that for commands which can use a reference mask as well as a target\nmask (e.g. rms, drmsd, symmrmsd, etc.) there must be a 1 to 1 corre-\nspondence between the atoms in each mask, i.e. the number of atoms and the\nordering of selected atoms must be the same.\nThefollowingActionsareavailable. IfanActionmaymodifycoordinate/topology\ninformation for subsequent Actions it is denoted with an X in the Mod column.\nCommand Description Mod\naddatom Temporarily add an atom to the system. X\nalign Align structure to a reference. X\nangle Calculate the angle between three points.\nareapermol Calculate area per molecule for molecules in a\nspecified plane.\natomiccorr Calculate average correlation between motions of\nspecified atoms.\natomicfluct, rmsf Calculate root mean square fluctuation of specified\natoms/residues.\natommap Attempt to create a map between atoms in X\nmolecules with different atom ordering.\nautoimage Automatically re-image coordinates. X\naverage Calculate average structure.\navgbox Calculate average unit cell (box), primarily for\nunwrapping NPT trajectories.\nbounds Calculate the min/max coordinates for specified\natoms. Can be used to create grid data sets.\nbox Set or overwrite box information for frames.\ncenter Center specified coordinates to box center or onto X\nreference structure.\n97"}, {"page": 98, "text": "check, Check for bad atomic overlaps or bond lengths.\ncheckoverlap,\ncheckstructure Can be used to skip corrupted frames.\ncheckchirality Report chirality around alpha carbons in amino\nacids (L, D).\nclosest, Retain only the specified number of solvent X\nclosestwaters molecules closest to specified solute.\nclusterdihedral Assign frames into clusters based on binning of\nbackbone dihedral angles in amino acids.\ncontacts Older version of nativecontacts, retained for\nbackwards compatibility.\ncreatecrd Create a COORDS data set from input frames.\ncreatereservoir Create a structure reservoir for use with reservoir\nREMD simulations.\ndensity Calculate density along a coordinate.\ndiffusion Calculate translational diffusion of molecules.\ndihedral Calculate the dihedral angle using four points.\ndihrms, Calculate the RMSD of dihedrals to dihedrals in a\ndihedralrms reference structure.\ndipole Bin dipoles of solvent molecules in 3D grid. Not\nwell tested, may be obsolete.\ndistance Calculate the distance between two points.\ndrms, drmsd Calculate the RMSD of distance pairs within\nselected atoms.\ndssp, secstruct Calculate secondary structure content using the\nDSSP algorithm\nenedecomp Perform per-atom energy decomposition.\nenergy Calculate simple bond, angle, dihedral, and\nnon-bonded energy terms (no PME).\nesander Calculate energies using via SANDER; requires\ncompilation with the SANDER API.\nfilter Filter frames for subsequent Actions using data\nsets and user defined criteria.\nfixatomorder Fix atom ordering so that all atoms in molecules X\nare sequential.\nfiximagedbonds Fix bonds which have been split across periodic\nboundaries by imaging.\ngist Perform grid inhomogenous solvation theory.\ngrid Bin selected atoms on a 3D grid.\nhbond Calculate hydrogen bonds using geometric criteria.\nimage Re-image coordinates. The autoimage command X\ntypically provides better results.\njcoupling Calculate J-coupling values from specified dihedral\nangles.\n98"}, {"page": 99, "text": "keep Keep specified atoms in system. X\nlessplit Split/average frames from LES trajectories.\nlie Calculate linear interaction energy between\nuser-specified ligand and surroundings.\nlipidorder Calculate order parameters for lipids in planar\nmembranes.\nlipidscd Calculate lipid order parameters SCD (|<P2>|) for\nlipid chains. Automatically identifies lipids.\nmakestructure Modify structure by applying dihedral values to X\nspecified residues.\nmask Print the results of selection by specified atom\nmask. Good for distnace-based masks.\nmatrix Calculate a matrix of the specified type from input\ncoordinates.\nmindist/maxdist Calculate the minimum or maximum distance\nbetween pairs of atoms/residues/molecules.\nminimage Calculate minimum non-self imaged distance\nbetween atoms in specified masks.\nmolsurf Calculate Connolly surface area of specified atoms.\nCannot do partial surface areas.\nmultidihedral Calculate multiple dihedral angles of\nspecified/given types.\nmultivector Calculate multiple vectors between specified atoms.\nnastruct Perform nucelic acid structure analysis.\nnativecontacts Calculate native contacts within a region or\nbetween two regions using a given reference.\nCan also be used to get min/max distances\nbetween groups of atoms.\nouttraj Write frames to a trajectory file within a list of\nActions.\npairdist Calculate pair distribution function.\npairwise Calculate pair-wise non-bonded energies.\nprincipal Calculate and optionally align system along X\nprincipal axes.\nprojection Project coordinates along given eigenvectors.\npucker Calculate ring pucker using five or six points.\nradgyr, rog Calculate radius of gyration (and optionally\ntensor) for specified atoms.\nradial, rdf Calculate radial distribution function.\nrandomizeions Swap specified ions with randomly selected solvent X\nmolecules.\nremap Re-map atoms according to a given data set. X\nreplicatecell Replicate unit cell in specified (or all) directions\nfor specfied atoms and write to trajectory.\n99"}, {"page": 100, "text": "rms, rmsd Perform best fit of coordinates to reference and X\ncalculate coordinate RMSD.\nFitting can be disabled.\nrotate Rotate the system around X/Y/Z axes, a specified X\naxis, or via given rotation matrices.\nrunavg, Calculate the running average of coordinates over X\nspecified window size.\nrunningaverage\nscale Scale coordinates in X/Y/Z directions by specified X\nfactors.\nsetvelocity Set velocities for specified atoms using Maxwellian\ndistribution based on given temperature.\nspam SPAM method for estimating relative free energies X\nof waters in hydration shell around proteins.\nstfcdiffusion Alternative translational diffusion calculation\nwhich can calculate diffusion in specified regions.\nstrip Remove specified atoms from the system. X\nsurf Calculate the LCPO surface area of specified\natoms. Can do partial surface areas.\nsymmrmsd Calculate symmetry-corrected RMSD. X\ntemperature Calculate system temperature using velocities of\nspecified atoms.\ntime Add/remove/modify time information in frames. X\ntordiff Calculate diffusion using the\ntoroidal-view-preserving scheme.\ntrans, translate Translate specified atoms by specified amounts in X\nX/Y/Z directions.\nunstrip Undo all previous strip Action commands.\nunwrap Reverse of image; unwrap selected atoms so they X\nhave continuous trajectories.\nvector Calculate various types of vector quantities.\nvelocityautocorr Calculate velocity autocorrelation function.\nvolmap Create volumetric map for specified coordinates;\nsimilar to grid but takes\ninto account atomic radii. Similar to VMD\nvolmap.\nvolume Calculate unit cell volume.\nwatershell Calculate the number of waters in the first and\nsecond solvation shells based on distance critera.\nxtalsymm Re-image coordinates based on crystal space group X\nsymmetry operations and asymmetric unit volume.\n100"}, {"page": 101, "text": "11.1 addatom\naddatom aname <name> [elt <element>] [rname <res name>]\n[xyz <X> <Y> <Z>] [mass <mass>] [charge <charge>]\n[outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\naname <name> New atom name. Required.\n[elt <element>] One or two character element string\n(e.g. ’H’ is hydrogen etc.). Default is ’H’.\n[rname <res name>] New residue name for the new atom.\nDefault is ’TMP’.\n[xyz <X> <Y> <Z>] Cartesian XYZ coordinates of the\nnew atom. Default is 0 0 0.\n[mass <mass>] Mass of the new atom. Default is based\noff of the element type.\n[charge <charge>] Charge of the new atom. Default is\n0.\n[outprefix <prefix>] Write out modified topology file\nwith name ’<prefix>.<Original Topology Name>’.\n[nobox] Remove any box information from the modified\ntopology.\n[parmout <file>] Write modified topology to file with\nname <file>.\n[parmopts <list>] Options for writing modified topology\nfile.\nTemporarily add an atom to the system. This is mostly useful for adding a\nplaceholder “dummy” atom for subsequent actions to use. For example, this\ncan be used in conjunction with the ’mask’ command in order to get a list of\natoms within a certain distance of a specified point in space. For example, the\nfollowing input gets a list of atoms within 3.0 Angstroms of coordinates 1, 1, 1:\nparm tz2.pdb\ntrajin tz2.pdb 1 1\naddatom aname TEMP rname TMP elt H xyz 1 1 1\nmask :TMP<@3.0&!:TMP out tz2.mask.dat name TZ2\n11.2 align\nalign <mask> [<refmask>] [move <mask>] [mass]\n[ first | reference | ref <name> | refindex <#> | previous |\nreftraj <name> [parm <name> | parmindex <#>] ]\n<mask> Target atoms to fit.\n101"}, {"page": 102, "text": "[<refmask>] Reference atoms to fit (default is target\nmask).\n[move <mask>] Atoms to move when aligning (default is\ntarget mask).\n[mass] Mass-weight the fit.\nReference keywords:\nfirst Use the first trajectory frame processed as\nreference.\nreference Use the first previously read in reference\nstructure (refindex 0).\nref <name> Use previously read in reference structure\nspecified by filename/tag.\nrefindex <#> Use previously read in reference\nstructure specified by <#> (based on order read in).\nprevious Use frame prior to current frame as reference.\nreftraj <name> Use frames from COORDS set <name> or\nread in from trajectory file <name> as references.\nEach frame from <name> is used in turn, so that\nframe 1 is compared to frame 1 from <name>, frame 2\nis compared to frame 2 from <name> and so on. If\n<trajname> runs out of frames before processing is\ncomplete, the last frame of <trajname> continues to\nbe used as the reference.\nparm <parmname> | parmindex <#> If reftraj\nspecifies a trajectory file, associate it with\nspecified topology; if not specified the first\ntopology is used.\nAlign structure using specified <mask> onto reference. If ’move’ is specified,\nonly move atoms in the move mask.\n11.3 angle\nangle [<dataset name>] <mask1> <mask2> <mask3> [out <filename>] [mass]\n[<dataset name>] Output data set name.\n<maskX> Three atom masks selecting atom(s) to\ncalculate angle for.\n[out <filename>] Output file name.\n[mass] Use center of mass of atoms in <maskX> instead of\ngeometric center.\n102"}, {"page": 103, "text": "Calculateangle(indegrees)betweenatomsin<mask1>,<mask2>,and<mask3>.\nForexample, tocalculatetheanglebetweenthefirstthreeatomsinthesystem:\nangle A123 @1 @2 @3 out A123.agr\n11.4 areapermol\nareapermol [<name>] {[<mask1>] [nlayers <#>] | nmols <#>} [out <filename>]\n[{xy | xz | yz}]\n[<name>] Data set name.\n[<mask1>] Atom mask for selecting molecules. If any\natom in a molecule is selected the whole molecule is\nselected.\n[nlayers <#>] Number of layers of molecules. Total\nnumber of molecules used will be # molecules divided\nby # layers.\n[nmols <#>] If <mask1> is not specified, the number of\nmolecules to use when calculating area per molecule.\n[out <filename>] Output file name.\n[{xy|xz|yz}] Cross-section of box to calculate area of.\nDefault is X-Y.\nCalculate area per molecule as Area / # molecules. The area is determined\nfrom the specified cross-section of the box (X-Y by default). Currently the\ncalculation is only guaranteed to work properly with orthorhombic unit cells.\nFor example, to get the area per molecule of residues named “OL” which are\narranged in 2 layers:\nareapermol OL_area :OL nlayers 2 out apm.dat\n11.5 atomiccorr\natomiccorr [<mask>] out <filename> [cut <cutoff>] [min <min spacing>]\n[byatom | byres]\n<mask> Atoms to calculate motion vectors for.\nout <filename> File to write results to.\ncut <cutoff> Only print correlations with absolute\nvalue greater than <cutoff>.\nmin <min spacing> Only calculate correlations for\nmotion vectors spaced <min spacing> apart.\nbyatom Default; calculate atomic motion vectors.\n103"}, {"page": 104, "text": "byres Calculate motion vectors for entire residues\n(selected atoms in residues only).\nCalculate average correlations between the motion of atoms in <mask>. For\neachframe,amotionvectoriscalculatedforeachselectedatomfromitsprevious\nposition to its current position. For each pair of motion vectors Va and Vb, the\naveragecorrelationbetweenthosevectorsiscalculatedastheaverageofthedot\nproduct of those vectors over all N frames.\n(cid:80)\nAvgCorr(a,b)=\nVa(i)·Vb(i)\nN\nThevalueofAvgCorrcanrangefrom1.0(correlated)to0.0(nocorrelation)\nto -1.0 (anti-correlated). For example, to calculate the correlation of motion\nvectors between residues 1 to 13, writing to a Gnuplot-readable formatted file:\natomiccorr :1-13 out acorr.gnu byres\n11.6 atomicfluct | rmsf\natomicfluct [<name>] [out <filename>] [<mask>] [bfactor] [calcadp [adpout <file>]]\n[{byres [pdbres] [resrange <range>] | byatom | bymask}]\n[start <start>] [stop <stop>] [offset <offset>]\n<name> Output data set name.\nout <filename> Write data to file named <filename>\n[<mask>] Calculate fluctuations for atoms in <mask>\n(all if not specified).\n[bfactor] Calculate atomic positional fluctuations\nsquared and weight by 8π2; this is similar but not\n3\nnecessarily equivalent to the calculation of\ncrystallographic B-factors.\n[calcadp [adpout <file>]] Calculate anisotropic\ndisplacement parameters and optionally output them\nto <file>.\nbyres [pdbres] Output the average (mass-weighted)\nfluctuation by residue.\npdbres If specified, the original residue numbering\nwill be used.\nresrange <range> Calculate fluctuations for\nresidues in the given range argument. The\nresidues selected by <mask> must be a subset of\nresidues specified with <range>. Unselected\nresidues will get a fluctuation of 0.0.\n104"}, {"page": 105, "text": "bymask Output the average (mass-weighted) fluctuation\nfor all atoms in <mask>.\nbyatom (default) Output the fluctuation by atom.\n[<start>] Frame to begin calculation at (default 1).\n[<stop>] Frame to end calculation at (default last).\n[<offset>] Frames to skip between calculations (default\n1).\nDataSets created\n<name> Hold atomic fluctuations.\n<name>[ADP] Hold anisotropic displacement parameters\nif ’calcadp’ specified.\nCompute the atomic positional fluctuations (also referred to as root-mean-\nsquare fluctuations, RMSF) for atoms specified in the <mask>. The RMSF\nof a given atom i is calculated as:\n(cid:114)\n(cid:68) (cid:69)\nRMSF = (x −⟨x ⟩)2\ni i i\nwherex denotesatomicpositionsandtheaveragesareoverallinputframes.\nNotethatRMSfittingisnotdoneimplicitly. Ifyouwantfluctuationswithout\nrotationsortranslations(forexampletotheaveragestructure),performanRMS\nfittotheaveragestructure(best)orthefirststructure(seermsd)priortothis\ncalculation. The units are (Å) for RMSF or Å2 × 8π2 if bfactor is specified.\n3\nIf byres or bymask are specified, the mass-weighted average of atomic\nfluctuations of each atom for either each residue or the entire mask will be\ncalculated respectively:\n(cid:80)\n⟨Fluct⟩=\nAto\n(cid:80)\nmFlucti∗Massi\nMassi\nIf calcadp is specified, anisotropic displacement factors for atoms will be\ncalculated and written to the file specified by adpout (or STDOUT if not\nspecified) using PDB ANISOU record format. The displacement factors will be\nsaved to a data set. Note that calcadp automatically implies bfactor.\nWithcpptraj itispossibletoperformcoordinateaveraging,thefittoaverage\ncoordinates, andtheatomicfluctuationcalculationinasingleexecutionlikeso:\nparm myparm.parm7\ntrajin mytrajectory.crd\nrms first\naverage crdset MyAvg\nrun\nrms ref MyAvg\natomicfluct out fluct.agr\n105"}, {"page": 106, "text": "To write the mass-weighted B-factors for the protein backbone atoms C, CA,\nand N, averaged by residue use the command:\natomicfluct out back.agr @C,CA,N byres bfactor\nTowritetheRMSForatomicpositionalfluctuationsofthesameatoms,usethe\ncommand:\natomicfluct out backbone-atoms.agr @C,CA,N\nTo write a PDB of averaged coordinates (after fitting to the first frame) with\nboth B-factors and anisotropic temperature factors:\nparm myparm.parm7\ntrajin mytraj.nc\nrms first\naverage crdset MyAvg\natomicfluct MyFluct calcadp\nrun\ncrdout MyAvg mypdb.pdb adpdata MyFluct[ADP] bfacdata MyFluct\n11.7 atommap\natommap <target> <reference> [mapout <filename>] [maponly]\n[rmsfit [ rmsout <rmsout> ]]\n[changenames] [chiralimpcut <cut>]\n<target> Reference structure whose atoms will be\nremapped.\n<reference> Reference structure that <target> should be\nmapped to.\nmapout <filename> Write atom map to <filename> with\nformat:\nTargetAtomNumber TargetAtomName ReferenceAtomNumber\nReferenceAtomName\nTarget atoms that cannot be mapped to a reference\natom are denoted “–-”.\nmaponly Write atom map but do not reorder atoms.\nrmsfit Any input frames using the same topology as\n<target> will be RMS fit to <reference> using\nwhatever atoms could be mapped.\nrmsout <rmsout> If rmsfit specified, write\nresulting RMSDs to <rmsout>.\nchangenames If specified, change names of mapped atoms\nin <target> to match those in <reference>.\n106"}, {"page": 107, "text": "chiralimpcut <cut> sets the improper dihedral angle\ncutoff for determining mapping via chirality\n(default 10 deg.).\nAttempttomaptheatomsof<target>tothoseof<reference>basedonstruc-\ntural similarity. This is useful e.g. when there are two files containing the same\nstructure but with different atom names or atom ordering. Both <target> and\n<reference> need to have been read in with a previous reference command.\nThe state will then be modified so that any trajectory read in with the same\nparameterfileas<target>willhaveitsatomsmapped(i.e. reordered)tomatch\nthose of <reference>. If the number of atoms that can be mapped in <target>\nare less than those in <reference>, the reference structure specified by <ref-\nerence> will be modified to include only mapped atoms; this is useful if for\nexample the reference structure is protonated with respect to the target. The\nrmsfitkeywordisusefulincaseswheretheatommappingwillnotbecomplete\n(e.g. two ligands with the same scaffold but different substituents). If change-\nnames is specified, in addition to remapping, the target atom names will be\nchanged to match the reference atom names.\nPart of the mapping process involves mapping atoms bonded to chiral cen-\nters, which in turn involves calculating so-called “improper” dihedral angles to\ndeterminetheorientationofthebondedatoms. Bydefaultthecodeassumesthe\norientation of the atoms around the chiral centers are fairly close, hence there\nis a small improper dihedral angle cutoff of 10 degrees. However, this can be\nincreased via the chiralimpcut keyword to handle cases where one structure\nis distorted with respect to the other.\nFor example, say you have the same ligand structure in two files, Ref.mol2\nand Lig.mol2, but the atom ordering in each file is different. To map the atoms\nin Lig.mol2 onto those of Ref.mol2 so that Lig.mol2 has the same ordering as\nRef.mol2:\nparm Lig.mol2\nreference Lig.mol2\nparm Ref.mol2\nreference Ref.mol2 parmindex 1\natommap Lig.mol2 Ref.mol2 mapout atommap.dat\ntrajin Lig.mol2\ntrajout Lig.reordered.mol2 mol2\n11.8 autoimage\nautoimage [{<mask> | anchor <mask> [fixed <mask>] [mobile <mask>]}]\n[origin] [firstatom] [{familiar|triclinic}] [moveanchor]\n[mode {bydist|byvec}]\n[<mask> | anchor <mask>] Atoms to image around; this\nis the region that will be centered. Default is the\nentire first molecule.\n107"}, {"page": 108, "text": "[fixed <mask>] Molecules that should remain ’fixed’ to\nthe anchor region; default is all\nnon-ion/non-solvent molecules.\n[mobile <mask>] Molecules that can be freely imaged;\ndefault is all ion/solvent molecules.\n[origin] Center anchor region at the origin; if not\nspecified, center at box center.\n[firstatom] Image based on molecule first atom; default\nis to image by molecule center of mass.\n[familiar] Image to familiar truncated-octahedral shape;\nthis is on by default if the original cell is\ntruncated octahedron.\n[triclinic] Force general triclinic imaging.\n[moveanchor] When treating “fixed” molecules, the\nanchor point will be set to the previous \"fixed\"\nmolecule; this is only expected to work well when\n\"fixed\" molecules that are sequential are also\ngeometrically close. Most useful in more condensed\nsystems like those containing membranes.\n[mode {bydist|byvec}] How to treat ’fixed’ molecules.\nbydist The default. ’Fixed’ molecules will use the\nimage closest to the \"anchor\" molecule.\nbyvec Fixed molecules will use the image closest to\ntheir orientation with respect to the anchor in\nthe first frame. May work better than ’bydist’\nfor systems containing membranes.\nAutomaticallycenterandimage(bymolecule)atrajectorywithperiodicbound-\naries. For most cases just specifying ’autoimage’ alone is sufficient. The\natoms of the ’anchor’ region (default the entire first molecule) will be cen-\ntered; all’fixed’moleculeswillbeimagedonlyifimagingbringsthemcloserto\nthe ’anchor’ molecule (default for ’fixed’ molecules is all non-solvent non-ion\nmolecules). All other molecules (referred to as ’mobile’) will be imaged freely.\nThe autoimage command works for the majority of systems; however, for\nvery densely packed systems the default anchor (entire first molecule) may not\nbe appropriate. In these cases, it is recommended to choose as the anchor a\nsmall region which should lie near the center of your system. For example, in a\nprotein dimer system one could choose a single residue that is near the center\nof the interface between the two monomers. The ’moveanchor’ and ’mode\nbyvec’optionsmayalsohelpincaseswherethesystemismorecondensed,such\nas those containing membranes.\n108"}, {"page": 109, "text": "11.9 average\naverage {crdset <set name> | <filename>} [<mask>]\n[start <start>] [stop <stop>] [offset <offset>]\n[Trajout Args]\n<filename> If specified, write averaged coordinates to\n<filename> (not compatible with crdset).\ncrdset <set name> If specified, save averaged\ncoordinates to COORDS set <set name> (not compatible\nwith <filename>).\n[<mask>] Average coordinates in <mask> (all atoms if\nnot specified).\n[<start>] Frame to begin calculation at (default 1).\n[<stop>] Frame to end calculation at (default last).\n[<offset>] Frames to skip between calculations (default\n1).\n[Trajout args] Output trajectory format argument(s)\n(default Amber Trajectory).\nCalculate the average of input coordinates and write out to file named <file-\nname>orsavetoCOORDSsetnamed<set name>inanytrajectoryformat\ncpptraj recognizes (Amber Trajectory if not specified). If the number of atoms\nin <mask> are less than the total number of atoms, the topology will be\nstripped to match <mask>.\nNote that since coordinates are being averaged over many frames, resulting\nstructures may appear distorted. For example, if one averages the coordinates\nof a freely rotating methyl group the average position of the hydrogen atoms\nwill be close to the center of rotation. Also note that typically one will want to\nremove global rotational and translation movement prior to this command by\nusing e.g. the rms (11.70 on page 194) command.\nAny arguments that are valid for the trajout command (10.5 on page 90)\ncan be passed to this command in order to control the format of the output\ncoordinates. For example, to write out a PDB file containing the averaged\ncoordinates over all frames:\naverage test.pdb pdb\nTo write out a mol2 file containing only the averaged coordinates of residues 1\nto 10 for frames 1 to 100:\naverage test.mol2 mol2 start 1 stop 100 :1-10\nTocreateanaveragestructureofatomsnamedCAandthenuseitasareference\nfor an rms command in a subsequent run:\n109"}, {"page": 110, "text": "trajin Input.nc\naverage crdset MyAvg @CA\nrun\nrms ref MyAvg @CA out RmsToAvg.dat\nrun\n11.10 avgbox\navgbox [name <setname>] [out <file>]\n[name <setname>] Average unit cell data set name.\n[out <file>] File to write average unit cell data to.\nDataSets created:\n<setname>[avg] Hold average unit cell as 3x3 matrix\ndata.\nCalculate the average unit cell vectors for incoming frames and store them\nin a 3x3 matrix data set, The average unit cell is particularly useful when\nunwrapping trajectories from NPT simulations where the box size fluctuates\n(see the unwrap command, 11.90 on page 212).\nIf writing to a .dat file, the output will look something like:\n#Frame MyBox[avg]\n1 42.433046075 0.000000000 0.000000000 -14.144347550 40.006259905 0.000000000 -14.144347550 -20.003127532 34.646438786\nWhere the first 3 floating point numbers are the average X unit cell vector, the\nnext 3 floating point numbers are the average Y unit cell vector, and the last 3\nfloating point numbers are the average Z unit cell vector.\n11.11 avgcoord\nThis command is deprecated. Use ’vector center’ (optionally with keyword\n’magnitude’) instead.\n11.12 bounds\nbounds [<mask>] [out <filename>]\n[dx <dx> [dy <dy>] [dz <dz>] name <gridname> [offset <bin offset>]]\n[<mask>] Mask of atoms to determine bounds of.\n[out <filename>] File to write bounds to (default\nSTDOUT if not specified).\n[dx <dx> [dy <dy>] [dz <dz>]] Triggers creation of a\ngrid data set from bounds. Spacings of generated\ngrid in the X, Y and Z directions. If only dx is\nspecified <dx> will be used for <dy> and <dz> as\nwell.\n110"}, {"page": 111, "text": "[name <gridname>] Name of generated data sets.\n[offset <bin offset>] Number of bins to add/subtract in\neach direction to generated grid.\nDataSets Generated\n<gridname> The 3D grid (only if ’dx’ etc specified).\n<gridname>[xmin] The minimum x coordinate\nencountered.\n<gridname>[xmax] The maximum x coordinate\nencountered.\n<gridname>[ymin] The minimum y coordinate\nencountered.\n<gridname>[ymax] The maximum y coordinate\nencountered.\n<gridname>[zmin] The minimum z coordinate encountered.\n<gridname>[zmax] The maximum z coordinate\nencountered.\nCalculate the boundaries (i.e. the max/min X/Y/Z coordinates) of atoms in\n<mask> and write to <filename> (STDOUT if not specified). Useful for\ndetermining dimensions for the grid command, and can be used to generate a\ngrid data set that can be used by grid (see 11.39 on page 149).\n11.13 box\nbox {[x <xval>] [y <yval>] [z <zval>] {[alpha <a>] [beta <b>] [gamma <g>] |\n[truncoct] [x <length>] |\nnobox |\nauto [offset <offset>] [radii {vdw|gb|parse|none}] |\ngetbox {ucell|frac|shape} [name <setname>] [out <file>]\n}\n[x <xval>] [y <yval>] [z <zval>] Change box length(s) to\nspecified value(s).\n[alpha <a>] [beta <b>] [gamma <g>] Change box\nangle(s) to specified value(s).\n[truncoct [x <length>] Set box angles (and optionally a\nlength) to truncated octahedron.\n[nobox] Remove any existing box information.\nauto Set an orthogonal bounding box enclosing all atoms\nby the specified radii and an optional offset.\noffset <offset> Offset in Angstroms to add to each\nbox length (both + and -).\n111"}, {"page": 112, "text": "radii {vdw|gb|parse|none} Radii to use for each\natom: van der Waals, generalized Born, PARSE,\nor no radii.\ngetbox Save existing box information to a 3x3 matrix\ndata set.\n{ucell|frac|shape} Specify the kind of box\ninformation to save: ucell saves the unit cell\nvectors, frac saves the fractional unit cell\nvectors, and shape saves the symmetric shape\nmatrix unit cell vectors.\nname <setname> The name of the 3x3 matrix data\nset.\nout <file> File to write the 3x3 matrix data to.\nModifyboxinformationduringtrajectoryprocessing. Notethatthiswillperma-\nnentlymodifytheboxinformationfortopologyfilesduringtrajectoryprocessing\nas well. It is possible to modify any number of the box parameters (e.g. only\nthe Z length can be modified if desired while leaving all other parameters in-\ntact). If no box is present, an orthogonal bounding box enclosing all atoms\ncan be created with the auto keyword. If ’getbox’ is specified, the existing box\ninformation will be saved to a data set.\n11.14 center\ncenter [<mask>] [origin] [mass]\n[ reference | ref <name> | refindex <#> [<refmask>]]\n[<mask>] Center based on atoms in mask; default is all\natoms.\n[origin] Center to origin (0, 0, 0); default is center to\nbox center (X/2, Y/2, Z/2).\n[mass] Use center of mass instead of geometric center.\n[reference | ref <name> | refindex <#> [<refmask]]\nCenter using coordinates in specified reference\nstructure selected by <refmask> (<mask> if not\nspecified.\nMove all atoms so that the center of the atoms in <mask> is centered at the\nspecified location: box center (default), coordinate origin, or reference coordi-\nnates.\nFor example, to move all coordinates so that the center of mass of residue 1\nis at the center of the box:\ncenter :1 mass\n112"}, {"page": 113, "text": "11.15 check | checkoverlap | checkstructure\ncheck [<mask>] [around <mask2>] [reportfile <report>] [noimage]\n[skipbadframes] [offset <offset>] [minoffset <minoffset>]\n[cut <cut>] [nobondcheck] [silent] [plcut <cut>]\n[<mask>] Check structure of atoms in <mask> (all if\nnot specified).\n[around <mask2>] If specified, only check for problems\nbetween atoms in <mask> and atoms in <mask2>.\n[reportfile <report>] Write any problems found to\n<report> (STDOUT if not specified).\n[noimage] Do not image distances.\n[skipbadframes] If errors are encountered for a frame,\nsubsequent actions/trajectory output will be\nskipped.\n[offset <offset>] Report bond lengths greater than the\nequilibrium value plus <offset> (default 1.15 Å).\n[minoffset <minoffset>] Report bond lengths less than\nthe equilibrium value minus <minoffset> (default 0.5\nÅ).\n[cut <cut>] Report atoms closer than <cut> (default 0.8\nÅ).\n[nobondcheck] Check overlaps only.\n[silent] Do not print information for bad frames - useful\nin conjunction with the skipbadframes option.\n[plcut <cut>] Pair list cutoff (default 4.0 Å); only\nmatters if box is present.\nCheck the structure and report problems related to atomic overlap/unusual\nbond length. Problems are reported when any two atoms in <mask> (or be-\ntween <mask> and <mask2> if using ’around’) are closer than <cut>;\natoms that are bonded to each other are ignored (except if using the ’around’\nmask). If bonds are being checked then bond lengths greater than their equi-\nlibriumvalueplus<offset>andlessthantheirequilibriumvalueminus<mi-\nnoffset> are reported as well. If box information is present and not using the\n’around’ mask, a pairlist will be used to speed up the calculation.\nThis command can also be used to skip corrupted frames in a trajectory\nduring processing. For example, if this message is encountered:\nWarning: Frame 10 coords 1 & 2 overlap at origin; may be corrupt.\n113"}, {"page": 114, "text": "One could use check so that e.g. a subsequent distance command is not\nprocessed for bad frames:\ncheck @1,2 skipbadframes silent\ndistance d1 :1 :10\nUsually frame corruption can be detected using only a few atoms, but this\nmay not catch all types of corruption. The more atoms that are used the\nbetter the corruption detection will be, but the slower it will be to process the\ncommand. Typically a good procedure to follow when corruption is suspected\nis to run check using all important atoms (e.g. all solute heavy atoms) with\nthe skipbadframes keyword followed by a trajout command to write all non-\ncorrupt frames, for example:\ntrajin corrupted.crd\ncheck :1-13 skipbadframes silent\ntrajout fixed.corrupted.nc\n11.16 checkchirality\ncheckchirality [<name>] [<mask>] [out <filename>]\n[<name>] Data set name.\n[<mask>] Atoms to check.\n[out <filename>] File to write results to.\nDataSet Aspects:\n[L] Number of frames ’L’ for each residue.\n[D] Number of frames ’D’ for each residue.\nCheck the chirality around the alpha carbon in amino acid residues selected by\n<mask>. Note that cpptraj expects atom names to correspond to the PDB V3\nstandard: N, CA, C, CB. For each residue, the number of frames in which the\namino acid is ’L’ or ’D’ will be recorded. For example, to check the chirality of\nall amino acids in a system and write to a file named chiral.dat with data set\nname DPDP:\ncheckchirality DPDP out chiral.dat\nOutput will have format similar to:\n#Res DPDP[L] DPDP[D]\n2.000 100 0\nSo in this example residue 2 was ’L’ for 100 frames and ’D’ for 0 frames.\n114"}, {"page": 115, "text": "11.17 closest | closestwaters\nclosest <# to keep> <mask> [solventmask <solvent mask>] [noimage]\n[first | oxygen] [center] [closestout <filename> [name <setname>]]\n[outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\n<# to keep> Number of solvent molecules to keep around\n<mask>\n<mask> Mask of atoms to search for closest waters\naround.\n[solventmask <solvent mask>] Optional mask for\nselecting solvent atoms. If not specified, atoms in\nall molecules marked as “solvent” will be used.\n[noimage] Do not perform imaging; only recommended if\ntrajectory has previously been imaged.\n[first | oxygen] Calculate distances between all atoms in\n<mask> and the first atom of solvent only\n(recommended for standard water models as it will\nincrease speed of calculation).\n[center] Search for waters closest to geometric center of\n<mask> instead of each atom in <mask>.\n[closestout <filename>] Write information on the closest\nsolvent molecules to <filename>.\n[outprefix <prefix>] Write corresponding topology to\nfile with name prefix <prefix>.\n[nobox] Remove any box information from the topology.\n[parmout <file>] Write corresponding topology to file\nwith name <file>.\n[parmopts <list>] Comma-separated list of options for\nwriting the topology file.\nDataSet Aspects:\n[Frame] Frame number.\n[Mol] Original solvent molecule number.\n[Dist] Solvent molecule distance in Å.\n[FirstAtm] First atom number of original solvent\nmolecule.\nSimilar to the strip command, but modify coordinate frame and topology by\nkeeping only the specified number of closest solvent molecules to the region\nspecified by the given mask. Solvent molecules can be determined automat-\nically by cpptraj (by default residues named WAT, HOH, or TIP3), can be\n115"}, {"page": 116, "text": "specified prior via the solvent command (9.21 on page 82), or can be selected\nby solventmask.\nThe format of the closestout file is:\nFrame Molecule Distance FirstAtom#\nForexample,toobtainthe10closestwaterstoresidues1-268bydistancetothe\nfirst atom of the waters, write out which waters were closest for each frame to\na file called “closestmols.dat”, and write out the stripped topology with prefix\n“closest” containing only the solute and 10 waters:\nclosest 10 :1-268 first closestout closestmols.dat outprefix closest\nAs of version 17 this command is CUDA-enabled in CUDA versions of CPP-\nTRAJ.\n11.18 cluster\nAlthough the ’cluster’ command can still be specified as an action, it is now\nconsidered an analysis. See 12.5 on page 229.\n11.19 clusterdihedral\nclusterdihedral [phibins <N>] [psibins <M>] [out <outfile>]\n[dihedralfile <dfile> | <mask>]\n[framefile <framefile>] [clusterinfo <infofile>]\n[clustervtime <cvtfile>] [cut <CUT>]\nCluster frames in a trajectory using dihedral angles. To define which dihedral\nangles will be used for clustering either an atom mask or an input file specified\nby the dihedralfile keyword should be used. If dihedral file is used, each line\nin the file should contain a dihedral to be binned with format:\nATOM#1 ATOM#2 ATOM#3 ATOM#4 #BINS\nwhere the ATOM arguments are the atom numbers (starting from 1) defining\nthe dihedral and #BINS is the number of bins to be used (so if #BINS=10\nthe width of each bin will be 36º). If an atom mask is specified, only protein\nbackbone dihedrals (Phi and Psi defined using atom names C-N-CA-C and N-\nCA-C-N)withinthemaskwillbeused,withthebinsizesspecifiedbythephibins\nand psibins keywords (default for each is 10 bins).\nOutputwilleitherbewrittentoSTDOUTorthefilespecifiedbytheoutkey-\nword. First, information about which dihedrals were clustered will be printed.\nThenthenumberofclusterswillbeprinted,followedbydetailedinformationof\neach cluster. The clusters are sorted from most populated to least populated.\nEach cluster line has format\n116"}, {"page": 117, "text": "Cluster CLUSTERNUM CLUSTERPOP [ dihedral1bin, dihedral2bin ... dihedralNbin ]\nfollowed by a list of frame numbers that belong to that cluster. If a cutoff\nis specified by cut, only clusters with population greater than CUT will be\nprinted.\nIf specified by the clustervtime keyword, the number of clusters for each\nframe will be printed to <cvtfile>. If specified by the framefile keyword, a file\ncontaining cluster information for each frame will be written with format\nFrame CLUSTERNUM CLUSTERSIZE DIHEDRALBINID\nwhereDIHEDRALBINIDisanumberthatidentifiestheuniquecombinationof\ndihedral bins this cluster belongs to (specifically it is a 3*number-of-dihedral-\ncharacters long number composed of the individual dihedral bins).\nIfspecifiedbytheclusterinfokeyword,afilecontaininginformationoneach\ndihedral and each cluster will be printed. This file can be read by SANDER\nfor use with REMD with a structure reservoir (-rremd=3). The file, which is\nessentiallyasimplifiedversionofthemainoutputfile,hasthefollowingformat:\n#DIHEDRALS\ndihedral1_atom1 dihedral1_atom2 dihedral1_atom3 dihedral1_atom4\n...\n#CLUSTERS\nCLUSTERNUM1 CLUSTERSIZE1 DIHEDRALBINID1\n...\n11.20 contacts\ncontacts [ first | reference | ref <ref> | refindex <#> ] [byresidue]\n[out <filename>] [time <interval>] [distance <cutoff>] [<mask>]\nNOTE:Usersareencouragedtotrythenativecontacts command(onpage180),\nan update version of this command.\nForeachatomgiveninmask,calculatethenumberofotheratoms(contacts)\nwithin the distance cutoff. The default cutoff is 7.0 A. Only atoms in mask are\npotential interaction partners (e.g., a mask @CA will evaluate only contacts\nbetween CA atoms). The results are dumped to filename if the keyword “out”\nis specified. Thereby, the time between snapshots is taken to be interval. In\nadditiontothenumberofoverallcontacts,thenumberofnativecontactsisalso\ndetermined. Native contacts are those that have been found either in the first\nsnapshotofthetrajectory(ifthekeyword“first” isspecified)orinareference\nstructure (if the keyword “reference” is specified). Finally, if the keyword\n“byresidue” is provided, results are output on a per-residue basis for each\nsnapshot, whereby the number of native contacts is written to filename.native.\n117"}, {"page": 118, "text": "11.21 createcrd\ncreatecrd [<name>] [ parm <name> | parmindex <#> ]\nThis command creates a COORDS data set named <name> using trajectory\nframes that are associated with the specified topology.\nForexample,tosaveframesthathavebeenpreviouslyRMS-fittoareference\nstructure into a COORDS set named MyCrd you would use the input:\nrms reference :1-12@CA\ncreatecrd MyCrd\nstrip :6-8\nNote that here the strip command will have no effect on the coordinates saved\nin MyCrd since it occurs after the createcrd command.\n11.22 createreservoir\ncreatereservoir <filename> ene <energy data set> [bin <cluster bin data set>]\ntemp0 <temp0> iseed <iseed> [velocity]\n[parm <parmfile> | parmindex <#>] [title <title>]\n<filename> File name of the reservoir to create.\nene <energy data set> Data set with energies\ncorresponding to frames.\n[bin <cluster bin data set>] Data set with bin numbers\n(for RREMD=3).\ntemp0 <temp0> Reservoir temperature.\niseed <iseed> Reservoir random number seed.\n[velocity] Include velocities in the reservoir.\n[parm <parmfile> | parmindex <#>] Associated\ntopology.\n[title <title>] Reservoir title.\nCreate structure reservoir for use with reservoir REMD simulations using en-\nergies in <energy data set>, temperature <temp0> and random seed <iseed>\nInclude velocities if [velocity] is specified. If <cluster bin data set> is specified\nfrom e.g. a previous ’clusterdihedral’ command, the reservoir can be used for\nnon-Boltzmann reservoir REMD (rremd==3).\n11.23 density\ndensity [out <filename>] [name <set name>]\n[ <mask1> ... <maskN> [delta <resolution>] [{x|y|z}]\n[{number|mass|charge|electron}] [{bincenter|binedge}]\n[restrict {cylinder|square} cutoff <cut>] ]\n118"}, {"page": 119, "text": "[out <filename>] Output file for histogram(s) (relative\ndistances vs. densities for each mask) or total\ndensity.\n[name <set name>] Output data set name.\n<mask1> ... <maskN> Arbitrary number of masks for\natom selection; a dataset is created and the output\nwill contain entries for each mask.\n[delta <resolution>] Resolution, i.e. determines\nnumber of slices (i.e. histogram bins).\n(default 0.25 Å)\n[{x|y|z}] Coordinate (axis) for density calculation.\n(default z)\n[{number|mass|charge|electron}] Number, mass,\npartial charge (q) or electron (Ne - q) density.\nElectron density will be converted to e-/Å3 by\ndividing the average area spanned by the other\ntwo dimensions. (default number)\n[{bincenter|binedge}] Determine whether histogram\nbin coordinates will be based on bin center\n(default) or bin edges.\n[restrict {cylinder|square}] If ’restrict’ is\nspecified, only calculate the density that is\nwithin a cylinder or square shape from the\nspecified axis as defined by a distance cutoff.\ncutoff <cut> The distance cutoff for\n’restrict’.\nDataSets Created if masks specified:\n<set name>[avg]:<idx> Average density over coordinate\nfor mask number <idx>.\n<set name>[sd]:<idx> Standard deviation of density\nover coordinate for mask number <idx>.\nDataSets Created if no masks specified:\n<set name> Total system density each frame.\nIf no atom masks are specified, calculate the total system density. Otherwise,\ncalculate specified density along the given axis for atoms in specified mask(s).\nDefaults are shown in parentheses above. The format of the file is as follows.\nComments are lines starting with ’#’ or empty lines. All other lines must\ncontain the atom type followed by an integer number for the electron number.\nEntries must be separated by spaces or ’=’. Example input:\ndensity out number_density.dat number delta 0.25 \":POPC@P1\" \":POPC@N\" \\\n119"}, {"page": 120, "text": "\":POPC@C2\" \":POPC\"\ndensity out mass_density.dat mass delta 0.25 \":POPC@P1\" \":POPC@N\" \\\n\":POPC@C2\" \":POPC\"\ndensity out charge_density.dat charge delta 0.25 \":POPC@P1\" \":POPC@N\" \\\n\":POPC@C2\" \":POPC\"\ndensity out electron_density.dat electron delta 0.25 efile Nelec.in \\\n\":POPC@P1\" \":POPC@N\" \":POPC@C2\" \":POPC\" \":TIP3\" \\\n\":POPC | :TIP3\" \"*\"\ndensity out ion_density.dat number delta 0.25 \":SOD\" \":CLA\"\nSee also $AMBERHOME/AmberTools/test/cpptraj/Test_Density.\nItcanbeusefultowriteouttheaverageandstandarddeviationasanXYDY\nset to a Grace data file, e.g.\ndensity :WAT@O out wato.agr xydy\n11.24 diffusion\nNote that although the syntax for diffusion has changed as of version 16, the\nold syntax is still supported.\ndiffusion [{out <filename>|separateout <suffix>}] [time <time per frame>] [noimage]\n[<mask>] [<set name>] [individual] [diffout <filename>] [nocalc]\n[avgucell <avg ucell set>]\n[allowmultipleorigins]\n[out <filename>] Write mean-square displacement (MSD)\ndata set output to file specified by <filename>.\n[separateout <suffix>] Write each MSD data set type to\nfiles with suffix <suffix>; see description below.\n[time <time_per_frame>] Time in-between each\ncoordinate frame in ps; default is 1.0.\n[noimage] If specified do not perform imaging. If this\nis specified coordinates should be unwrapped prior\nto this command.\n[<mask>] Mask of atoms to calculate diffusion for;\ndefault all atoms.\n[<set name>] MSD data set name.\n[individual] Write diffusion for each individual atom as\nwell as average diffusion for atoms in mask.\n[diffout <filename>] Write diffusion contants calculated\nfrom fits of MSD data sets to <filename>.\n[nocalc] Do not calculate diffusion constants.\n120"}, {"page": 121, "text": "[avgucell] Remove periodic box fluctuations from imaged\nNPT trajectories using average unit cell vectors.\n[allowmultipleorigins] (MPI only). For imaged\ntrajectories in parallel, calculate diffusion by\naveraging over multiple time origins. Should be\nused with caution.\nDataSet Aspects:\n[X] MSD(s) in the X direction.\n[Y] MSD(s) in the Y direction.\n[Z] MSD(s) in the Z direction.\n[R] Overall MSD(s).\n[A] Overall displacement(s).\n[D] Diffusion constants (1x10-5 cm2/s).\n[Label] Diffusion constant lablels.\n[Slope] Linear regression slopes.\n[Intercept] Linear refression Y-intercepts.\n[Corr] Linear regression correlation coefficients.\n[aX]:<atomN> (individual only) Atom <N> MSD in the X\ndirection.\n[aY]:<atomN> (individual only) Atom <N> MSD in the Y\ndirection.\n[aZ]:<atomN> (individual only) Atom <N> MSD in the Z\ndirection.\n[aR]:<atomN> (individual only) Atom <N> overall MSD.\n[aA]:<atomN> (individual only) Atom <N> overall\ndisplacement.\nCompute mean-squared displacement (MSD, in Angstroms squared) plots (us-\ning distance traveled from initial position) for the atoms in <mask>. By\ndefault only the diffusion averaged over all atoms in <mask> is calculated; if\nindividual is specified diffusion for individual atoms is calculated as well.\nIn order to correctly calculate diffusion molecules should take continuous\npaths, so imaging of atoms is autoimatically performed. If the trajectory is al-\nready unwrapped (or the unwrap command is used prior to this command) the\nnoimage keyword can be used. To remove the “noise” caused by box fluctua-\ntions in NPT trajectories, the average unit cell vectors describing the average\nbox can be provided with the avgucell keyword; see the avgbox command\n(11.10 on page 110). Alternatively, the trajectory can be unwrapped prior us-\ning the unwrap command, 11.90 on page 212. If the trajectory is unwrapped,\nthe noimage keyword should be specified.\n121"}, {"page": 122, "text": "Note that in parallel, imaging becomes difficult because there is no way to\ncorrectforanywrappingthathasbeendoneonprecedingMPIranks. Therefore\nthiscommandwillnotworkonimagedtrajectoriesinparallelbydefault. There\nare two workarounds. 1) Unwrap the trajectory prior to diffusion with the\nunwrap command, then run the diffusion calculation with the noimage key-\nword (this is the recommended way), or 2) specify the allowmultipleorigins\nkeyword to calculate MSD separately on each MPI rank, then averaging over\nall MSD plots. This means the maximum length of any given MSD plot will be\n<#frames>/<#MPIranks>, andthecalculateddiffusionconstantswillnot\nbe as accurate.\nThefollowingtypesofdisplacementsarecalculated. If separateoutisspec-\nfied the following files will be created:\nx_<suffix> Mean square displacement(s) in the X direction (in Å2/ps).\ny_<suffix> Mean square displacement(s) in the Y direction (in Å2/ps).\nz_<suffix> Mean square displacement(s) in the Z direction (in Å2/ps).\nr_<suffix> Overall mean square displacement(s) (in Å2/ps).\na_<suffix> Total distance traveled (in Å/ps).\nThe diffusion coefficient D can be calculated using the Einstein relation:\nMSD\n2nD = lim\nt→∞ t\nWhere n is the number of dimensions; for overall MSD n = 3, for single\ndimension MSD (e.g. X) n = 1, etc. Unless nocalc is specified, the diffusion\nconstant is calculated automatically from MSD data sets (and written to the\nfile specified by diffout) in the following manner. The slope the plot of MSD\nversus time is obtained via linear regression. To convert from units of Å2/ps to\n1x10-5 cm2/s,theslopeismultipliedby10.0/(2n). Boththecalculateddiffusion\nconstants as well as the results of the fit are reported.\nDue to the fact that diffusion is currently calculated from initial positions\nonly, diffusioncalculatedforsmallnumbersofatomswillbeinherentlystochas-\ntic, so the results are most sensible when averaged over many atoms; for exam-\nple, the diffusion of water should be calculated using all waters in the system.\nIf more averaging is needed, the calcdiffusion Analysis command (12.3 on\npage 225) can be used to calculate diffusion from multiple time origins.\nFor example, to calculate the diffusion of water in a system:\ndiffusion :WAT@O out WAT_O.agr WAT_O diffout DC.dat\n11.25 dihedral\ndihedral [<name>] <mask1> <mask2> <mask3> <mask4> [out <filename>] [mass]\n[type {alpha|beta|gamma|delta|epsilon|zeta|chi|c2p|h1p|phi|psi|omega|pchi}]\n[range360]\n122"}, {"page": 123, "text": "[<name>] Output data set name.\n<maskX> Four atom masks selecting atom(s) to\ncalculate dihedral for.\n[out <filename>] Output file name.\n[mass] Use center of mass of atoms in <maskX>; default\nis geometric center.\n[range360] Output dihedral angle values from 0 to 360\ndegrees instead of -180 to 180 degrees.\n[type <type>] Label dihedral as <type> for use with\nstatistics analysis; note ’chi’ is nucleic acid chi\nand ’pchi’ is protein chi.\nCalculate dihedral angle (in degrees) between the planes defined by atoms in\n<mask1>, <mask2>, <mask3> and <mask2>, <mask3>, <mask4>. To cal-\nculate multiple dihedral angles see the multidihedral command on page 171.\n11.26 dihedralrms | dihrms\ndihedralrms [<name>] <dihedral types> [out <file>]\n[ first | reference | ref <name> | refindex <#> | previous |\nreftraj <name> [parm <name> | parmindex <#>] ]\n[dihtype <name>:<a0>:<a1>:<a2>:<a3>[:<offset>] ...]\n[tgtrange <range> [refrange <range>]]\n[<name>] Output data set name.\n<dihedral types> Dihedral types to look for. Note that\nchip is ’protein chi’, chin is ’nucleic chi’.\n[out <filename>] Output file name.\n[dihtype <name>:<a0>:<a1>:<a2>:<a3>[:<offset>]\nSearch for a custom dihedral type called <name>\nusing atom names <a0>, <a1>, <a2>, and <a3>.\nOffset: -2=<a0><a1> in previous res, -1=<a0> in\nprevious res, 0=All <aX> in single res, 1=<a3> in\nnext res, 2=<a2><a3> in next res.\n[tgtrange <range>] Residue range to look for target\ndihedrals in. Default is all solute residues.\n[refrange <range>] Residues range to look for reference\ndihedrals in. If not specified, use target range.\nCalculate RMSD of selected dihedrals to dihedrals in a reference structure.\nSee the multidihedral command syntax on page 171 for a list of all available\ndihedral types.\n123"}, {"page": 124, "text": "11.27 dihedralscan\nThis command has been replaced by permutedihedrals; see 7.12 on page 42.\n11.28 dipole\ndipole [out <filename>]\n{ data <dsname> | boxref <ref name/tag> <nx> <ny> <nz> |\n<nx> <dx> <ny> <dy> <nz> <dz>\n[ { gridcenter <cx> <cy> <cz> |\nboxcenter |\nmaskcenter <mask> |\nrmsfit <mask> [noxalign]} ]\n[box|origin|center <mask>] [negative] [name <gridname>]\n<mask1> [max <max_percent>]\n[out <filename>] File to write out grid to. Use\n“.grid” or “.xplor” extension for XPLOR format,\n“.dx” for OpenDX format.\nOptions for setting up grid:\ndata <dsname> Use previously calculated/loaded grid\ndata set named <dsname>. When using this option\nthere is no need to specify grid\nbins/spacing/center.\nboxref <ref name/tag> <nx> <ny> <nz> Set up grid\nusing box information from a previously loaded\nreference structure. Currently the only way to set\nup non-orthogonal grids.\n<nx> <dx> <ny> <dy> <nz> <dz> Number of grid\nbins and spacing in the X/Y/Z directions.\n[gridcenter <cx> <cy> <cz>] Location of grid center,\ndefault is origin (0.0, 0.0, 0.0).\n[boxcenter] Center grid on box center.\n[maskcenter <mask>] Center the grid on the atoms\nselected by <mask>.\n[rmsfit <mask>] Perform a best-fit rotation of the grid\nusing the coordinates selected by <mask>.\n[noxalign] If specified, grid will not be\nre-oriented to align with Cartesian axes once\nbinning is finished. Will affect file formats\nthat do not store full unit cell vectors (like\nXplor).\nOptions for offset during grid binning (must center grid\nat origin):\n124"}, {"page": 125, "text": "[box] Offset each point by location of box center prior\nto gridding. Cannot be used with ’gridcenter’.\n[origin] No offset (default)\n[center <mask>] Offset each point by center of atoms in\n<mask> prior to gridding. Cannot be used with\n’gridcenter’.\nOther options:\n[negative] Grid negative density instead of positive\ndensity.\n[name <gridname>] Grid data set name.\n<mask1> Mask selecting solvent atoms to bin.\n[max <max percent>] Only keep density >= to\n<max_percent> of the maximum density.\nNOTE: This command is not well-tested and may be obsolete.\nSimilar to grid (see 11.39 on page 149 below) except that dipoles of the\nsolvent molecules are binned. The output file format is for Chris Bayly’s dis-\ncern delegate program that comes with Midas/Plus. Consult the code in Ac-\ntion_Dipole.cpp for more information.\n11.29 distance\ndistance [<name>] <mask1> [<mask2>] [point <X> <Y> <Z>]\n[ reference | ref <name> | refindex <#> ]\n[out <filename>] [geom] [noimage] [type noe]\nOptions for ’type noe’:\n[bound <lower> bound <upper>] [rexp <expected>] [noe_strong] [noe_medium] [noe_weak]\n[<name>] Output data set name\n<mask1> Atom mask selecting atom(s) to calculate\ndistance between.\n<mask2> If specified, second atom mask selection\natom(s) to calculate distance from <mask1>.\npoint <X> <Y> <Z> If specified instead of second\nmask, calculate distance between <mask1> and\nspecified XYZ coordinates.\nreference | ref <name> | refindex <#> If specified,\ncalculate distance between <mask1> in each input\nframe and <mask2> in the specified reference.\n[out <filename>] Output filename.\n[geom] Use geometric center of atoms in <mask1>/<mask2>;\ndefault is to use center of mass.\n125"}, {"page": 126, "text": "[noimage] Do not image distances across periodic\nboundaries.\n[type noe] Mark distance as ’noe’ for use with\nstatistics analysis.\n[bound <lower> bound <upper>] Lower and upper\nbounds for NOE (in Angstroms); must specify\nboth.\n[rexp <expected>] Expected value for NOE (in\nAngstroms); if not given ’(<lower> + <upper>)’ /\n2.0 is used.\n[noe_strong] Set lower and upper bounds to 1.8 and\n2.9 Å respectively.\n[noe_medium] Set lower and upper bounds to 2.9 and\n3.5 Å respectively.\n[noe_weak] Set lower and upper bounds to 3.5 and\n5.0 Å respectively.\nCalculate distance between the center of mass of atoms in <mask1> to atoms\nin <mask2>, between atoms in <mask1> from each input frame and atoms in\n<mask2> in specified reference, or atoms in <mask1> and the specified point.\nIf geom is specified use the geometric center instead. For periodic systems\nimaging is turned on by default; the noimage keyword disables imaging.\nA distance can be labeled using ’type noe’ for further analysis as an NOE\nusing the ’statistics’ analysis command (12.37 on page 276).\n11.30 drms | drmsd (distance RMSD)\ndrmsd [<dataset name>] [<mask> [<refmask>]] [out <filename>]\n[ first | ref <refname> | refindex <#> |\nreftraj <trajname> [parm <trajparm> | parmindex <parm#>] ]\n[<dataset name>] Output data set name.\n[<mask>] Atoms to calculate DRMSD for.\n[<refmask>] Mask corresponding to atoms in reference;\nif not specified, <mask> is used.\n[out <filename>] Output file name.\n[first] Use the first trajectory frame processed as\nreference.\n[reference] Use the first previously read in reference\nstructure.\n[ref <refname>] Use previously read in reference\nstructure specified by <refname>.\n126"}, {"page": 127, "text": "[refindex <#>] Use previously read in reference\nstructure specified by <#> (based on order read in).\nprevious Use frame prior to current frame as reference.\nreftraj <name> Use frames from COORDS set <name> or\nread in from trajectory file <name> as references.\nEach frame from <name> is used in turn, so that\nframe 1 is compared to frame 1 from <name>, frame 2\nis compared to frame 2 from <name> and so on. If\n<trajname> runs out of frames before processing is\ncomplete, the last frame of <trajname> continues to\nbe used as the reference.\nparm <parmname> | parmindex <#> If reftraj\nspecifies a file associate trajectory <name>\nwith specified topology; if not specified the\nfirst topology is used.\nCalculate the distance RMSD (i.e. the RMSD of all pairs of internal distances)\nbetweenatomsintheframedefinedby <mask>(allifno<mask>specified)\nto atoms in a reference defined by <refmask> (<mask> if <refmask> not\nspecified). Both <mask> and <refmask> must specify the same number of\natoms, otherwise an error will occur.\nBecause this method compares pairs of internal distances and not absolute\ncoordinates, it is not sensitive to translations and rotations the way that a no-\nfit RMSD calculation is. It can be more time consuming however, as (N2-N)/2\ndistances must be calculated and compared for both the target and reference\nstructures.\nFor example, to get the DRMSD of a residue named LIG to its structure in\nthe first frame read in:\ndrmsd :LIG first out drmsd.dat\n11.31 dssp\nSee 11.78 on page 199.\n11.32 enedecomp\nenedecomp [<name>] [<mask>] [out <filename>] [savecomponents]\n[ pme [cut <cutoff>] [dsumtol <dtol>] [ewcoeff <coeff>]\n[erfcdx <dx>] [skinnb <skinnb>] [ljswidth <width>]\n[order <order>] [nfft <nfft1>,<nfft2>,<nfft3>]\n[<name>] Data set name.\n[<mask>] Mask of atoms to calculate energy for.\n[out <filename>] File to write results to.\n127"}, {"page": 128, "text": "[savecomponents] If specified, also save the individual\ncomponents of the total energy (bond, angle,\ndihedral, etc.).\n[pme] Use particle mesh Ewald for electrostatics; van\nder Waals energy will be calculated using a\nlong-range correction for periodicity.\ncut <cutoff> Direct space cutoff in Angstroms\n(default 8.0).\ndsumtol <dtol> Direct sum tolerance (default\n0.00001). Used to determine Ewald coefficient.\newcoeff <coeff> Ewald coefficient in 1/Ang.\nerfcdx <dx> Spacing to use for the ERFC splines\n(default 0.0002 Ang.).\nskinnb Used to determine pairlist atoms (added to\ncut, so pairlist cutoff is cut + skinnb);\nincluded in order to maintain consistency with\nresults from sander.\nljswidth <width> If specified, use a\nforce-switching form for the Lennard-Jones\ncalculation from <cutoff>-<width> to <cutoff>.\norder <order> Spline order for charges.\nnfft <nfft1>,<nfft2>,<nfft3> Explicitly set the\nnumber of FFT grid points in each dimension.\nWill be determined automatically from unit cell\ndimensions if not specified.\nDataSets created:\n<name> Set containing atom index and the\ncorresponding average total energy over frames.\nIf ’savecomponents’ is specified:\n<name>[bond] Set containing atom index and average\nbond energy over frames.\n<name>[angle] Set containing atom index and average\nangle energy over frames.\n<name>[dih] Set containing atom index and average\ndihedral energy over frames.\n<name>[vdw14] Set containing atom index and average\n1-4 van der Waals energy over frames.\n<name>[elec14] Set containing atom index and average\n1-4 electrostatic energy over frames.\n<name>[elec] Set containing atom index and average\nelectrostatic energy over frames.\n128"}, {"page": 129, "text": "<name>[vdw] Set containing atom index and average van\nder Waals energy over frames.\nPerform per-atom energy decomposition for selected atoms. The energy is cal-\nculated for the entire system but only the energies for selected atoms will be\nreported. The energy is composed of the regular bond, angle, torsion, 1-4 non-\nbonded, and nonbonded terms. If ’savecomponents’ is specified, each of the\nenergy components will be saved in addition to the total energy for each atom.\nIf ’pme’ is specified the non-bonded terms will us PME for electrostatics and\na long-range periodic correction for van der Waals, otherwise a simple model\nwith no cutoff will be used.\n11.33 energy\nenergy [<name>] [<mask1>] [out <filename>] [nobondstoh] [openmm [<mdopts>]]\n[bond] [angle] [dihedral] {[nb14]|[e14]|[v14]} {[nonbond]|[elec] [vdw]}\n[{nokinetic|kinetic [ketype {vel|vv}] [dt <dt>]}]\n[ etype { simple |\ndirectsum [npoints <N>] |\newald [cut <cutoff>] [dsumtol <dtol>] [ewcoeff <coeff>]\n[erfcdx <dx>] [skinnb <skinnb>] [ljswidth <width>]\n[rsumtol <rtol>] [maxexp <max>] [mlimits <X>,<Y>,<Z>] |\npme [cut <cutoff>] [dsumtol <dtol>] [ewcoeff <coeff>]\n[erfcdx <dx>] [skinnb <skinnb>] [ljswidth <width>]\n[order <order>] [nfft <nfft1>,<nfft2>,<nfft3>]\n[ljpme [ewcoefflj <ljcoeff>]]\n} ]\n[<name>] Data set name.\n[<mask1>] Mask of atoms to calculate energy for.\n[out <filename>] File to write results to.\n[nobondstoh] Skip calculating the energy of bonds to\nhydrogen.\n[openmm] If specified and CPPTRAJ is compiled with\nOpenMM support, use OpenMM to calculate energy.\nNote this will only calculate total energy; any\nkeywords pertaining to individual energy components\nare not available. For a list of potential options\nthat can be used with ’openmm’, see 7.7 on page 38.\n[bond] Calculate bond energy.\n[angle] Calculate angle energy.\n[dihedral] Calculate dihedral energy.\n[nb14] Calculate nonbonded 1-4 energy.\n129"}, {"page": 130, "text": "[e14] Calculate 1-4 electrostatics.\n[v14] Calculate 1-4 van der Waals.\n[nonbond] Calculate nonbonded energy (electrostatics\nand van der Waals).\n[elec] Calculate electrostatic energy (Coulomb\npotential).\n[vdw] Calculate van der Waals energy (Lennard-Jones 6-12\npotential).\n[nokinetic] Do not calculate kinetic energy even if\nvelocity/force information present.\n[kinetic] Attempt to calculate kinetic energy. Requires\nforce and/or velocity information.\nketype {vel|vv} Specify kinetic energy type. If not\nspecified, if velocity and force information use\na velocity verlet-type calculation (vv), i.e.\nassume velocities are a half-step ahead of the\nforces. If only velocity information is\npresent, calculate from on-step velocities\n(vel).\ndt <dt> Time step for vv calculation in ps.\n[etype <type>] Calculate electrostatics via specified\ntype.\n[simple] Use simple Coulomb term for electrostatics, no\ncutoff.\n[directsum] Use direct summation method for\nelectrostatics.\n[npoints <N>] Number of cells in each direction to\ncalculate the direct sum.\n[ewald] Use Ewald summation for electrostatics. If van\nder Waals energy will be calculated a long-range\ncorrection for periodicity will be applied.\ncut <cutoff> Direct space cutoff in Angstroms\n(default 8.0).\ndsumtol <dtol> Direct sum tolerance (default\n0.00001). Used to determine Ewald coefficient.\newcoeff <coeff> Ewald coefficient in 1/Ang.\nerfcdx <dx> Spacing to use for the ERFC splines\n(default 0.0002 Ang.).\nskinnb Used to determine pairlist atoms (added to\ncut, so pairlist cutoff is cut + skinnb);\nincluded in order to maintain consistency with\nresults from sander.\n130"}, {"page": 131, "text": "ljswidth <width> If specified, use a\nforce-switching form for the Lennard-Jones\ncalculation from <cutoff>-<width> to <cutoff>.\nrsumtol <rtol> Reciprocal sum tolerance (default\n0.00005). Used to determine number of\nreciprocal space vectors.\nmlimits <X>,<Y>,<Z> Explicitly set the number\nof reciprocal space vectors in each dimension.\nWill be determined automatically if not\nspecified.\n[pme] Use particle mesh Ewald for electrostatics. If\nvan der Waals energy will be calculated a long-range\ncorrection for periodicity will be applied.\ncut <cutoff> Direct space cutoff in Angstroms\n(default 8.0).\ndsumtol <dtol> Direct sum tolerance (default\n0.00001). Used to determine Ewald coefficient.\newcoeff <coeff> Ewald coefficient in 1/Ang.\nerfcdx <dx> Spacing to use for the ERFC splines\n(default 0.0002 Ang.).\nskinnb Used to determine pairlist atoms (added to\ncut, so pairlist cutoff is cut + skinnb);\nincluded in order to maintain consistency with\nresults from sander.\nljswidth <width> If specified, use a\nforce-switching form for the Lennard-Jones\ncalculation from <cutoff>-<width> to <cutoff>.\norder <order> Spline order for charges.\nnfft <nfft1>,<nfft2>,<nfft3> Explicitly set the\nnumber of FFT grid points in each dimension.\nWill be determined automatically from unit cell\ndimensions if not specified.\nljpme If specified use particle mesh Ewald for\ncalculating Lennard-Jones interactions.\newcoefflj Ewald coefficient for Lennard-Jones PME\n(implies ljpme).\nDataSet Aspects:\n[bond] Bond energy.\n[angle] Angle energy.\n[dih] Dihedral energy.\n[vdw14] 1-4 van der Waals energy.\n[elec14] 1-4 electrostatic energy.\n131"}, {"page": 132, "text": "[vdw] van der Waals energy.\n[elec] Electrostatic energy.\n[kinetic] Kinetic energy.\n[total] Total energy.\nCalculate the energy for atoms in <mask>. If no terms are specified, all terms\narecalculated. Notethatthenon-bondedenergytermsfor’simple’donottake\ninto account periodicity and there is no distance cut-off. Electrostatics can also\nbe determined via the direct sum, Ewald, or particle-mesh Ewald summation\nprocedures. The particle mesh Ewald functionality requires that CPPTRAJ be\ncompiled with FFTW and a C++11 compliant compiler.\nCalculation of energy terms requires that the associated topology file have\nparameters for any of the calculated terms, so for example angle calculations\nare not possible when using a PDB file as a topology, etc. All nonbonded\ncalculations methods other than simple require unit cell parameters.\nFor example, to calculate all energy terms and write to a Grace-format file:\nparm DPDP.parm7\ntrajin DPDP.nc\nenergy DPDP out ene.agr\n11.34 esander\nesander [<name>] [out <filename>] [saveforces] [parmname <file>] [keepfiles]\n[<namelist vars> ...]\n[<name>] Data set name.\n[out <filename>] File to write results to.\n[saveforces] If specified, save forces to frames.\nRequires writing frames in NetCDF format.\n[parmname <file>] Name of temporary topology file\n(default: ’CpptrajEsander.parm7’).\n[keepfiles] Keep temporary topology file after program\nexecution.\n[<namelist vars>] Namelist variables supported by the\nsander API in format ’var <value>’; see below.\nCalculate energies for input frames using the sander API. It requires compila-\ntion with the SANDER API (sanderlib). This can be considered as a faster\nalternative to energy post-processing with sander (imin = 5). Currently the\nfollowing sander namelist variables are supported: extidel, intdiel, rgbmax,\nsaltcon,cut,dielc,igb,alpb,gbsa,lj1264,ipb,inp,vdwmeth,ew_type,\nntb, ntf, ntc. See ?? on page ?? for details.\n132"}, {"page": 133, "text": "Ifntb/cut/igbarenotspecifiedcpptrajwillattempttopickreasonablevalues\nbased on the input system. The defaults for a non-periodic system are ntb=0,\ncut=9999.0, igb=1. The defaults for a periodic system are ntb=1, cut=8.0,\nigb=0. This currently requires writing a temporary Amber topology, the name\nof which can be set by parmname. If keepfiles is specified this temporary\ntopology will not be deleted after execution.\nFor example, to calculate energies for a non-periodic system using igb=1\n(the default) with GB surface area turned on (gbsa=1):\nparm DPDP.parm7\ntrajin DPDP.nc\nesander DPDP out Edpdp.dat gbsa 1\n11.35 filter\nfilter {<dataset arg> min <min> max <max> ...} [out <file>] [name <setname>]\n{[multi] | [filterset <set> [newset <newname>]] [countout <countfile>]}\n<dataset arg> Data set name(s) to use for filtering\nmin <min> Allow values greater than <min> in\ndataset(s).\nmax <max> Allow values greater than <max> in\ndataset(s).\n[out <file>] File containing 1 for frames that were\nallowed, 0 for frames that were filtered.\n[name <setname>] Filtered data set name containing 1\nfor allowed frames, 0 for filtered frames.\n[multi] Filter each set separately instead of all\ntogether (creates filter set for each input set).\nCannot be used with ’filterset’.\n[filterset <set>] If specified, <set> will be filtered to\nonly contain data that satisfies cutoffs. Cannot be\nused with ’multi’.\n[newset <newname>] If specified a new set will be\ncreated from ’filterset’ instead of replacing\n’filterset’.\n[countout <count>] If specified, write number of\nelements passed and filtered to <countfile>. Cannot\nbe used with ’multi’.\nSets Created (not ’multi’)\n<setname> For each input element contains 1 for\nelements that “passed”, 0 otherwise.\n133"}, {"page": 134, "text": "<setname>[npassed] Number of elements that passed.\n<setname>[nfiltered] Number of elements filtered out.\nSets Created (’multi’)\n<setname>:<idx> For each input set (number with\n<idx>, starting from 0) contains 1 for elements that\n“passed”, 0 otherwise.\nFor all following actions, only include frames that are between <min> and\n<max> of data sets in <dataset arg>. There must be at least one <min>\nand <max> argument, and there must be as many <min>/<max> argu-\nments as there are specified data sets. If ’multi’ is specified then only filter\ndatasetswillbecreatedforeachdatasetinstead. If ’filterset’isspecified, the\nspecified<set>willbemodifiedtoonlycontain’1’frames;cannotbeusedwith\n’multi’. If ’newset’ is also specified, a new set will be created containing the\n’1’ frames instead. The ’filterset’ functionality only works for 1D scalar sets.\nIf ’countout’ is specified, the final number of elements passed and filtered out\nwill be written to <countfile>.\nForexample,towriteonlyframesin-betweenanRMSDof0.7-0.8Angstroms\nfor a given input trajectory:\ntrajin ../tz2.truncoct.nc\nrms R1 first :2-11\nfilter R1 min 0.7 max 0.8 out filter.dat\nouttraj maxmin.crd\nThe output trajectory will only contain frames that meet the RMSD require-\nment, and the filter.dat file can be used to see which frames those were that\nwere output.\nA similar command that can be used with data that already exists (e.g. it\nhas been read in with readdata) is datafilter (see page 54).\n11.36 fixatomorder\nfixatomorder [outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\n[pdborder [hetatm <mask>]] (EXPERIMENTAL)\noutprefix <prefix> Write re-ordered topology to\n<prefix>.<originalname>\n[nobox] Remove any box information from the re-ordered\ntopology.\nparmout <filename> Write re-ordered topology to\n<filename>\nparmopts <list> Options for writing topology file\n134"}, {"page": 135, "text": "pdborder (EXPERIMENTAL) Try to reorder atoms according\nto PDB information.\nhetatm <mask> Mark atoms in mask as HETATM, order\nthem after other atoms.\nCpptraj(andmostofAmber)expectsthatatomindicesinmoleculestoincrease\nmonotonically. However, occasionally atom indices in molecules can become\ndisordered or non-sequential, in which case cpptraj will print an error message\nsuch as the following:\nError: Atom 45 was assigned a lower molecule # (1) than previous atom (2).\nand:\nError: Could not determine molecule information for <topology file>.\n. This command fixes atom ordering so that all atoms in molecules are sequen-\ntial. The outprefix keyword will write out the re-ordered topology with name\n<name>.<original name>.\nFor example, given an out of order topology named ’outoforder.parm7’ and\na corresponding trajectory ’min1.crd’, the following will produce a reordered\ntopology named ’reorder.outoforder.parm7’ and a reordered trajectory named\n’reorder.mdcrd’:\nparm outoforder.parm7\ntrajin min1.crd 1 10\nfixatomorder outprefix reorder\ntrajout reorder.mdcrd\nIf ’pdborder’isspecified,attempttoorganizeatomsbyPDBinformation(i.e.\nChain ID, original residue numbering, and insertion codes). Atoms optionally\nspecified by ’hetatm <mask>’ will be placed after all other atoms. Note that\nthe ’pdborder’ keyword is still experimental, and requires that the Topology have\nPDB-type information present.\n11.37 fiximagedbonds\nfiximagedbonds [<mask>]\n<mask> Mask expression of atoms to check.\nFixbondsthathavebeensplitacrossperiodicboundaryconditionsbyimaging.\nIt may be desirable to reimage the coordinates after this with autoimage.\n135"}, {"page": 136, "text": "11.38 gist (Grid Inhomogeneous Solvation Theory)\ngist [name <dataset name>] [doorder [nopl] [plcut <plcut>]]\n[doeij] [skipE] [skipS] [refdens <rdval>] [temp <tval>]\n[noimage] [gridcntr <xval> <yval> <zval>]\n[rmsfit <fitmask>]\n[griddim <nx> <ny> <nz>] [gridspacn <spaceval>] [neighborcut <ncut>]\n[prefix <filename prefix>] [ext <grid extension>] [out <output suffix>]\n[floatfmt {double|scientific|general}] [floatwidth <fw>] [floatprec <fp>]\n[intwidth <iw>]\n[info <info suffix>]\n[nopme|pme [cut <cutoff>] [dsumtol <dtol>] [ewcoeff <coeff>]\n[erfcdx <dx>] [skinnb <skinnb>] [ljswidth <width>]\n[order <order>] [nfft <nfft1>,<nfft2>,<nfft3>]]\n[name <dataset name>] Name for output data sets.\n[doorder] Calculate the water order parameter [7] for\neach voxel.\n[nopl] If specified, do not use the pair list for\nthe order calculation (may be much slower).\n[plcut <plcut>] Pair list cutoff for order\ncalculation (default 10.0 Ang.).\n[doeij] Calculate the triangular matrix representing the\nwater-water interactions between pairs of voxels\n(see below).\n[skipE] Skip all energy calculations (cannot be\nspecified with ’doeij’).\n[skipS] Skip all entropy calculations.\n[refdens rdval>] Reference density of bulk water, used\nin computing g_O, g_H, and the translational\nentropy. Default is 0.0334 molecules/Å3.\n[temp <tval>] Temperature of the input trajectory.\n[noimage] Disable distance imaging in energy\ncalculation.\n[gridcntr <xval> <yval> <zval>] Coordinates (Å) of the\ncenter of the grid (default 0.0, 0.0, 0.0).\n[rmsfit <fitmask>] If specified, grid will be centered\nand rotated to follow atoms selected by <fitmask>.\n[griddim <nx> <ny> <nz>] Grid dimensions (number of\nbins/voxels) along each coordinate axis (default 40,\n40, 40).\n136"}, {"page": 137, "text": "[gridspacn <spaceval>] Grid spacing (linear dimension\nof each voxel) in Angstroms. Values greater than\n0.75 Å are not recommended (default 0.5 Å).\n[neighborcut <ncut>] Cutoff in Å for determining\nsolvent O-O neighbors (default 3.5 Å).\n[prefix <filename prefix>] Output file name prefix\n(default “gist”).\n[ext <grid extension>] Output grid file name extension\n(default “.dx”).\n[out <output suffix>] Suffix for main GIST output file\nname. If not specified, output file will be set to\n’<prefix>-output.dat’.\n[floatfmt {double|scientific|general}] Format for floating\npoint values in GIST output file: double (regular\nfixed decimal point), scientific, or general\n(default, chooses fixed or scientific, whichever\nfits better).\n[floatwidth <fw>] Changes width of floating point values\nin GIST output file. Default is no width\nrestriction.\n[floatprec <fp>] Changes precision of floating point\nvalues in GIST output file. Default is to use\nwhatever the system default is.\n[intwidth <iw>] Changes width of integer values in GIST\noutput file. Default is no width restriction.\n[info <info suffix>] Suffix for main GIST info file name.\nIf not specified, info will be written to standard\noutput.\n[oldnnvolume] Use the old reference volume for the\nnearest neighbor entropy, instead of the more\nprecice new implementation.\n[nnsearchlayers <nlayers>] Number of layers of\nneighboring voxels that should be used when\nsearching for nearest neighbors. This has to be at\nleast 1 to obtain the correct entropy. Higher\nvalues can help to obtain better convergence of the\ntranslational and 6D entropy with little sampling or\nfine grid spacings, but increase the calculation\ntime (default 1).\n[solute <mask>] Selection mask for the solute. All\nother molecules will be solvent. If this is\nomitted, the standard solute/solvent assignment will\nbe used.\n137"}, {"page": 138, "text": "[solventmols <MOLS>] Comma-separated list of names of\nsolvent molecules. Energies will be computed per\nsolvent molecule. For the entropy, only the main\nsolvent (the first one) will be used. Use, e.g.,\nsolventmols WAT,NA,CL for a GIST calculation\nincluding ions. This needs to be specified if there\nis more than one solvent species.\n[nocom] Do not use the center of mass to define the\nmolecular position. Instead, use the first atom in\nrigidatoms. Use this flag to restore the behavior\nof old GIST runs.\n[rigidatoms <CENTRAL> <SUBST1> <SUBST2>]\nSpecifies how to define the molecular orientation\nfor the entropy. By default, a simple heuristic\nwill be used. This works for water, but not for all\nsolvents. The atoms should be representative of the\nmolecular orientation and should not be collinear.\nNote that the central atom goes first. For water,\nthe default is equivalent to rigidatoms O H1 H2,\ncorresponding to H1-O-H2 as the rigid substructure.\n[nopme] Do not use particle mesh Ewald for the\nnon-bonded calculation (default).\n[pme] Use particle mesh Ewald for the non-bonded\nelectrostatics calculation. The van der Waals\nenergy will be calculated using a long-range\ncorrection for periodicity. Does not support doeij.\ncut <cutoff> Direct space cutoff in Angstroms\n(default 8.0).\ndsumtol <dtol> Direct sum tolerance (default\n0.00001). Used to determine Ewald coefficient.\newcoeff <coeff> Ewald coefficient in 1/Ang.\nerfcdx <dx> Spacing to use for the ERFC splines\n(default 0.0002 Ang.).\nskinnb Used to determine pairlist atoms (added to\ncut, so pairlist cutoff is cut + skinnb);\nincluded in order to maintain consistency with\nresults from sander.\nljswidth <width> If specified, use a\nforce-switching form for the Lennard-Jones\ncalculation from <cutoff>-<width> to <cutoff>.\norder <order> Spline order for charges.\nnfft <nfft1>,<nfft2>,<nfft3> Explicitly set the\nnumber of FFT grid points in each dimension.\n138"}, {"page": 139, "text": "Will be determined automatically if not\nspecified.\nDataSet Aspects:\n[gO] Number density of oxygen centers found in the\nvoxel, in units of the bulk density.\n[gH] Number density of hydrogen centers found in the\nvoxel in units of the reference bulk density.\n[Esw] Mean solute-water interaction energy density.\n[Eww] Mean water-water interaction energy density.\n[dTStrans] First order translational entropy density.\n[dTSorient] First order orientational entropy density.\n[dTSsix] First order six-dimensional entropy density.\n[neighbor] Mean number of waters neighboring the water\nmolecules found in this voxel multiplied by the\nvoxel number density.\n[dipole] Magnitude of mean dipole moment (polarization).\n[order] Average Tetrahedral Order Parameter.\n[dipolex] x-component of the mean water dipole moment\ndensity\n[dipoley] y-component of the mean water dipole moment\ndensity\n[dipolez] z-component of the mean water dipole moment\ndensity\n[Eij] Water-water interaction matrix.\n[PME] (pme only) Mean water energy on the GIST grid.\n[U_PME] (pme only) Mean solute energy on the GIST\ngrid.\nDataSets if the main solvent is not water:\n[gELEM] For every element ELEM in the main solvent,\nthe atomic density relative to rho0 (e.g., gC and gH\nfor benzene).\nDataSets if there are multiple solvents:\n[g_mol_NAME] for every solvent species NAME (e.g.,\ng_mol_WAT and g_mol_NA if solventmols WAT,NA was\nspecified).\n[Esw_mol_NAME] for every solvent species NAME\n(e.g., Esw_mol_WAT and Esw_mol_NA if solventmols\nWAT,NA was specified).\n139"}, {"page": 140, "text": "Figure 1: Diagram, in 2D, of GIST’s gridded water properties in a binding site.\n[Eww_mol_NAME] for every solvent species NAME\n(e.g., Eww_mol_WAT and Eww_mol_NA if solventmols\nWAT,NA was specified).\nGrid Inhomogeneous Solvation Theory [8, 9] (GIST) is a method for analyzing\nthestructureandthermodynamicsofsolventinthevicinityofasolutemolecule.\nThecurrentimplementationworksforonlywater,butthemethodcanbegener-\nalizedtoothersolventswhosemoleculesarerigidlikewater, suchaschloroform\nor dimethylsulfoxide (DMSO). GIST post-processes explicit solvent simulation\ndata to create a three-dimensional mapping of water density and thermody-\nnamicpropertieswithinaregionofinterest, whichisdefinedbyauser-specified\n3D rectangular grid. The small grid boxes are referred to as voxels, and each\nvoxel is associated with solvent properties. (See Fig. 1.) The GIST implemen-\ntation incorporated into AmberTools cpptraj also calculates a number of other\nlocalwaterproperties,aslistedbelow. GISTworksforthenonpolarizablewater\nmodels currently supported by AMBER.\nIn order to carry out a GIST calculation, you must have a trajectory file\ngenerated with explicit water, as well as the corresponding topology file. To\ngeneratethemostreadilyinterpretableresults,itisrecommendedthatthesolute\n(e.g.,aprotein)berestrainedintoessentiallyoneconformation. GISTwillthen\nprovide information about the structure and thermodynamics of the solvent for\nthat conformation. For a room-temperature simulation of a solvent-exposed\nbindingsite, andagrid-spacingof0.5Å,itisrecommendedthatthesimulation\nbeatleast10-20nsinduration,anditisalsoagoodideatocheckforconvergence\nof the GIST properties you are interested in by loading and then processing\nsuccessively more frames of your trajectory file. Because GIST assumes that\nthe solute of interest comprises all molecules in the simulation that are not\nwaters, it is a good idea to remove all counterions and cosolutes with cpptraj’s\nstrip command before running GIST. A sample series of cpptraj commands for\nrunning GIST is provided below.\nAlthough it is not mandatory to supply values of gridcntr, griddim and\ngridspcn,theseparametersshouldbecarefullychosen,becausetheydetermine\nthe region to be analyzed (gridcntr and griddim) and the spatial resolution\nand convergence properties of the results (gridspcn). In particular, although\n140"}, {"page": 141, "text": "smaller grid spacings will give finer spatial resolution, longer simulation times\nwill be needed to converge the properties in the smaller voxels that result. A\nlarger grid spacing will allow earlier convergence, but will smooth the spatial\ndistributions. Whencomputingthesumovervoxelvaluesinalargerregion,the\nresult is independent of the grid spacing as long as nlayers is high enough for\nthe given sampling.\nThe reference density of water (rdval) is taken by default to be the exper-\nimental number density of pure water at 300 K and 1 atm. However, different\nwater models may yield slightly different bulk densities under these conditions,\nand the density also depends on T and P. If you know that the bulk density\nof the water model you are using, at the T and P of your simulation, deviates\nsignificantly from 0.0334 water molecules/Å3, it would be advisable to supply\nthe actual value with the refdens keyword, instead of allowing GIST to supply\nthe default value.\nFor GIST, a GPU accelerated version is available, in which the interaction\nenergy is calculated using CUDA. When using the GPU accelerated version of\nGIST, the doeij keyword is not available. It is recommended to use a grid\ncovering the entire box, when using the GPU implementation. You may also\nchooseasmallergrid,butallinteractionenergies,i.e.,eachatomwitheachatom,\nwillalwaysbecalculatedindependentofthechosengrid. Thisensuresoptimum\nperformance when calculating the interaction energies. Thus, the additional\ntime required to calculate the order parameters (doorder) is negligible.\nThe nonbonded energy calculation can also be accelerated using particle\nmesh Ewald via the pme keyword (CPU only).[10]\nGIST Output\nGIST generates a main output file and a collection of grid data files that\nby default are in Data Explorer format (.dx); this can be changed via the ext\nkeyword. These grid files enable visualization of the various gridded quantities,\nsuch as with the program VMD [11]. If the doeij keyword is provided, GIST\nalso writes out a matrix of water-water interactions between pairs of voxels. In\naddition, run details are written to stdout, which can be redirected into a log\nfile.\nNote that a number of quantities are written out as both densities and\nnormalized quantities. For example, the output file includes both the solute-\nwaterenergydensityandthenormalized(perwater)solute-waterenergy. Inall\ncases,thenormalizedquantityatvoxeli,X isrelatedtothecorresponding\ni,norm\ndensity,X ,bytherelationshipX =ρ X ,whereρ isthenumber\ni,dens i,norm i i,dens i\ndensity of water in the voxel. The normalized quantity provides information\nregarding the nature of the water found in the voxel. The density has the\nproperty that, if the grid extended over the entire simulation volume, the total\n141"}, {"page": 142, "text": "systemquantitywouldbegivenbyX =V (cid:80) X ,whereV isthe\ntot voxel i i,dens voxel\nvolume of one grid voxel.\nThe main output file takes the form of a space-delimited-variable file, where\neach row corresponds to one voxel of the grid. This file can easily be opened\nwith and manipulated with spreadsheet programs like Excel and LibreOffice\nCalc. The columns are as follows.\n• index - A unique, sequential integer assigned to each voxel\n• xcoord - x coordinate of the center of the voxel (Å)\n• ycoord - y coordinate of the center of the voxel (Å)\n• zcoord - z coordinate of the center of the voxel (Å)\n• population - Number of water molecule, n , found in the voxel over the\ni\nentire simulation. A water molecule is deemed to populate a voxel if its\noxygen coordinates are inside the voxel. The expectation value of this\nquantity increases in proportion to the length of the simulation.\n• g_O - Number density of oxygen centers found in the voxel, in units of\nthe bulk density (rdval). Thus, the expectation value of g_O for a neat\nwater system is unity.\n• g_H - Number density of hydrogen centers found in the voxel in units\nof the reference bulk density (2×rdval). Thus, the expectation value of\ng_H for a neat water system would be unity.\n• g_ELEM - (if the main solvent is not water) Density of every further\nelement ELEM in the main solvent. Scaled such that the expectation\nvalue in pure solvent is unity.\n• g_mol_NAME - (if there is more than one solvent) Density of every\nsolvent species NAME specified in solventmols. Scaled by rho0.\n• dTStrans-dens-Firstordertranslationalentropydensity(kcal/mole/Å3),\nreferenced to the translational entropy of bulk water, based on the value\nrdval.\n• dTStrans-norm - First order translational entropy per water molecule\n(kcal/mole/molecule), referenced to the translational entropy of bulk wa-\nter, based on the value rdval. The quantity dTStrans-norm equals\ndTStrans-dens divided by the number density of the voxel in units of\nnumber/Å3.\n• dTSorient-dens-Firstorderorientationalentropydensity(kcal/mole/Å3),\nreferenced to bulk solvent (see below).\n• dTSorient-norm - First order orientational entropy per water molecule\n(kcal/mole/water), referenced to bulk solvent (see below). This quantity\nequals dTSorient-dens divided by the number density of the voxel.\n142"}, {"page": 143, "text": "• Esw-dens-Meansolute-waterinteractionenergydensity(kcal/mole/Å3).\nThisistheinteractionofthesolventinagivenvoxelwiththeentiresolute.\nBoth Lennard-Jones and electrostatic interactions are computed without\nanycutoff,withintheminimumimageconventionbutwithoutEwaldsum-\nmation. This quantity is referenced to bulk, in the trivial sense that the\nsolute-solvent interaction energy is zero in bulk.\n• Esw-norm - Mean solute-water interaction energy per water molecule\n(kcal/mole/molecule). ThisequalsEsw-densdividedbythenumberden-\nsity of the voxel.\n• Eww-dens - Mean water-water interaction energy density, scaled by ½\nto prevent double-counting, and not referenced to the corresponding bulk\nvalue of this quantity (see below). This quantity is one half of the mean\ninteraction energy of the water in a given voxel with all other waters\nin the system, both on and off the GIST grid, divided by the volume\nof the voxel (kcal/mole/Å3). Unless PME is used, both Lennard-Jones\nandelectrostaticinteractionsarecomputedwithoutanycutoff,withinthe\nminimum image convention.\n• Esw_mol_NAME-dens and Esw_mol_NAME-norm - (if there\nare multiple solvent species) Mean solute-solvent energy per molecule of\nspecies NAME, for each solvent specified in solventmols. Follows the\nsame conventions as Esw.\n• Eww_mol_NAME-dens and Eww_mol_NAME-norm - (if there\nare multiple solvent species) Mean solvent-solvent energy per molecule of\nspecies NAME, for each solvent specified in solventmols. Follows the\nsame conventions as Eww.\n• PME-norm - (Only if PME was used) Mean PME solvent energy per\nwater molecule (kcal/mole/molecule). This equals PME-dens divided\nby the number density of the voxel.\n• PME-dens - (Only if PME was used) Mean PME solvent energy density\n(kcal/mole/Å3). This corresponds roughly to Eww-norm plus one half\nof Esw-norm in a non-PME GIST calculation.\n• Eww-norm - Mean water-water interaction energy, normalized to the\nmean number of water molecules in the voxel (kcal/mole/water). See\nprior column definition for details.\n• Dipole_x-dens-x-componentofthemeanwaterdipolemomentdensity\n(Debye/Å3).\n• Dipole_y-dens-y-componentofthemeanwaterdipolemomentdensity\n(Debye/Å3).\n• Dipole_z-dens-z-componentofthemeanwaterdipolemomentdensity\n(Debye/Å3).\n143"}, {"page": 144, "text": "• Dipole-dens-Magnitudeofmeandipolemoment(polarization)(Debye/Å3).\n• Neighbor-dens-Meannumberofwatersneighboringthewatermolecules\nfound in this voxel multiplied by the voxel number density. Two waters\nareconsideredneighborsiftheiroxygensarewithin3.5angstromsofeach\nother. For any given frame, the contribution to the average is set to zero\nif no water is found in the voxel (units of number/Å3).\n• Neighbor-norm-Meannumberofneighboringwatermolecules,perwa-\nter molecule found in the voxel (units of number per water).\n• Order-norm - Average Tetrahedral Order Parameter [7], q , for water\ntet\nmolecules found in the voxel, normalized by the number of waters in the\nvoxel. The order parameter for water i in a given frame is given by:\nq (i) = 1− 3(cid:80)3 (cid:80)4 (cosϕ + 1)2 where j and k index the 4\ntet 8 j=1 k=j+1 ijk 3\nclosest water neighbors to water i, and ϕ is the angle formed by water\nijk\ni, j, and k. If the doorder keyword is not provided or is set to FALSE,\nthen this calculation will not be done, and the entries in this column will\nbe set to zero.\nGrid files are provided for all computed quantities listed above, except that the\nnormalizedquantitiesarenotincluded. Thefilenamesareasfollows: gist-gO.dx,\ngist-gH.dx,gist-dTStrans-dens.dx,gist-dTSorient-dens.dx,gist-Esw-dens.dx,gist-\nEww-dens.dx, gist-dipolex-dens.dx, gist-dipoley-dens.dx, gist-dipolez-dens.dx,\ngist-dipole-dens.dx,gist-neighbor-dens.dx,gist-neighbor-norm.dx,gist-order-norm.dx.\nIf the doorder keyword is not provided, then the data in gist-order-norm.dx\nwill all be zeroes. Note that the file of voxel water densities, gist-gO.dx, can\nbe used as input to the program Placevent [12], in order to define spherical\nhydration sites based on the density distribution. More detailed descriptions of\nthe files follows:\n• gist-dipole-dens.dx - Magnitude of mean dipole moment (polarization)\n(Debye/Å3).\n• gist-dipolex-dens.dx - X-component of the mean water dipole moment\ndensity (Debye/Å3).\n• gist-dipoley-dens.dx- Y-component of the mean water dipole moment\ndensity (Debye/Å3).\n• gist-dipolez-dens.dx - Z-component of the mean water dipole moment\ndensity (Debye/Å3).\n• gist-dTSorient-dens.dx - Density weighted first order orientational en-\ntropy(kcal/mole/Å3). Themorenegativetheisovalue,themorerestricted\nor unfavorable the orientation of the water is. The value at bulk density\nis expected to be zero, most values will fall between 0 and -1.\n144"}, {"page": 145, "text": "• gist-dTStrans-dens.dx - Density weighted first order tranlational en-\ntropy (kcal/mole/Å3). The more negative isovalue, the more restricted\nthewaterispositionally. Thevalueatbulkdensityisexpectedtobezero,\nmost values will fall between 0 and -1.\n• gist-Esw-dens.dx - Density weighted Solute-Water interaction energy\n(kcal/mole/Å3). Thisistheinteractionofthewaterwiththeentiresolute.\nBoth Lennard-Jones and electrostatic interactions are computed without\nanycutoff,withintheminimumimageconvention. Themorenegativethe\nnumber, the more favorable the interaction between the water and solute\nis. Isovalues will be negative numbers.\n• gist-Eww-dens.dx - Density weighted Water-Water interaction energy\n(kcal/mole/Å3). This is the interaction of the water with the entire sol-\nvent. Both Lennard-Jones and electrostatic interactions are computed\nwithout any cutoff, within the minimum image convention. The more\nnegative the number, the more favorable the interaction between water\nwithin this voxel and all other water molucules is. Isovalues will be nega-\ntive numbers.\n• gist-gH.dx - Number density of hydrogen centers found in the voxel, in\nunits of the bulk density. The expectation value of g_H for a neat water\nsystem is unity. The units of this dx file are the density/bulk density.\nTherefore an isovalue of 1 represents every voxel at or greater than bulk\ndensity.\n• gist-gO.dx - Number density of oxygen centers found in the voxel, in\nunits of the bulk density. The expectation value of g_O for a neat water\nsystem is unity. The units of this dx file are the density/bulk density.\nTherefore an isovalue of 1 represents every voxel at or greater than bulk\ndensity.\n• gist-neighbor-norm.dx-Meannumberofneighboringwatermolecules,\nper water molecule found in the voxel (units of number per water). Two\nwatersareconsideredneighborsiftheiroxygensarewithin3.5Angstroms\nof each other.\n• gist-order-norm.dx - Average Tetrahedral Order Parameter for water\nmolecules found in the voxel, normalized by the number of waters in the\nvoxel. doorder must be declared in the command line for this file to be\nproduced.\nSimilar grid files with other computed quantities can be generated by reading\nthegist.outfileintoaspreadsheetprogram,processingthenumberstogenerate\na new column of voxel data of interest, and writing this column to an ascii text\nfile. Then the Perl script write_dx_file.pl, which should be available on the\nGIST tutorial web-site, may be used to read in the column of data and create\nthe corresponding dx file. The input format, and an example, are as follows:\n145"}, {"page": 146, "text": "./write_dx_file.pl [filename] [x-dimension y-dimension z-dimension]\n[x-origin y-origin z-origin] [grid spacing]\n./write_dx_file.pl file.dat 40 40 40 13.0 13.0 13.0 0.75\nIf the doeij keyword is provided, GIST also writes a large file, Eww_ij.dat,\ncontaining the mean water-water interaction energies between pairs of voxels,\nscaledby½. (Seebelow.) Thisfilehasthreecolumns. Thefirsttwocolumnsare\nvoxel indexes, i, j, where j > i, so that no pair appears more than once, and\nthe third column is the mean interaction energy (kcal/mole) of water in voxels\ni andj, scaled by ½. If the occupancy of either voxel is 0, such as for voxels\ncovered by solute atoms, then the interaction energy is zero. In order to save\nspace, such interactions are omitted from the file.\nSample cpptraj input file to run GIST\nThefollowinginputfile,gist.in,causescpptrajtoreadaparameterfilenamed\ntopology.top; read in the first 5000 frames of the trajectory file named trajecto-\nryfile.mdcrd;stripoutallNaandClions;andcarryoutaGISTrunwhichcom-\nputesorderparameters, usesa41x41x45gridcenteredat(25.0, 31.0, 30.0)with\na spacing of 0.5 Å, uses the default bulk water density of 0.0334 molecules/Å3,\nand generates the main output file gist.out.\nparm topology.top\ntrajin trajectoryfile.mdcrd 1 5000\nstrip @Na\nstrip @Cl\ngist doorder doeij gridcntr 25.0 31.0 30.0 griddim 41 41 45\ngridspacn 0.50 out gist.out\ngo\nTo execute this run in the background, use\ncpptraj<gist.in>gist.log& or cpptraj –i gist.in>gist.log&\nReferencing GIST results to unperturbed (bulk) water\nInhomogeneousfluidsolvationtheory,whichisthebasisofGIST,isdesigned\nto provide information on how water structure and thermodynamics around a\nsolute molecule, such as a protein, are changed relative to the structure and\n146"}, {"page": 147, "text": "Water Model Mean Energy (Eww-norm) (kcal/mol/water) Number Density (Å−3)\nTIP3P -9.533 0.0329\nTIP4PEW -11.036 0.0332\nTIP4P -9.856 0.0332\nTIP5P -9.596 0.0329\nTip3PFW -11.369 0.0334\nSPCE -11.123 0.0333\nSPCFW -11.873 0.0329\nTable 3: Water model energy and density.\nthermodynamics of unperturbed (bulk) water. Accordingly, the quantities re-\nported by GIST are most informative when the results are referenced to the\ncorresponding bulk water properties. For the orientational entropy, the refer-\nencevalueisthesameregardlessofwatermodelorconditions, becausethefirst\norder orientational distribution of water in the bulk is always uniform. There-\nfore,theGISTresultsfororientationalentropiesarealreadyreferencedtobulk.\nHowever, cpptraj reports unreferenced values for those GIST quantities whose\nreference values depend upon the water model and the simulation conditions;\ni.e., the energies. The translational entropy as well as the number densities will\nbe referenced to bulk using the input referenced density or the default density\nvalueof0.0334. Thetablebelowprovidesusefulreferencevaluesforthesequan-\ntities, computed for various water models at P=1atm, T=300K, using GIST in\norder to ensure a consistent minimum image treatment of periodic boundary\nconditions.\nUsers running calculations under significantly different conditions, or with\ndifferent water models, should consider generating their own reference quanti-\nties by applying GIST to a simulation of pure water under their conditions of\ninterest. The quantities of interest can then be obtained in their most precise\navailable form by averaging over voxels, for the pure water simulation. If the\n(cid:80)\nquantityofinterestisQ,thenitsaveragereferencevalueisQ reference = (cid:80) niQi,\nni\nwhere Q and n are, respectively, GIST’s reported values of the quantity and\ni i\nthe population in voxel i. The densities, ρ , are referenced to the correspond-\ni\ning bulk densities, ρo, as g = ρ /ρo, while the energy and entropy terms are\ni i\nreferenced by subtracting their bulk values.\nNote that the Eww reference needs to be subtracted from the normal-\nized water-water energy Eww-norm. A referenced Eww-dens can be ob-\ntained by multiplying the referenced Eww-norm by the solvent number den-\nsity ρ=g O ρo = Nf N V w vox , where N w is the number of water molecules in a voxel\n(population), N is the number of frames, and V is the voxel volume.\nf vox\nInterpreting GIST results\n147"}, {"page": 148, "text": "GIST provides access to the first order entropies and the first- and second-\norder energies of inhomogeneous fluid solvation theory. Non-zero higher-order\nentropies exist but are not yet computationally accessible. However, for a pair-\nwise additive force-field, such as those listed in the Table above, the energy is\nfully described at the second order provided by GIST.\nGISTisaresearchtool, anditsapplications(to, forexample, protein-ligand\nbinding and protein function) are still being explored. The following general\ncomments may be helpful to users studying GIST results.\n1. The water in voxels near a solute (e.g., a protein) almost always has\nunfavorable water-water interaction energies, relative to bulk, simply because\nthe solute displaces water, resulting in fewer proximal water-water interactions.\n2. The unfavorable water-water energies mentioned in [8] may be balanced\nby favorable water-solute interactions. If they are not, as may occur especially\nfor voxels in small, hydrophobic pockets, then the net energy of the water in\nthe voxel may be unfavorable relative to bulk, in which case a ligand which\ndisplaces water from the voxel into bulk may get a boost in affinity.\n3. Becausethefirstorderorientationaldistributionofbulkwaterisuniform,\nandanonuniformdistributionalwayshaslowerentropythanauniformone,the\nsolute can only lower the orientational entropy of water, relative to bulk. Thus,\nthis term always opposes solvation, and displacing oriented water into the bulk\nis always favorable from the standpoint of orientational entropy.\n4. Localized water, which corresponds to voxels with high water density,\nhasalowfirstordertranslationalentropy,andthetranslationalentropyaround\na solute is lower than that in bulk, as a nonuniform translational distribution\ntakes the place of the uniform translational distribution of bulk water.\n5. The displacement of highly oriented (low orientational entropy) and lo-\ncalized (low translational entropy) water into bulk leads to a favorable increase\nin these entropy terms.\n6. However, highly oriented and localized water is often the consequence of\nstrongly favorable polar interactions, such as hydrogen-bonding, between water\nandthesolute. Asaconsequence,thenetfavorabilityofdisplacingsuchwateris\nfrequently a balance between favorable entropic consequences and unfavorable\nenergetic consequences.\n7. The water-water energy associated with a given voxel accounts for the\ninteractions of the waters in this voxel with all other waters in the system,\nincluding waters in other voxels. This quantity is multiplied by ½, so that, in\na pure-water system where the GIST grid covers the entire simulation box, the\nsumoverallvoxelsequalsthecorrectmeanwater-waterinteractionenergy. Note\nthat Reference [9] does not include this factor of ½.\n8. ForatypicalGISTapplication,inwhichthegridoccupiesonlypartofthe\nsimulationbox,theenergybookkeepingcanbecomecomplicated,asdiscussedin\nSection II.B.3 (page 044101-6) of Reference [9]. That section also explains how\none can compute the water-water energy associated with a region R defined by\na set of voxels, ER . The regional water-water energy, on a normalized (per\nWW\nwater) basis, is given by ER = 2( (cid:80) E − (cid:80) (cid:80) E )\nWW i∈R i,WW i∈R j∈R,j>i i,j,WW\nwhere i ∈ R means that voxel i is in region R, E is the value of Eww-\ni,WW\n148"}, {"page": 149, "text": "normforvoxeli,andE isthevalueofthewater-waterinteractionenergy\ni,j,WW\nbetween voxels i and j, taken from the file Eww_ij.dat. The extra factor of 2\nin the present formula, relative to that in the paper, results from application of\nan extra factor of ½ to the reported water-water interaction energies here.\n9. If the GIST grid contains the entire solute and the calculation is suffi-\ncently converged, the energy and first-order entropy of hydration can be cal-\nculated by numerical integration. E.g., the energy of hydration is ∆E =\nhyd\n(cid:80)voxels(Edens +Edens)×V . For this, Eww has to be referenced carefully,\nsw ww vox\nsince numerical inaccuracies can add up quickly. It can be advisable to omit all\nvoxels above a certain distance to the solute to obtain more stable results.\n11.39 grid\ngrid [out <filename>]\n{ data <dsname> | boxref <ref name/tag> <nx> <ny> <nz> |\n<nx> <dx> <ny> <dy> <nz> <dz>\n[ { gridcenter <cx> <cy> <cz> |\nboxcenter |\nmaskcenter <mask> |\nrmsfit <mask> [noxalign]} ]\n[box|origin|center <mask>] [negative] [name <gridname>]\n<mask> [normframe | normdensity [density <density>]]\n[pdb <pdbout> [max <fraction>]] [{byres|mymol}]\n[[smoothdensity <value>] [invert]] [madura <madura>]\n[out <filename>] File to write out grid to. Use\n“.grid” or “.xplor” extension for XPLOR format,\n“.dx” for OpenDX format.\nOptions for setting up grid:\ndata <dsname> Use previously calculated/loaded grid\ndata set named <dsname>. When using this option\nthere is no need to specify grid\nbins/spacing/center.\nboxref <ref name/tag> <nx> <ny> <nz> Set up grid\nusing box information from a previously loaded\nreference structure. Currently the only way to set\nup non-orthogonal grids.\n<nx> <dx> <ny> <dy> <nz> <dz> Number of grid\nbins and spacing in the X/Y/Z directions.\n[gridcenter <cx> <cy> <cz>] Location of grid center,\ndefault is origin (0.0, 0.0, 0.0).\n[boxcenter] Center grid on box center.\n[maskcenter <mask>] Center the grid on the atoms\nselected by <mask>.\n149"}, {"page": 150, "text": "[rmsfit <mask>] Perform a best-fit rotation of the grid\nusing the coordinates selected by <mask>.\n[noxalign] If specified, grid will not be\nre-oriented to align with Cartesian axes once\nbinning is finished. Will affect file formats\nthat do not store full unit cell vectors (like\nXplor).\nOptions for offset during grid binning (must center grid\nat origin):\n[box] Offset each point by location of box center prior\nto gridding. Cannot be used with ’gridcenter’.\n[origin] No offset (default)\n[center <mask>] Offset each point by center of atoms in\n<mask> prior to gridding. Cannot be used with\n’gridcenter’.\nOther options:\n[negative] Grid negative density instead of positive\ndensity.\n[name <gridname>] Grid data set name.\n<mask> Mask of atoms to grid.\n[normframe] Normalize grid bins by the number of\nframes.\n[normdensity [density <density>]] Normalize grid bins by\ndensity: GridBin = GridBin / (Nframes * BinVolume *\ndensity). Default particle density\n(molecules/Ang^3) for water based on 1.0 g/mL.\n[pdb <pdbout> [max <fraction>]] Write a pseudo-PDB of\ngrid points that have density greater than\n<fraction> (default 0.80) of the grid max value.\n[{byres|bymol}] Grid the centers of mass of residues or\nmolecules selected by <mask>.\nLess common options:\n[smoothdensity <smooth>] Used to smooth density. The\nsmoothing takes the form of GridBin = 0 if GridBin <\nsmooth, otherwise GridBin = GridBin - (GridBin *\nexp[-(GridBin - smooth)^2 / (0.2 * smooth^2)]).\n[invert] (Only used if smoothdensity also used) Do\ninverse smoothing (i.e. if GridBin > smooth).\n[madura <madura>] Grid values lower than <madura>\nbecome flipped in sign, exposes low density.\n150"}, {"page": 151, "text": "Data Sets Created:\n<dsname> Grid data set.\nCreateagridrepresentingthehistogramofatomsinmask1 onthe3Dgridthat\nis \"nx * x_spacing by ny * y_spacing by nz * z_spacing angstroms (cubed).\nBy default the grid is centered at the origin unless gridcenter is specified.\nGrid points can be offset by either the box center (using box) or the center of\nspecified atoms (using center <mask>); if either of these options are used\nthe grid must be centered at the origin. Note that the bounds command ( on\npage 110) can be very useful for determining grid dimensions.\nNote that when calculating grid densities for things like solvent/ions, the\nsolute of interest (about which the atomic densities are binned) should be rms\nfit,centeredandimagedpriortothegrid callinordertoprovideanymeaningful\nrepresentationofthedensity. Iftheoptionalkeywordnegativeisalsospecified,\nthen these density will be stored as negative numbers. Output can be in the\nXPLOR or OpenDX data formats.\nExamples\nExample 1: Grid water density around a solute. The solute is imaged to the\norigin and rms fit to the first frame. The grid will be centered on the origin as\nwell.\ntrajin tz2.truncoct.nc\nautoimage origin\nrms first :1-13\n# Create average of solute to view with grid.\naverage avg.mol2 :1-13\ngrid out.dx 20 0.5 20 0.5 20 0.5 :WAT@O\nExample 2: Grid water density around a solute. The grid is centered on the\nsolute.\ntrajin tz2.truncoct.nc\nautoimage\ngrid out.dx 20 0.5 20 0.5 20 0.5 :WAT@O maskcenter :1-13\nExample 3: Grid water density around a solute. The grid is centered on the\nsolute and rms-fit. The density obtained should be equivalent to the first ex-\nample.\ntrajin tz2.truncoct.nc\nimage :WAT\ngrid out.dx 20 0.5 20 0.5 20 0.5 :WAT@O rmsfit :1-13\nExample 4: Generate grid from bounds command.\n151"}, {"page": 152, "text": "trajin tz2.ortho.nc\nautoimage\nrms first :1-13&!@H= mass\nbounds :1-13 dx .5 name MyGrid out bounds.dat\naverage bounds.mol2 :1-13\n# Save coordinates for second pass.\ncreatecrd MyCoords\nrun\n# Grid using grid data set from bounds command.\ncrdaction MyCoords grid bounds.xplor data MyGrid :WAT@O\nExample 5: Create non-orthogonal grid based on the box.\ntrajin tz2.truncoct.nc\nreference ../tz2.truncoct.nc [REF]\nautoimage triclinic\ngrid nonortho.dx boxref [REF] 50 50 50 :WAT@O pdb nonortho.pdb\n11.40 hbond\nhbond [<dsname>] [out <filename>] [<mask>] [angle <acut>] [dist <dcut>]\n[donormask <dmask> [donorhmask <dhmask>]] [acceptormask <amask>]\n[avgout <filename>] [printatomnum] [nointramol] [image]\n[solventdonor <sdmask>] [solventacceptor <samask>]\n[solvout <filename>] [bridgeout <filename>] [bridgebyatom]\n[series [uuseries <filename>] [uvseries <filename>]]\n[bseries [bseriesfile <filename>]]\n[uuresmatrix [uuresmatrixnorm {none|frames}] [uuresmatrixout <file>]]\n[splitframe <comma-separated-list>]\n[<dsname>] Data set name.\n[out <filename>] Write # of solute-solute hydrogen\nbonds (aspect [UU]) vs time to this file. If\nsearching for solute-solvent hydrogen bonds, write #\nof solute-solvent hydrogen bonds (aspect [UV]) and #\nof bridging solvent molecules (aspect [Bridge]), as\nwell as the residue # of the bridging solvent and\nthe solute residues being bridged with format\n’<solvent resnum>(<solute res1>+<solute\nres2>+...+),...’ (aspect [ID]).\n[<mask>] Atoms to search for solute hydrogen bond\ndonors/acceptors.\n[angle <acut>] Angle cutoff for hydrogen bonds (default\n135°). Can be disabled by specifying -1.\n152"}, {"page": 153, "text": "[dist <dcut>] Distance cutoff for hydrogen bonds\n(acceptor to donor heavy atom, default 3.0 Å).\n[donormask <dmask>] Use atoms in <dmask> as solute\ndonor heavy atoms. If ’donorhmask’ not specified\nonly atoms bonded to hydrogen will be considered\ndonors.\n[donorhmask <dhmask>] Use atoms in <dmask> as solute\ndonor hydrogen atoms. Should only be specified if\n’donormask’ is. Should be a 1 to 1 correspondence\nbetween donormask and donorhmask.\n[acceptormask <amask>] Use atoms in <amask> as solute\nacceptor atoms.\n[avgout <filename>] Write solute-solute hydrogen bond\naverages to <filename>.\n[printatomnum] Add atom numbers to the output, in\naddtion to residue name, residue number and atom\nname.\n[nointramol] Ignore intramolecular hydrogen bonds.\n[image] Turn on imaging of distances/angles.\n[solventdonor <sdmask>] Use atoms in <sdmask> as\nsolvent donors. Can specify ions as well.\n[solventacceptor <samask>] Use atoms in <samask> as\nsolvent acceptors. Can specify ions as well.\n[solvout <filename>] Write solute-solvent hydrogen bond\naverages to <filename>. If not specified and\n’avgout’ is, solute-solvent hydrogen bonds averages\nwill be written to that file.\n[bridgeout <filename>] Write information on detected\nsolvent bridges to <filename>. If not specified,\nwill be written to same place as ’solvout’.\n[bridgebyatom] Report bridging results by atom instead\nof by residue.\n[series] Save hydrogen bond formed (1.0) or not formed\n(0.0) per frame for any detected hydrogen bond.\nSolute-solute hydrogen bonds are saved with aspect\n[solutehb], solute-solvent hydrogen bonds are saved\nwith aspect [solventhb].\n[uuseries <filename>] File to write solute-solute\nhbond time series data to.\n[uvseries <filename>] File to write solute-solvent\nhbond time series data to.\n153"}, {"page": 154, "text": "[bseries] Save bridge formed (1.0) or not formed (0.0)\nper frame for any detected bridge. Bridges are\nsaved with aspect [bridge_<indexlist>], where\n<indexlist> is an underscore (’_’) delimited list of\nbridged atom/residue numbers (depending on\nbridgebyatom).\n[bseriesfile <filename>] File to write bridge time\nseries data to.\n[uuresmatrix] If specified, create a matrix with aspect\n[UUresmat] containing # of hydrogen bonds between\neach possible solute residue pair.\n[uuresmatrixnorm {none|frames}] Control how matrix\nis normalized: none=no normalization,\nframes=normalize by total # frames.\n[uuresmatrixout <file>] If specified, write matrix\ndata to specified file.\n[splitframe <comma-separated-list>] If specified, aplit\nthe average hydrogen bond (avgout, solvout,\nbridgeout) analysis into sections delimited by the\nframe numbers. For example, ’splitframe\n250,500,1000’ will divide analysis into frames\n1-249, 250-499, 500-999, and 1000 to end.\nData Sets Created:\n<dsname>[UU] Number of solute-solute hydrogen bonds.\n<dsname>[UV] (only for solventdonor/solventacceptor)\nNumber of solute-solvent hydrogen bonds.\n<dsname>[Bridge] (only for\nsolventdonor/solventacceptor) Number of bridging\nsolvent molecules.\n<dsname>[ID] (only for solventdonor/solventacceptor)\nString identifying bridging solvent residues and the\nsolute residues they bridge.\n<dsname>[solutehb] (series only) Time series for\nsolute-solute hydrogen bonds; 1 for present, 0 for\nnot present.\n<dsname>[solventhb] (series only) Time series for\nsolute-solvent hydrogen bonds; 1 for present, 0 for\nnot present.\n<dsname>[bridge_<indexlist>] (bseries only) Time\nseries for bridge; 1 for present, 0 for not present.\nThe <indexlist> is an underscore (’_’) delimited\nlist of bridged atom/residue numbers (depending on\nbridgebyatom).\n154"}, {"page": 155, "text": "<dsname>[UUresmatr] (uuresmatrix only) Solute\nresidue hydrogen bond matrix.\nNote that series data sets are not generated until hydrogen bonds are actually\ndetermined (i.e. run is called).\nDetermine hydrogen bonds in each coordinate frame using simple geometric\ncriteria. A hydrogen bond is defined as being between an acceptor heavy atom\nA,adonorhydrogenatomH,andadonorheavyatomD.IftheAtoDdistance\nis less than or equal to the distance cutoff and the A-H-D angle is greater than\nor equal to the angle cutoff a hydrogen bond is considered formed. Imaging of\ndistances/angles is not performed by default, but can be turned on using the\nimage keyword.\nPotential hydrogen bond donor/acceptor atoms are searched for as follows:\n1. If just <mask> is specified donors and acceptors will be automatically\ndetermined from <mask>.\n2. If donormask is specified donors will be determined from <dmask>\n(only atoms bonded to hydrogen will be considered valid). Optionally,\ndonorhmask can be used in conjunction with donormask to explicitly\nspecify the hydrogen atoms bonded to donor atoms. Acceptors will be\nautomatically determined from <mask>.\n3. Ifacceptormaskisspecifiedacceptorswillbedeterminedfrom<amask>.\nDonors will be automatically determined from <mask>.\n4. If both acceptormask and donormask are specified only <amask>\nand <dmask> will be used; no searching will occur in <mask>.\nAutomaticdeterminationofhydrogenbonddonors/acceptorsusesthesimplistic\ncriterionthat“hydrogenbondsareFON”,i.e.,hydrogensbondedtoF,O,andN\natoms are considered donors, and F, O, and N atoms are considered acceptors.\nIntra-molecularhydrogenbondscanbeignoredusingthenointramolkeyword.\nThe number of hydrogen bonds present at each frame will be determined\nand written to the file specified by out. If desired, the bridge [ID] data can be\nused in conjunction with the keep command to generate structures that only\ncontain bridging solvent (11.43 on page 159). If the series keyword is specified\nthetimeseriesforeachhydrogenbond(1forpresent,0fornotpresent)willalso\nbe saved for subsequent analysis (e.g. with lifetime, see on page 257); solute-\nsolute hydrogen bonds will be saved to ’<dataset name>[solutehb]’ and solute-\nsolventhydrogenbondswillbesavedto’<datasetname>[solventhb]’. Thedata\nsetlegendsaresetwiththeresiduesandatomsinvolvedinthehydrogenbonds.\nInthecaseofsolutetonon-specificsolventhydrogenbonds,aVisusedinplace\nof solvent.\nIfavgoutisspecifiedtheaverageofeachsolute-solutehydrogenbond(sorted\nby population) formed over the course of the trajectory is printed with the for-\nmat:\n155"}, {"page": 156, "text": "Acceptor DonorH Donor Frames Frac AvgDist AvgAng\nwhere Acceptor, DonorH, and Donor are the residue and atom name of the\natomsinvolvedinthehydrogenbond,Frames isthenumberofframesthebond\nis present, Frac is the fraction of frames the bond is present, AvgDist is the\naverage distance of the bond when present, and AvgAng is the average angle\nof the bond when present. The printatomnum keyword can be used to print\natom numbers as well.\nSolute to non-specific solvent hydrogen bonds can be tracked by using the\nsolventdonor and/or solventacceptor keywords. The number of solute-\nsolventhydrogenbondsandnumberof“bridging” solventmolecules(i.e. solvent\nthat is hydrogen bonded to two or more different solute residues at the same\ntime) will also be written to the file specified by out. These keywords can also\nbe used to track non-specific interactions with ions. If avgout or solvavg is\nspecified the average of each solute solvent hydrogen bond will be printed with\nthe format:\nAcceptor DonorH Donor Count Frac AvgDist AvgAng\nwhere Acceptor, DonorH, and Donor are either the residue and atom name of\nthesoluteatomsor“SolventAcc”/”SolventH”/”SolventDnr” representingsolvent,\nCount is the total number of interactions between solute and solvent (note\nthis can be greater than the total number of frames since for any given frame\nmore than one solvent molecule can hydrogen bond to the same place on solute\nand vice versa), AvgDist is the average distance of the bond when present, and\nAvgAng istheaverageangleofthebondwhenpresent. If avgoutorbridgeout\nisspecifiedinformationonresiduesthatwerebridgedbyasolventmoleculeover\nthe course of the trajectory will be written to <bfilename> with format:\nBridge Res <N0:RES0> <N1:RES1> ... , <X> frames.\nwhere’<N0:RES0>...’ isalistofresiduesthatwerebridged(residue#followed\nby residue name) and <X> is the number of frames the residues were bridged.\nhbond Examples\nTo search for all hydrogen bonds within residues 1-22, writing the number of\nhydrogenbondsperframeto“nhb.dat” andinformationoneachhydrogenbond\nfound to “avghb.dat”:\nhbond :1-22 out nhb.dat avgout avghb.dat\nTo search for all hydrogen bonds formed between donors in residue 1 and ac-\nceptors in residue 2:\nhbond donormask :1 acceptormask :2 out nhb.dat avgout avghb.dat\n156"}, {"page": 157, "text": "To search for all intermolecular hydrogen bonds only and solute-solvent hydro-\ngen bonds, saving time series data to HB:\nhbond HB out nhb.dat avgout solute_avg.dat \\\nsolventacceptor :WAT@O solventdonor :WAT \\\nsolvout solvent_avg.dat bridgeout bridge.dat \\\nseries uuseries uuhbonds.agr uvseries uvhbonds.agr\nTosearchfornon-specifichydrogenbondsbetweensoluteandionsnamedNa+:\nhbond HB-Ion out nhb.agr avgout ion_avg.dat \\\nsolventacceptor :Na+ solventdonor :Na+\n11.41 image\nimage [origin] [center] [triclinic | familiar [com <commask>]] [<mask>]\n[ bymol | byres | byatom ] [xoffset <x>] [yoffset <y>] [zoffset <z>]\n[origin] Image to coordinate origin (0.0, 0.0, 0.0);\ndefault is to image to box center.\n[center] For bymol/byres, image by center of mass;\ndefault is to image by first atom position.\n[triclinic] Force imaging with triclinic code. This is\nthe default for non-orthorhombic cells.\n[familiar [com <commask>]] Image to truncated\noctahedron shape. If ’com <commask>’ is given,\nimage with respect to the center of mass of atoms in\n<commask>.\n[<mask>] Image atoms/residues/molecules in mask.\n[bymol] Image by molecule (default).\n[byres] Image by residue.\n[byatom] Image by atom.\n[xoffset <x>] Shift atoms by a factor of <x> in the\nX-direction.\n[yoffset <y>] Shift atoms by a factor of <y> in the\nY-direction.\n[zoffset <z>] Shift atoms by a factor of <z> in the\nZ-direction.\nNotethiscommandisintendedforadvanceduse;formostcasestheautoimage\ncommand should be sufficient.\nFor periodic systems only, image molecules/residues/atoms that are out-\nside of the box back into the box. Currently both orthorhombic and non-\northorhombicboxesaresupported. Atypicaluseof image istomovemolecules\n157"}, {"page": 158, "text": "back into the box after performing center. For example, the following com-\nmandsmoveallatomssothatthecenterofresidue1isatthecenterofthebox,\nthen image so that all molecules that are outside the box after centering are\nwrapped back inside:\ncenter :1\nimage\nThe xoffset etc. keywords can be used to shift the entire unit cell in a certain\ndirection by the given factor, which can be useful for visualizing trajectories\nwith periodic boundary conditions. For example, to generate a trajectory that\nis offset by 1.0 box length in the X direction, one could use:\nimage xoffset 1.0\ntrajout traj.offsetx1.nc\n11.42 jcoupling\njcoupling <mask> [outfile <filename>] [kfile <param file>] [out <filename>]\n[name <dsname>]\n<mask> Atom mask in which to search for dihedrals\nwithin.\n[outfile <filename>] File to write j-coupling values to\nwith fixed format.\n[kfile <param file>] File containing Karplus parameters.\nIf not specified will check CPPTRAJHOME, AMBERHOME,\nand KARPLUS environment variables (see below).\n[out <filename>] File to write data set output to.\n[name <dsname>] Data set name.\nNote data sets are not generated until run is called.\nCalculate J-coupling values for all dihedrals found within <mask> (all\natoms if no mask given). In order to use this function, Karplus parameters\nfor all dihedrals which will be calculated must be loaded. By default cpptraj\nwill use the data found in either $CPPTRAJHOME/dat/Karplus.txt or $AM-\nBERHOME/dat/Karplus.txt; if this is not found cpptraj will look for the file\nspecified by the $KARPLUS environment variable.\nIn the Karplus parameter file each parameter set consists of two lines for\neach dihedral with the format:\n[<Type>]<Name1><Name2><Name3><Name4><A><B><C>[<D>]\n<Resname1>[<Resname2>...]\n158"}, {"page": 159, "text": "The first line defines the parameter set for a dihedral. <Type> is optional;\nif not given the form for calculating the J-coupling will be as described by\nChou et al.[13]; if ’C’ the form will be as described by Perez et al.[14]. The\n<NameX> parameters define the four atoms involved in the dihedral. Each\n<NameX>parameteris5characterswide,startingwithaplus’+’,minus’-’or\nspace ’ ’ character indicating the atom belongs to the next, previous, or current\nresidue. Theremaining4charactersaretheatomname. Theparameters<A>,\n<B>, <C>, and <D> are floating point values 6 characters wide describing\nthe Karplus parameters. For the ’C’ form A, B, and C correspond to C0, C1,\nand C2; D is unused and should not be specified. The second line is a list of\nresidue names (4 characters each) to which the dihedral applies. For example:\nC HA CA CB HB 5.40 -1.37 3.61\nILE VAL\nDescribesadihedralbetweenatomsHA-CA-CB-HBusingthePerezetal. form\nwith constants C0=5.40, C1=-1.37, C2=3.61 applied to ILE and VAL residues.\nOutput can be in both a fixed format (outfile <filename>) and using\ncpptrajdataset/datafileformatting(out <filename>). Thefixedformathas\neach dihedral that is defined from <mask1> printed along with its calculated\nJ-coupling value for each frame, e.g.:\n#Frame 1\n1 SER HA CA CB HB2 45.334742 4.024759\n1 SER HA CA CB HB3 -69.437134 1.829510\n...\nFirst the frame number is printed, then for each dihedral: Residue number,\nresidue name, atom names 1-4 in the dihedral, the value of the dihedral, the\nJ-coupling value.\nIn cpptraj format, only the J-coupling value is written.\n11.43 keep\nkeep [ bridgedata <bridge data set> [nbridge <#>] [nobridgewarn]\n[bridgeresname <res name>] bridgeresonly <resrange>] ]\n[keepmask <atoms to keep>] [charge <new charge>]\n[outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\nbridgedata <bridge data set> Data set containing bridge\nID strings from the hbond command (11.40 on\npage 152).\nnbridge <#> Number of bridging residues to keep\n(default 1).\n159"}, {"page": 160, "text": "nobridgewarn If specified, suppress warnings for\nwhen active # bridges does not equal requested\nnumber.\nbridgeresname <res name> Name of bridging\nresidues (default ’WAT’).\nbridgeresonly <range> If specified, only keep\nbridges that bridge residues in the <resrange>\nlist.\nkeepmask <atoms to keep> Mask of atoms to keep.\ncharge <new charge> Scale charges so total charge of\nremaining atoms matches the specified <new charge>.\noutprefix <prefix> Write modified topology to\n<prefix>.<originalname>\n[nobox] Remove any box information from the modified\ntopology.\nparmout <filename> Write modified topology to\n<filename>.\nparmopts <list> Options for writing topology file.\nKeeponlyspecifiedatoms(oppositeof strip). Thiscanalsobeusedinconjunc-\ntion with output from the hbond command to retain solute and only bridging\nresidues (e.g. bridging waters). Forexample, thefollowingrun generatesbridg-\ning data with the ’hbond’ command in a first pass, then uses the bridge ID\ndata to retain only 1 single bridging water between residues 10 and 11:\nparm tz2.ortho.parm7\ntrajin tz2.ortho.nc\n# First pass, generate bridge time series\nhbond hb solventacceptor :WAT@O solventdonor :WAT out hb.dat\nrun\n# Second pass, retain only frames where the bridge is present\n# for residues 10 and 11.\nkeep bridgedata hb[ID] nbridge 1 bridgeresonly 10,11 parmout keep.parm7\n# Write trajectory\ntrajout keep.nc\nrun\nThis run reads in bridge ID data from a previous hbond run and uses it to keep\nonly residues 10, 11, and a bridging water:\nparm tz2.ortho.parm7\ntrajin tz2.ortho.nc\nreaddata hb.dat\nkeep keepmask :10,11 bridgedata hb.dat:5 nbridge 1 bridgesonly 10,11 \\\nparmout keep.10.11.parm7\ntrajout keep.10.11.nc\nrun\n160"}, {"page": 161, "text": "11.44 lessplit\nlessplit [out <filename prefix>] [average <avg filename>] <trajout args>\n[out <filename prefix>] Write split LES trajectories to\n<filename prefix>.X, where X is an integer.\n[average <avg filename>] Write trajectory of averaged\nLES regions to <avg filename>.\n<trajout args> Arguments for output trajectories.\nSplit and/or average LES trajectory. At least one of ’out’ or ’average’ must\nbe specified. If both are specified they share <trajout args>.\n11.45 lie\nlie [<name>] <Ligand mask> [<Surroundings mask>] [out <filename>] [nopbc]\n[noelec] [novdw] [cutvdw <cutoff>] [cutelec <cutoff>] [diel <dielc>]\nDataSet Aspects:\n[EELEC] Electrostatic energy (kcal/mol).\n[EVDW] van der Waals energy (kcal/mol).\nFor each frame, calculate the non-bonded interactions between all atoms in\n<Ligand mask> with all atoms in <Surroundings mask>. Electrostatic and\nvanderWaalsinteractionswillbecalculatedforallatompairs. Aseparateelec-\ntrostaticandvanderWaalscutoffcanbeapplied,thedefaultis12.0Angstroms\nfor both. <dielc> is an optional dielectric constant. Either the electrostatic\nor van der Waals calculations can be suppressed via the keywords noelec and\nnovdw, respectively. Periodic boundary conditions (and the minimum image\nconvention) can be abandoned with the “nopbc” keyword. Note, however, that\nno prior imaging is performed if the frames contain periodic boundaries. This\nmay be useful for instances when you are simulating a microscopic droplets.\nThe electrostatic interactions are calculated according to a simple shifting\nfunctionshownbelow. Thedatafilewillcontaintwodatasets—oneforelectro-\nstatic interactions and one for van der Waals interactions. Periodic topologies\nand trajectories are required (i.e., explicit solvent is necessary). The minimum\nimage convention is followed.\n(cid:32) (cid:33)2\nq q r2\nE =k i j 1− ij\nelec r r2\nij cut\n11.46 lipidorder\norder out <filename> [x|y|z] [scd] [unsat <mask>]\n[taildist <filename> [delta <resolution>] tailstart <mask>\ntailend <mask>] <mask0> ... <maskN>\n161"}, {"page": 162, "text": "out Output file for order parameters: Sx, Sy, Sz (each\nsucceeded by the standard deviation), and two\nestimates for the deuterium-order parameter |SCD| =\n0.5Sz and |SCD| = -(2Sx + Sy)/3. If scd is set then\nthe order parameter directly computed from the C-H\nvectors is output.\nx|y|z Reference axis. (z)\nunsat Mask for unsaturated bonds. Sz is calculated for\nvector Cn-Cn+1. This is only relevant if scd\n(below) is not set, i.e. order parameters are\ncalculated from carbon position only.\nscd Calculate the deuterium-order parameter |SCD|\ndirectly from the C-H vectors (masks must contain\nC-H-H triplets, see below). Otherwise the order\nparameter is estimated from carbon positions only\n(masks must contain only relevant carbons). (false)\ntaildist Optional output file for end-to-end distances.\ndelta Optional resolution for taildist. (0.1)\ntailstart Mask for the start of the tail. Must be given\nif taildist.\ntailend Mask for the end of the tail. Must be given if\ntaildist.\nmask0 ... maskN Masks for each group in the lipid\nchain.\nThe order parameters S , S , S and |SCD| are calculated. Carbons must be\nx y z\ngiven in bonding order. If scd the masks must be made up of C-H-H triples,\nhencehydrogenstodoublebondsmustbeenumeratedtwicewhilemethylgroups\nrequire an additional mask which will also create two entries in the output.\nS is the vector joining carbons C and C , S the vector normal to\nz n−1 n+1 x\nthe C −C and C −C plane and S is the third axis in the molecular\nn−1 n n n+1 y\ncoordinate system. The order parameter is then calculated from Sc = 0.5 <\n3cos(2θ) > −1, where θ is the angle to the chosen reference axis. See example\ninput file.\nExample input (all atom names according to CHARMM27 force field for\nPOPC).\nsn1 chain: order parameters S , S , S and |SCD|=0.5×S and |SCD|=\nx y z z\n−(2S +S )/3\nx y\nlipidorder out sn1.dat z taildist e2e_sn1.dat delta 0.1 \\\ntailstart \":POPC@C32\" tailend \":POPC@C316\" \\\n\":POPC@C32\" \":POPC@C33\" \":POPC@C34\" \":POPC@C35\" \\\n\":POPC@C36\" \":POPC@C37\" \":POPC@C38\" \":POPC@C39\" \\\n162"}, {"page": 163, "text": "\":POPC@C310\" \":POPC@C311\" \":POPC@C312\" \":POPC@C313\" \\\n\":POPC@C314\" \":POPC@C315\" \":POPC@C316\"\nSee also $AMBERHOME/AmberTools/test/cpptraj/Test_LipidOrder.\n11.47 lipidscd\nlipidscd [<name>] [<mask>] [{x|y|z}] [out <file>] [p2]\n<name> Output data set name.\n<mask> Atom mask specifying where to search for\nlipids.\nx|y|z Axis to calculate order parameters with respect to\n(default z).\nout <file> File to write order parameters to.\np2 If specified, report raw <P2> values.\nDataSets Generated:\n<name>[H1]:<idx> Hold lipid order parameters for\neach C-H1. Each lipid type will have a different\n<idx> starting from 0.\n<name>[H2]:<idx> Hold lipid order parameters for\neach C-H2. If no H2, the C-H1 value will be used.\n<name>[H3]:<idx> Hold lipid order parameters for\neach C-H3. If no H3, the C-H2/C-H1 value will be\nused.\n<name>[SDHX]:<idx> Hold standard deviation of lipid\norder parameters for each C-HX.\nCalculatelipidorderparametersSCD(|<P2>|)forlipidchainsinmask<mask>.\nLipid chains are identified by carboxyl groups, i.e. O-(C=O)-C1-...-CN, where\nC1 is the first carbon in the acyl chain and CN is the last. Order parameters\nwillbedeterminedforeachhydrogenbondedtoeachcarbon. If’p2’isspecified\nthe raw <P2> values will be reported.\n11.48 makestructure\nmakestructure <List of Args>\nApply dihedrals to specified residues using arguments found in <List of Args>,\nwhere an argument is 1 or more of the following arg types:\n163"}, {"page": 164, "text": "<sstype keyword>:<res range>\nApply secondary structure type (via phi/psi backbone angles) to residues in\ngiven range. If the secondary structure type is a turn, the residue range must\ncorrespond to a multiple of 2 residues.\nKeyword phi, psi (deg.) # residues\nalpha -57.8, -47.0 1\nleft -57.8, 47.0 1\npp2 -75.0, 145.0 1\nhairpin -100.0, 130.0 1\nextended -150.0, 155.0 1\ntypeI -60.0, -30.0 | -90.0, 0.0 2\ntypeII -60.0, 120.0 | 80.0, 0.0 2\ntypeVIII -60.0, -30.0 | -120.0, 120.0 2\ntypeI’ 60.0, 30.0 | 90.0, 0.0 2\ntypeII 60.0, -120.0 | -80.0, 0.0 2\ntypeVIa1 -60.0, 120.0 | -90.0, 0.0 2\ntypeVIa2 -120.0, 120.0 | -60.0, 0.0 2\ntypeVIb -135.0, 135.0 | -75.0, 160.0 2\n<custom ss name>:<res range>[:<phi>:<psi>]\nIf <phi> and <psi> are given, define a custom secondary structure conforma-\ntionnamed<custom_ss>andapplytoresiduesinrange. If<custom_ss>has\nbeen previously defined then apply it to residues in range.\n<customturnname>:<resrange>[:<phi1>:<psi1>:<phi2>:<psi2>]\nIf <phi1>, <psi1>, <phi2>, and <psi2> are given, defined a custom turn\nconformation named <custom_turn> and apply to residues in range (range\nmust correspond to a multiple of 2 residues). If <custom_turn> has been\npreviously defined then apply it to residues in range.\n<custom dih name>:<res range>[:<dih type>:<angle>]\n<dih type> = alpha beta gamma delta epsilon zeta nu0 nu1 nu2 nu3 nu4\nh1p c2p chin phi psi chip omega chi2 chi3 chi4 chi5\nIf <dih type> and <angle> are given, apply <angle> to selected dihedrals of\ntype in range. If <custom dih> has been previously defined then apply it to\nresidues in range.\n<customdihname>:<resrange>[:<at0>:<at1>:<at2>:<at3>:<angle>[:<offset>]]\nApply <angle> to dihedral defined by atoms <at1>, <at2>, <at3>, and\n<at4>, or use previously defined <custom_dih>.\n164"}, {"page": 165, "text": "<offset> Description\n-2 <at0> and <at1> in previous residue.\n-1 <at0> in previous residue.\n0 All atoms in single residue.\n1 <at3> in next residue.\n2 <at2> and <at3> in next residue.\nref:<range>:<refname>[:<refrange>[:<dihtypes>]][refvalsout<file>]\n[founddihout <file>]\nApplydihedralsfromresidues<ref_range>inpreviouslyloadedreferencestruc-\nture<refname>todihedralsin<range>. If<refrange>isspecified, usethose\nresidues from reference. The dihedral types to be used (see <dih_type>\nabove) can be specified in a comma-separated list; default is phi/psi. Note that\nin order to specify <dih types>, <ref range> must be specified. The ’refval-\nsout’ and ’founddihout’ keywords can be used to print dihedrals found in the\nreference and target structures respectively to files.\nExamples\nAssign polyproline II structure to residues 1 through 13:\nmakestructure pp2:1-13\nMake residues 1 and 12 ’extended’, residues 6 and 7 a type I’ turn, and two\ncustom assignments, one (custom1) for residues 2-5, the other (custom2) for\nresidues 8-11:\nmakestructure extended:1,12 \\\ncustom1:2-5:-80.0:130.0:-130.0:140.0 \\\ntypeI’:6-7 \\\ncustom2:8-11:-140.0:170.0:-100.0:140.0\nAssign residue 5 phi 90 degrees, residues 6 and 7 phi=-70 and psi=60 degrees:\nmakestructure customdih:5:phi:90 custom:6,7:-70:60\nCreate a new dihedral named chi1 and assign it a value of 35 degrees in residue\n8:\nmakestructure chi1:8:N:CA:CB:CG:35\nAssign ’extended’ structure to residues 1 and 12, a custom turn to residues 2-5\nand 8-11, and a typeI’ turn to residues 6-7:\nmakestructure extended:1,12 \\\ncustom1:2-5:-80.0:130.0:-130.0:140.0 \\\ntypeI’:6-7 \\\ncustom1:8-11\n165"}, {"page": 166, "text": "Assign secondary structure from reference structure:\nparm ../tz2.parm7\nreference ../tz2.rst7\ntrajin pp2.rst7.save\nmakestructure \"ref:1-13:tz2.rst7\" rmsd reference\ntrajout fromref.pdb multi\n11.49 mask\nmask <mask> [maskout <filename>] [out <filename>] [nselectedout <filename>]\n[name <setname>] [ {maskpdb <filename> | maskmol2 <filename>}\n[trajargs <comma-separated args>] ]\n<mask> Atom mask to process.\nmaskout <filename> Write information on atoms in\n<mask> to <filename>.\nout <filename> Write the frame, atom number, atom name,\nresidue number, residue name, and molecule number\nfor each selected atom to file.\nnselectedout <filename> Write the total number of\nselected atoms to file.\nname <setname> Name for output data sets.\nmaskpdb <filename> Write PDB of atoms in <mask> to\n<name>.X.\nmaskmol2 <filename> Write Mol2 of atoms in <mask> to\n<name>.X.\ntrajargs <comma-separated args> When writing output\nPDB/Mol2, additional trajectory arguments to pass to\nthe output trajectory.\nDataSets Created\n<name> Number of atoms selected each frame.\n<name>[Frm] Frame number for each selected atom.\n<name>[AtNum] Atom number for each selected atom.\n<name>[Aname] Atom name for each selected atom.\n<name>[Rnum] Residue number for each selected atom.\n<name>[Rname] Residue name for each selected atom.\n<name>[Mnum] Molecule number for each selected atom.\nFor each frame determine all atoms that correspond to <mask>. This is\nmost useful when using distance-based masks, since the atoms in the mask are\nupdated for every frame read in. If maskout is specified information on all\natoms in <mask> will be written to <filename> with format:\n166"}, {"page": 167, "text": "#Frame AtomNum Atom ResNum Res MolNum\nwhere#Frameistheframenumber,AtomNumisthenumberoftheselectedatom,\nAtom is the name of the selected atom, ResNum is the residue number of the\nselected atom, Res is the residue name, and MolNum is the molecule number of\nthe selected atom.\nIf maskpdb or maskmol2 are specified a PDB/Mol2 file corresponding to\n<mask> will be written out every frame with name “<name>.frame#”.\nFor example, to write out all residues within 3.0 Angstroms of residue 195\nthat are named WAT to “Res195WAT.dat”, as well as write out corresponding\nPDB files:\nmask “(:195<:3.0)&:WAT” maskout Res195WAT.dat maskpdb Res195WAT.pdb\nTowritealloutatomsoutsideof5.0AngstromsofresiduesnamedARGtoPDB\nfiles with a chain ID of ’B’:\nmask :ARG>@5.0 maskpdb Outside5Arg.pdb trajargs “chainid ’B’”\n11.50 matrix\nmatrix [out <filename>] [start <#>] [stop|end <#>] [offset <#>]\n[name <name>] [ byatom | byres [mass] | bymask [mass] ]\n[ ired [order <#>] ]\n[ {distcovar | idea} <mask1> ]\n[ {dist | correl | covar | mwcovar} <mask1> [<mask2>] ]\n[ dihcovar dihedrals <dataset arg> ]\n[out <filename>] If specified, write matrix to\n<filename>.\n[start <#>] [stop|end <#>] [offset <#>] Start, stop,\nand offset frames to use (as a subset of all frames\nread in).\n[name <name>] Name of the matrix dataset (for\nreferral in subsequent analysis).\nbyatom Write results by atom (default). This is the\nsole option for covar, mwcovar, and ired.\nbyres Write results by calculating an average for each\nresidue (mass weighted if mass is specified).\nbymask Write average over <mask1>, and if <mask2> is\nspecified <mask1> x <mask2> and <mask2> as well\n(mass weighted if mass is specified).\n167"}, {"page": 168, "text": "Calculate matrix of the specified type from input coordinate frames:\ndist <mask1> [<mask2>] Distance matrix (default).\ncorrel <mask1> [<mask2>] Correlationmatrix(akadynamiccrosscorrelation[15]).\ncovar <mask1> [<mask2>] Coordinate covariance matrix.\nmwcovar <mask1> [<mask2>] Mass-weighted coordinate covariance ma-\ntrix.\ndistcovar <mask1> Distance covariance matrix.\nidea <mask1> Isotropically Distributed Ensemble Analysis matrix.[16]\nired [order <#>] IsotropicReorientationalEigenmodeDynamicsmatrix[17]\nwith Legendre polynomials of specified order (default 1). IRED vectors\nmust have been specified previously with ’vector ired’ (see 11.91 on\npage 214).\ndihcovar dihedrals <dataset arg> Dihedral covariance matrix. Dihedral\ndata sets must have been previously defined with e.g. dihedral or mul-\ntidihedral commands or read in externally with readdata and marked\nas dihedrals.\nMatrix dimensions will be of the order of N x M for dist, correl, idea, and\nired, 2N x 2N for dihcovar, 3N x 3M for covar and mwcovar, and N(N-1) x\nN(N-1) / 4 for distcovar (with N being the number of data sets in the case of\nired and dihcovar and the number of atoms in <mask1> otherwise, and M\nbeing the number of atoms in <mask2> if specified or <mask1> otherwise).\nNo mask is required for ired; the matrix will be made up of previously defined\nIRED vectors (see the vector command on page 214). Similarly no mask is\nrequired for dihcovar; dihedral data sets must have been previously defined.\nOnly one mask can be used with distcovar and idea matrices (i.e. they can\nbe symmetric only), otherwise one or two masks can be used (for symmetric\nand full matrices respectively). If two masks are specified the number of atoms\ncoveredbymask1 mustbegreaterthanorequaltothenumberofatomscovered\nby mask2, and on output <mask1> corresponds to columns while <mask2>\ncorresponds to rows.\nNotethatforbackwardscompatibility,outputfileswrittenwith’out <file-\nname>’ will have the options ’noheader noxcol square2d’ applied to them\n(see 6 on page 27 for more details). To prevent any of these from taking effect,\nsimplyspecify’header’,’xcol’,and/or’nosquare2d’after’out<filename>’.\nAs a simple example, a distance matrix of all CA atoms is generated and\noutput to ’distmat.dat’.\nmatrix dist @CA out distmat.dat\n168"}, {"page": 169, "text": "11.51 mindist/maxdist\n{min|max}dist mask1 <mask1> [mask2 <mask2>] [{byatom|byres|bymol}]\n[noimage] [name <setname>] [out <file>] [resoffset <#>]\nmask1 <mask1> First mask for selecting atoms.\n[mask2 <mask2>] Optional second mask for selecting\natoms.\n[{byatom|byres|bymol}]\nbyatom Report the minimum or maximum distance\nbetween atoms in <mask1> or between atoms in\n<mask1> and atoms in <mask2>.\nbyres Report the minimum or maximum distance\nbetween all residue pairs selected by <mask1>,\nor pairs of residues selected by <mask1> and\nresidues selected by <mask2> (excluding certain\npairs, see resoffset).\nbymol Report the minimum or maximum distance\nbetween all molecule pairs selected by <mask1>,\nor pairs of molecules selected by <mask1> and\nmolecules selected by <mask2>.\n[noimage] Do not use the minimum image convention for\ndistances.\n[name <setname>] Output data set name.\n[out <file>] Write data to <file>.\n[resoffset <#>] For byres, ignore residue pairs if the\ndifference in residue numbers is greater than the\ncutoff (default 1).\nData Sets Created:\n<name> For byatom, a set containing the minimum or\nmaximum distance for each frame.\n<name>[<#>_<#>] For byres/bymol, a set containing\nthe minimum or maximum distance between the\nresidue/molecule pair specified by the numbers in\nthe aspect, e.g. ’<name>[1_3]’ for byres would be\nbetween residues 1 and 3.\nCalculate the minimum or maximum distance in Angstroms between atoms or\nresidue/molecule pairs.\n11.52 minimage\nminimage [<name>] <mask1> <mask2> [out <filename>] [geom] [maskcenter]\n169"}, {"page": 170, "text": "<name> Data set name.\n<mask1> First atom mask.\n<mask2> Second atom mask.\nout <filename> File to write to.\ngeom (maskcenter only) If specified, use geometric\ncenter instead of center of mass.\nmaskcenter Calculate distance from center of masks\ninstead of between each atom.\nData Sets Created:\n<name> Minimum distance to an image in Ang.\n<name>[A1] Atom number in mask 1 involved in minimum\ndistance.\n<name>[A2] Atom number in mask 2 involved in minimum\ndistance.\nCalculate the shortest distance to an image, i.e. the distance to a neighboring\nunitcell,aswellasthenumbersoftheatomsinvolvedinthedistance. Bydefault\nthe distance between each atom in <mask1> and <mask2> is considered; if\nmaskcenter is specified the center of the masks is used. By convention, the\nlower atom number is saved as A1 and the higher is saved as A2.\n11.53 molsurf\nmolsurf [<name>] [<mask>] [out filename] [probe <probe_rad>]\n[radii {gb | parse | vdw}] [offset <rad_offset>]\n[<name>] Name of surface area data set.\n[<mask>] Atoms to calculate surface area of.\n[out <filename>] File to write values to.\n[probe <probe_rad>] Probe radius (default 1.4\nAngstrom).\n[offset <rad_offset>] Add <rad_offset> to each atom\nradius (default 0.0).\n[radii {gb|parse|vdw}] Specify radii to use:\ngb GB radii (default).\nparse PARSE radii.\nvdw van der Waals radii.\nCalculate the Connolly surface area[18] of atoms in <mask> (default all atoms\nif no mask specified) using routines from molsurf (originally developed by Paul\nBeroza)usingtheproberadiusspecifiedbyprobe(1.4Åifnotspecified). Note\n170"}, {"page": 171, "text": "that if GB/VDW radii are not present in the topology file (e.g. for PDB files),\nthen PARSE[19] radii can be used. Also note that this routine only calculate\nabsolute surface areas, i.e. it cannot be used to get the contribution of a subset\nof atoms to overall surface area; if such functionality is needed try the surf\ncommand (11.83 on page 207).\n11.54 multidihedral\nmultidihedral [<name>] <dihedral types> [resrange <range/mask>] [out <filename>] [range360]\n[dihtype <name>:<a0>:<a1>:<a2>:<a3>[:<offset>] ...]\nOffset -2=<at0><at1> in previous res, -1=<at0> in previous res,\n0=All <atX> in single res,\n1=<at3> in next res, 2=<at2><at3> in next res.\n<dihedral types> = alpha beta gamma delta epsilon zeta\nnu0 nu1 nu2 nu3 nu4 h1p c2p chin\nphi psi chip omega chi2 chi3 chi4 chi5\n[<name>] Output data set name.\n<dihedral types> Dihedral types to look for. Note that\nchip is ’protein chi’, chin is ’nucleic chi’.\n[resrange <range/mask>] Residue range to look for\ndihedrals in. Default is all solute residues. If a\nmask expression is given, use residues selected by\nthe mask expression; if any part of a residue is\nselcted it will be used.\n[out <filename>] Output file name.\n[range360] Wrap torsion values from 0.0 to 360.0\n(default is -180.0 to 180.0).\n[dihtype <name>:<a0>:<a1>:<a2>:<a3>[:<offset>]\nSearch for a custom dihedral type called <name>\nusing atom names <a0>, <a1>, <a2>, and <a3>.\nOffset: -2=<a0><a1> in previous res, -1=<a0> in\nprevious res, 0=All <aX> in single res, 1=<a3> in\nnext res, 2=<a2><a3> in next res.\nDataSets Generated:\n<name>[<dihedral type>]:<#> Aspect corresponds to\nthe dihedral type name (e.g. [phi], [psi], etc).\nThe index is the residue number.\nNote data sets are not generated until run is called.\nCalculatespecifieddihedralangletypesforresiduesingivenrange/mask. By\ndefault, dihedral angles are identified based on standard Amber atom names.\nThe resulting data sets will have aspect equal to [<dihedral type>] and index\n171"}, {"page": 172, "text": "equal to residue #. To differentiate the chi angle, chip is used for proteins and\nchinfornucleicacids. Forexample,tocalculateallphi/psidihedralsforresidues\n6 to 9:\nmultidihedral MyTorsions phi psi resrange 6-9 out PhiPsi_6-9.dat\nThis will generate data sets named MyTorsions[phi]:6, MyTorsions[psi]:6, My-\nTorsions[phi]:7, etc. Dihedrals other than those defined in <dihedral types>\ncan be searched for using dihtype. For example to create a custom dihedral\ntypecalledchi1usingatomsN,CA,CB,andCG(allinthesameresidue),then\nsearch for and calculate the dihedral in all residues:\nmultidihedral dihtype chi1:N:CA:CB:CG out custom.dat\n11.55 multipucker\nmultipucker [<name>] [<pucker types>] [out <filename>] [resrange <range>]\n[altona|cremer] [puckertype <name>:<a0>:<a1>:<a2>:<a3>:<a4>[:<a5>] ...]\n[amplitude [ampout <ampfile>]] [theta [thetaout <thetafile>]]\n[range360] [offset <offset>]\n<pucker types> = nucleic furanose pyranose\n[<name>] Output data set name.\n<pucker types> Pucker types to look for.\n[out <filename>] Output file name to write pucker data\nto.\n[resrange <range>] Residue range to look for puckers\nin. Default is all solute residues.\n[puckertype <name>:<a0>:<a1>:<a2>:<a3>:<a4>[:<a5>]\nSearch for a custom pucker type called <name> using\natom names <a0>, <a1>, <a2>, <a3>, and <a4> (also\n<a5> for 6 atom puckers).\n[altona] Use method of Altona & Sundaralingam (5 atoms\nonly). This is the default when pucker has 5 atoms.\n[cremer] Use method of Cremer and Pople (5 or 6 atoms).\nThis is the default when pucker has 6 atoms.\n[amplitude] Also calculate amplitude (in degrees).\nampout <ampfile> File to write amplitude sets to.\n[theta] (Valid for 6 atoms only) Also calculate theta (in\ndegrees).\nthetaout <thetafile> File to write theta sets to.\n172"}, {"page": 173, "text": "[range360] Wrap pucker values from 0.0 to 360.0 (default\nis -180.0 to 180.0).\n[offset <offset>] Add <offset> to pucker values.\nDataSets Generated:\n<name>[<pucker type>]:<#> Aspect corresponds to\nthe pucker type name (e.g. [nucleic], [furanose],\netc). The index is the residue number.\n<name>[<pucker type>Amp]:<#> amplitude only.\nData set for pucker amplitude.\n<name>[<pucker type>Theta]:<#> theta only. Data\nset for pucker theta.\nNote data sets are not generated until run is called.\nCalculate specified pucker types for residues in given range. By default,\npuckers are identified based on standard Amber atom names. The resulting\ndata sets will have aspect equal to [<pucker type>] and index equal to residue\n#. In order to be identified as a pucker, all consecutive atoms in the pucker\nmust be bonded, and the last atom of the pucker must be bonded to the first.\nFor example, to calculate all nucleic acid ribose puckers for residues 6 to 9:\nmultipucker MyPuckers nucleic resrange 6-9 out Pucker_6-9.dat\nThiswillgeneratedatasetsnamedMyPuckers[nucleic]:6,MyPuckers[nucleic]:7,\netc. Puckers other than those defined in <pucker types> can be searched for\nusingpuckertype. Forexampletocreateacustompuckertypecalledfuranoid\nusing atoms C2, C3, C4, C5, and O2, then search for and calculate that pucker\n(with amplitudes) using the method of Cremer and Pople in all residues:\nmultipucker Furanoid puckertype furanoid:C2:C3:C4:C5:O2 cremer \\\nout furanoid.dat amplitude ampout furanoid.dat\n11.56 multivector\nmultivector [<name>] [resrange <range>] name1 <name1> name2 <name2> [out <filename>]\n[ired]\n[<name>] Data set name.\n[resrange <range>] Range of residues to look for\nvectors in.\nname1 <name1> Name of first atom in each residue.\nname2 <name2> Name of second atom in each residue.\n[out <filename>] File to write results to.\n173"}, {"page": 174, "text": "Search for and calculate atomic vectors between atoms named <name1> and\n<name2> in residues specified by the given <range>; each one is equivalent to\nthecommand’vector <name1> <name2>’. Forexample,tocalculateallvectors\nbetween atoms named ’N’ and atoms named ’H’ in residues 5-20, storing the\nresults in data sets named NH and writing to NH.dat:\nmultivector NH name1 N name2 H ired out NH.dat resrange 5-20\n11.57 nastruct\nnastruct [<dataset name>] [resrange <range>] [sscalc] [naout <suffix>]\n[noheader] [resmap <ResName>:{A,C,G,T,U} ...] [calcnohb]\n[noframespaces] [baseref <file>] ...\n[bpmode {3dna|babcock}] [allhb]\n[hbcut <hbcut>] [origincut <origincut>] [altona | cremer]\n[zcut <zcut>] [zanglecut <zanglecut>] [groovecalc {simple | 3dna}]\n[axesout <file> [axesoutarg <arg> ...] [axesparmout <file>]]\n[bpaxesout <file> [bpaxesoutarg <arg> ...] [bpaxesparmout <file>]]\n[stepaxesout <file> [stepaxesoutarg <arg> ...] [stepaxesparmout <file>]]\n[axisnameo <name>] [axisnamex <name>] [axisnamey <name>] [axisnamez <name>]\n[{ first | reference | ref <name> | refindex <#> |\nallframes |\nspecifiedbp pairs <b1>-<b2>,... }]\n[<dataset name>] Output data set name.\n[resrange <range>] Residue range to search for nucleic\nacids in (default all).\n[sscalc] Calculate parameters between consectuive bases\nin strands.\n[naout <suffix>] File name suffix for output files;\nBP.<suffix> for base pair parameters,\nBPstep.<suffix> for base pair step parameters, and\nHelix.<suffix> for base pair step helical\nparameters. If sscalc is specified, also\nSS.<suffix> for parameters of consecutive bases in\nstrands.\n[noheader] Do not print header to naout file.\n[resmap <ResName>:{A,C,G,T,U}] Attempt to treat\nresidues named <ResName> as if it were A, C, G, T,\nor U; useful for residues with modifications or\nnon-standard residue names. This will only work if\nenough reference atoms are present in <ResName>.\n174"}, {"page": 175, "text": "[calcnohb] Calculate parameters between bases in base\npairs even if no hydrogen bonds present between\nthem.\n[noframespaces] If specified there will be no spaces\nbetween frames in the naout files.\n[baseref <file>] Specify a custom nucleic acid base\nreference. One file per custom residue; multiple\n’baseref’ keywords may be present. See below for\ndetails.\n[bpmode {3dna|babcock}] Specify axis conventions for\ncalculating base pair parameters. If ’3dna’\n(default), use conventions of 3DNA[20]; flip Y and Z\nof complimentary base for antiparallel. If\n’babcock’, use conventions of Babcock et al.[21] ;\nflip Y and Z of complimentary base for antiparallel,\nflip X and Y for parallel.\n[allhb] Report the total number of hydrogen bonds\ndetected instead of just the number of\nWatson-Crick-Franklin hydrogen bonds.\n[hbcut <hbcut>] Distance cutoff (in Angstroms) for\ndetermining hydrogen bonds between bases (default\n3.5).\n[origincut <origincut>] Distance cutoff (in Angstroms)\nbetween base pair axis origins for determining which\nbases are eligible for base-pairing (default 2.5).\n[altona] Use method of Altona & Sundaralingam to\ncalculate sugar pucker (default, see pucker\ncommand).\n[cremer] Use method of Cremer and Pople to calculate\nsugar pucker (see pucker command).\n[zcut] Distance cutoff (in Angstroms) between base\nreference axes along the Z axis (i.e. stagger) for\ndetermining base pairing (default 2).\n[zanglecut] Angle cutoff (in degrees) between base\nreference Z axes for determining base pairing\n(default 65).\n[groovecalc] Groove width calculation method:\nsimple Use P-P distance for major groove, O4-O4\ndistance for minor groove. Output to\n’BP.<suffix>’.\n3dna Use groove width calculation of El Hassan and\nCalladine[22]. Output to ’BPstep.<suffix>’.\n175"}, {"page": 176, "text": "[axesout <file>] Trajectory file to write base axes to.\n[axesoutarg <arg>] Trajectory argument to pass to\nbase axes trajectory file (can specify more than\nonce).\n[axesparmout <file>] Topology file to write base\naxes pseudo topology to.\n[bpaxesout <file>] Trajectory file to write base pair\naxes to.\n[bpaxesoutarg <arg>] Trajectory argument to pass\nto base pair axes trajectory file (can specify\nmore than once).\n[bpaxesparmout <file>] Topology file to write base\npair axes pseudo topology to.\n[stepaxesout <file>] Trajectory file to write base pair\nstep axes to.\n[stepaxesoutarg <arg>] Trajectory argument to pass\nto base pair step axes trajectory file (can\nspecify more than once).\n[stepaxesparmout <file>] Topology file to write\nbase pair step axes pseudo topology to.\n[axisnameo <name>] Change name of axis origin pseudo\natom (default ’Orig’).\n[axisnamex <name>] Change name of axis origin pseudo\natom (default ’X’).\n[axisnamey <name>] Change name of axis origin pseudo\natom (default ’Y’).\n[axisnamez <name>] Change name of axis origin pseudo\natom (default ’Z’).\nHow to determine base pairing:\n[first] Use first frame to determine base pairing\n(default).\n[reference | refindex <#> | ref <name>] Reference\nstructure to use to determine base pairing.\n[allframes] If specified determine base pairing each\nframe.\n[specifiedbp pairs <b1>-<b2>,...] User specified base\npairing. Base pairs are specified in a\ncomma-separated list after the ’pairs’ keyword as\n<b1>-<b2>, where <b1> and <b2> are the residue\nnumbers of bases in the base pair, e.g. ’pairs\n1-16,2-15,3-14,4-13’. Can specify ’pairs’ multiple\ntimes.\n176"}, {"page": 177, "text": "DataSets Created:\n<name>[pucker]:X Base X (residue number) sugar\npucker.\nBase pairs:\n<name>[shear]:X Base pair X (starting from 1) shear.\n<name>[stretch]:X Base pair stretch.\n<name>[stagger]:X Base pair stagger.\n<name>[buckle]:X Base pair buckle.\n<name>[prop]:X Base pair propeller.\n<name>[open]:X Base pair opening.\n<name>[hb]:X Number of WC hydrogen bonds between\nbases in base pair.\n<name>[bp]:X Contain 1 if bases are base paired, 0\notherwise.\n<name>[major]:X (If groovecalc simple) Major groove\nwidth calculated between P atoms of each base.\n<name>[minor]:X (If groovecalc simple) Minor groove\nwidth calculated between O4 atoms of each base.\nBase pair steps:\n<name>[shift]:X Base pair step X (starting from 1)\nshift.\n<name>[slide]:X Base pair step slide.\n<name>[rise]:X Base pair step rise.\n<name>[title]:X Base pair step tilt.\n<name>[roll]:X Base pair step roll.\n<name>[twist]:X Base pair step twist.\n<name>[zp]:X Base pair step Zp value.\n<name>[major]:X (If groovecalc 3dna) Major groove\nwidth, El Hassan and Calladine.\n<name>[minor]:X (If groovecalc 3dna) Minor groove\nwidth, El Hassan and Calladine.\nHelical steps:\n<name>[xdisp]:X Helical step X (starting from 1) X\ndisplacement.\n<name>[ydisp]:X Helical Y displacement.\n<name>[hrise]:X Helical rise.\n<name>[incl]:X Helical inclination.\n177"}, {"page": 178, "text": "<name>[tip]:X Helical tip.\n<name>[htwist]:X Helical twist.\nStrands (sscalc only):\n<name>[dx]:X Strand pair X (starting from 1) X\ndisplacement.\n<name>[dy]:X Y displacement.\n<name>[dz]:X Z displacement.\n<name>[rx]:X Relative rotation around X axis.\n<name>[ry]:X Relative rotation around Y axis.\n<name>[rz]:X Relateive rotation around Z axis.\nNote that base pair data sets are not created until base pairing is determined.\nCalculatebasicnucleicacid(NA)structureparametersforallresiduesinthe\nrange specified by resrange (or all NA residues if no range specified). Residue\nnamesarerecognizedwiththefollowingpriority: standardAmberresiduenames\nDA, DG, DC, DT, RA, RG, RC, and RU; 3 letter residue names ADE, GUA,\nCYT,THY,andURA;andfinally1letterresiduenamesA,G,C,T,andU.Non-\nstandard/modified NA bases can be recognized by using the resmap keyword.\nFor example, to make cpptraj recognize all 8-oxoguanine residues named ’8OG’\nas a guanine-based residue:\nnastruct naout nastruct.dat resrange 274-305 resmap 8OG:G\nThe resmap keyword can be specified multiple times, but only one mapping\nper unique residue name is allowed. Note that resmap may fail if the residue\nis missing heavy atoms normally present in the specified base type.\nBasepairsaredeterminedeitheroncefromthefirstframeorfromareference\nstructure, or can be determined each frame if allframes is specified. Base\npairing can also be specified via the specifiedbp and pairs keywords. Base\npairingisdeterminedfirstbybasereferenceaxisorigindistance,thenbystagger,\nthen by angle between base Z axes, then finally by hydrogen bonding (at least\nonehydrogenbondmustbepresent). Basepairparameterswillonlybewritten\nfor determined base pairs. Both Watson-Crick and other types of base pairing\ncan be detected. Note that although all possible hydrogen bonds are searched\nfor, only WC hydrogen bonds are reported in the BP.<suffix> file.\nThe procedure used to calculate NA structural parameters is the same as\n3DNA[20], with algorithms adapted from Babcok et al.[21] and reference frame\ncoordinates from Olson et al.[23]. Given the same base pairs are determined,\ncpptraj nastruct should give the exact same numbers as 3DNA. One notable\nexception are parameters for G-quadruplex structures.\nCalculated NA structure parameters are written to three separate files, the\nsuffix of which is specified by naout. Base pair parameters (shear, stretch,\nstagger, buckle, propeller twist, opening, # WC hydrogen bonds, base pairing,\n178"}, {"page": 179, "text": "and simple groove widths) are written to BP.<suffix>, along with the number\nof WC hydrogen bonds detected. Base pair step parameters (shift, slide, rise,\ntilt, roll, twist, Zp, and El Hassan and Calladine groove widths) are written\nto BPstep.<suffix>, and helical parameters (X-displacement, Y-displacement,\nrise, inclination, tip, and twist) are written to Helix.<suffix>. If noheader is\nspecified a header will not be written to the output files. Note that although\nbase puckering is calculated, it is not written to an output file by default. You\ncan output pucker to a file via the create or write/writedata commands after\nthe data has been generated, e.g.:\nnastruct NA naout nastruct.dat resrange 1-3,28-30\nrun\nwritedata NApucker.dat NA[pucker]\nNote that while the underlying procedure is geared towards calculating param-\neters for base pairs, the code can be made to calculate parameters between\nconsecutive bases in single strands by specifying sscalc.\nBaseaxes,basepairaxes,andbasepairstepaxescanbewrittentotrajectory\nfiles using the axesout, bpaxesout, and stepaxesout and related keywords.\nThe axes are written using 4 points: an origin, and X Y and Z which are\nbonded to the origin. The names of these pseudo atoms can be changed using\nthe axisnameo, axisnamex, axisnamey, and axisnamez keywords.\nCustom Nucleic Acid Base References\nUsers can now specify baseref <file> to load a custom nucleic acid base ref-\nerence. The base reference files are white-space delimited, begin with the line\nNASTRUCT REFERENCE, and have the following format:\nNASTRUCT REFERENCE\n<base character> <res name 0> [<res name 1> ...]\n<atom name> <X> <Y> <Z> <HB type> <RMS fit>\n...\nThere is a line for each reference atom. Lines beginning with ’#’ are ignored as\ncomments.\n<base character> Used to identify the underlying base type: A G C T or\nU. If none of these, it will be considered an unknown residue (which just\nmeans WC hydrogen bonding will not be identified).\n<res name X> Specifies what residue names this reference corresponds to.\nThere must be at least one residue name. There can be any number of\nthese specified.\n<atom name> A reference atom name.\n<X> <Y> <Z> The X Y and Z coordinates of the reference atom.\n179"}, {"page": 180, "text": "<HB type> Denotes if and how the atom participates in hydrogen bonding.\nCan be’d’onor, ’a’cceptor, or ’n’one (or thenumbers1, 2, 0 respectively).\nOnly the first character of the word actually matters.\n<RMS fit> Denotes whether the atom is involved in RMS-fitting.\nHere is an example for GUA:\nNASTRUCT REFERENCE\nG G G5 G3\n# Modified into format readable by cpptraj nastruct\nC1’ -2.477 5.399 0.000 0 0\nN9 -1.289 4.551 0.000 0 1\nC8 0.023 4.962 0.000 0 1\nN7 0.870 3.969 0.000 accept 1\nC5 0.071 2.833 0.000 0 1\nC6 0.424 1.460 0.000 0 1\nO6 1.554 0.955 0.000 accept 0\nN1 -0.700 0.641 0.000 donor 1\nC2 -1.999 1.087 0.000 0 1\nN2 -2.949 0.139 -0.001 donor 0\nN3 -2.342 2.364 0.001 accept 1\nC4 -1.265 3.177 0.000 0 1\n11.58 nativecontacts\nnativecontacts [<mask1> [<mask2>]] [writecontacts <outfile>] [resout <resfile>]\n[noimage] [distance <cut>] [out <filename>] [includesolvent]\n[ first | reference | ref <name> | refindex <#> ]\n[resoffset <n>] [contactpdb <file>] [pdbcut <cut>] [mindist] [maxdist]\n[name <dsname>] [byresidue] [map [mapout <mapfile>]]\n[series [seriesout <file>]]\n[savenonnative [seriesnnout <file>] [nncontactpdb <file>]]\n[resseries { present | sum } [resseriesout <file>]] [skipnative]\n<mask1> First mask to calculate contacts for.\n[<mask2>] (Optional) Second mask to calculate contacts\nfor.\n[writecontacts <outfile>] Write information on native\ncontacts to <outfile> (STDOUT if not specified).\n[resout <resfile>] File to write contact residue pairs\nto.\n[noimage] Do not image distances.\n[distance <cut>] Distance cutoff for determining native\ncontacts in Angstroms (default 7.0 Ang).\n180"}, {"page": 181, "text": "[out <filename>] File to write number of native\ncontacts and non-native contacts.\n[includesolvent] By default solvent molecules are\nignored; this will explicitly include solvent\nmolecules.\n[first | reference | ref <name> | refindex <#>] Reference\nstructure to use for determining native contacts.\n[resoffset <n>] (byresidue only) Ignore contacts between\nresidues spaced less than <n> residues apart in\nsequence.\n[contactpdb <file>] Write PDB with B-factor column\ncontaining relative contact strength for native\ncontacts (strongest is 100.0).\n[pdbcut <cut>] If writing contactpdb, only write\ncontacts with relative contact strength greater than\n<cut>.\n[mindist] If specified, determine the minimum distance\nbetween any atoms in the mask(s).\n[maxdist] If specified, determine the maximum distance\nbetween any atoms in the mask(s).\n[name <dsname>] Data set name.\n[byresidue] Write out the contact map by residue instead\nof by atom.\n[map] Calculate matrices of native contacts\n([nativemap]) and non-native contacts ([nonnatmap]).\nThese matrices are normalized by the total number of\nframes, so that a value of 1.0 means “contact always\npresent”. If byresidue specified, the values for\neach individual atom pair are summed over the\nresidues they belong to (this means for byresidue\nvalues greater than 1.0 are possible).\n[mapout <mapfile>] Write native/non-native matrices to\n’native.<mapfile>’ and ’nonnative.<mapfile>’\nrespectively.\n[series] Calculate native contact time series data, 1 for\ncontact present and 0 otherwise.\n[seriesout <file>] Write native contact time series data\nto file.\n[savenonnative] Save non-native contacts; series must\nalso be specified. This is enabled by default if\nskipnative specified.\n181"}, {"page": 182, "text": "[seriesnnout <file>] Write non-native contact time\nseries data to file.\n[nncontactpdb <file>] Write PDB with B-factor\ncolumn containing relative contact strength for\nnon-native contacts (strongest is 100.0).\n[resseries {present | sum} Create contacts time series by\nresidue; series must also be specified.\npresent Record a 1 if any contact is present and 0\nif no contact is present for the residue pair.\nsum The sum of all individual contacts is recorded\nfor the residue pair.\n[resseriesout <file>] Write residue time series data\nto <file>.\n[skipnative] If specified, skip native contacts\ndetermination, i.e. treat all sonctacts as\nnon-native contacts. Implies savenonnative.\nData Sets Created:\n<dsname>[native] Number of native contacts.\n<dsname>[nonnative] Number of non-native contacts.\n<dsname>[mindist] (mindist only) Minimum observed\ndistance each frame.\n<dsname>[maxdist] (maxdist only) Maximum observed\ndistance each frame.\n<dsname>[nativemap] (map only) Native contacts matrix\n(2D).\n<dsname>[nonnatmap] Non-native contacts matrix (2D).\n<dsname>[NC] Native contacts time series.\n<dsname>[NN] Non-native contacts time series.\n<dsname>[NCRES] Residue native contacts time\nseries.\n<dsname>[NNRES] Residue non-native contacts time\nseries.\nDefine and track “native” contacts as determined by a simple distance cut-off,\ni.e. any atoms which are closer than <cut> in the specified reference frame\n(the first frame if no reference specified) are considered a native contact. If one\nmask is provided, contacts are looked for within <mask1>; if two masks are\nprovided, only contacts between atoms in <mask1> and atoms in <mask2>\nare looked for (useful for determining intermolecular contacts). By default only\nnative contacts are tracked. This can be changed by specifying the savenon-\nnative keyword or by specifying skipnative. The time series for contacts\n182"}, {"page": 183, "text": "can be saved using the series keyword; these can be further consolidated by\nresidue using the resseries keyword. When using <resseries> the data set in-\ndex is calculated as (r2 * nres) + r1 so that indices can be matched between\nnative/non-native contact pairs. Non-native residue contact legends have an\nnn_ prefix.\nNative contacts that are found are written to the file specified by writecon-\ntacts (or STDOUT) with format:\n# Contact Nframes Frac. Avg Stdev\nWhere Contact takes the form ’:<residue1 num>@<atom name>_:<residue2\nnum>@<atomname>, Nframesisthenumberofframesthecontactispresent,\nFrac. is the total fraction of frames the contact is present, Avg is the average\ndistance of the contact when present, and Stdev is the standard deviation of\nthe contact distance when present. If resout is specified the total fraction of\ncontacts is printed for all residue pairs having native contacts with format:\n#Res1 #Res2 TotalFrac Contacts\nWhere #Res1 is the first residue number, #Res2 is the second residue number,\nTotalFrac is the total fraction of contacts for the residue pair, and Contacts\nis the total number of native contacts involved with the residue pair. Since\nTotalFrac is calculated for each pair as the sum of each contact involving that\npair divided by the total number of frames, it is possible to have TotalFrac\nvalues greater than 1 if the residue pair includes more than 1 native contact.\nDuring trajectory processing, non-native contacts (i.e. any pair satisfying\nthe distance cut-off which is not already a native contact) are also searched\nfor. The time series for native contacts can be saved as well, with 1 for contact\npresent and 0 otherwise (similar to the hbond command). This data can be\nsubsequently analyzed using e.g.12.21 on page 257.\nContact maps (matrices) are generated for native and non-native contacts.\nIf byresidueisspecified,contactmapsaresummedoverresidues,andcontacts\nbetween residues spaced <resoffset> residues apart in sequence are ignored.\nIf contactpdb is specified a PDB is generated containing relative contact\nstrengths in the B-factor column. The relative contact strength is normalized\nso that a value of 100 means that atom participated in the most contacts with\nother atoms.\nExample command looking for contacts between residues 210 to 260 and\nresidue named NDP, using reference structure ’FtuFabI.WT.pdb’ to define na-\ntive contacts:\nparm FtuFabI.parm7\ntrajin FtuFabI.nc\nreference FtuFabI.WT.pdb\nnativecontacts name NC1 :210-260&!@H= :NDP&!@H= \\\nbyresidue out nc.all.res.dat mindist maxdist \\\ndistance 3.0 reference map mapout resmap.gnu \\\n183"}, {"page": 184, "text": "contactpdb Loop-NDP.pdb \\\nseries seriesout native.dat\n11.59 outtraj\nouttraj <filename> [ trajout args ]\n[maxmin <dataset> min <min> max <max>] ...\n<filename> Output trajectory file name.\n[trajout args] Output trajectory arguments (see 10.5 on\npage 90).\n[maxmin <dataset> min <min> max <max>] Only write\nframes to <filename> if values in <dataset> for\nthose frames are between <min> and <max>. Can be\nspecified for one or more data sets.\nTheouttrajcommandissimilarinfunctiontotrajout,andtakesallofthesame\narguments. However, instead of writing a trajectory frame after all actions are\ncompleteouttrajwritesthetrajectoryframeatitspositionintheActionqueue.\nFor example, given the input:\ntrajin mdcrd.crd\ntrajout output.crd\nouttraj BeforeRmsd.crd\nrms R1 first :1-20@CA out rmsd.dat\nouttraj AfterRmsd.crd\nthreetrajectorieswillbewritten: output.crd,BeforeRmsd.crd,andAfterRmsd.crd.\nThe output.crd and AfterRmsd.crd trajectories will be identical, but the Befor-\neRmsd.crd trajectory will contain the coordinates of mdcrd.crd before they are\nRMS-fit.\nThe maxmin keyword can be used to restrict output using one more more\ndata sets. For example, to only write frames for which the RMSD value is\nbetween 0.7 and 0.8:\ntrajin tz2.truncoct.nc\nrms R1 first :2-11\nouttraj maxmin.crd maxmin R1 min 0.7 max 0.8\n11.60 pairdist\npairdist out <filename> mask <mask> [delta <resolution>]\nCalculate pair distribution function. In the following, defaults are given in\nparentheses. The out keyword specifies output file for histogram: distance,\nP(r), s(P(r)). The mask option specifies atoms for which distances should be\ncomputed. The delta option specifies resolution. (0.1 Å)\n184"}, {"page": 185, "text": "11.61 pairwise\npairwise [<name>] [<mask>] [out <filename>] [cuteelec <ecut>] [cutevdw <vcut>]\n[ reference | ref <name> | refindex <#> ] [cutout <cut mol2 prefix>]\n[vmapout <vdw map>] [emapout <elec map>] [avgout <avg file>]\n[eout <eout file>] [pdbout <pdb file>] [scalepdbe] [printmode {only|or|and|}]\n[<name>] Data set name; van der Waals energy will get\naspect [EVDW] and electrostatic energy will get\naspect [EELEC].\n[<mask>] Atoms to calculate energy for.\n[out <filename>] File to write total EELEC and EVDW to.\n[eout <eout file>] File to write individual EELEC and\nEVDW interactions to.\n[reference | ref <name> | refindex <#> ] Specify a\nreference to compare frames to (i.e. calculate Eref\n- Eframe).\n[cuteelec <cut>] Only report interaction EELEC (or delta\nEELEC) if absolute value is greater than <ecut>\n(default 1.0 kcal/mol).\n[cutevdw <cutv>] Only report interaction EVDW (or\ndelta EVDW) if absolute value is greater than <vcut>\n(default 1.0 kcal/mol).\n[cutout <cut mol2 prefix>] Write out mol2 containing\nonly atom pairs which satisfy <ecut> and <vcut>.\n[vmapout <vdw map>] Write out interaction EVDW (or\ndelta EVDW) matrix to file <vdw map>.\n[emapout <elec map>] Write out interaction EELEC (or\ndelta EELEC) matrix to file <elec map>.\n[avgout <avg file>] Print average interaction EVDW|EELEC\n(or average delta EVDW|EELC) to <avg file>.\n[pdbout <pdb file>] Write PDB with EVDW|EELEC in\noccupancy|B-factor columns to <pdb file>.\n[scalepdbe] Scale energies written to PDB from 0 to 100.\n[printmode {only|or|and}] Control when/how average\nenergies are written\nData Sets Created:\n<name>[EELEC] Electrostatic energy in (kcal/mol).\n<name>[EVDW] van der Waals energy in (kcal/mol).\n<name>[VMAP] van der Waals energy matrix.\n185"}, {"page": 186, "text": "<name>[EMAP] Electrostatic energy matrix.\nThis action has two related functions: 1) Calculate pairwise (i.e. non-bonded)\nenergy (in kcal/mol) for atoms in <mask>, or 2) Compare pairwise energy of\nframes to a reference frame. This calculation does use an exclusion list but is\nnot periodic.\nWhen comparing to a reference frame, the eout file will contain the differ-\nencesforeachindividualinteraction(i.e. Eref-Eframe),otherwisetheeoutfile\nwill contain the absolute value of each individual interaction. The cuteelc and\ncutevdw keywords can be used to restrict printing of individual interactions\nto those for which the absolute value is above a cutoff. The VMAP and EMAP\nmatrixelementswillcontainthesevaluesaswell(differencesforreference,abso-\nlutevalueotherwise)averagedoverallframes. Theavgoutfilewillcontainonly\nthese values averaged over all frames that satisfy the cutoffs. The printmode\nkeywordcontrolswhentheaverageenergiesarewritten: only meansonlyaver-\nage energy components that satisfy cutoffs will be printed, or means that both\nenergy components will be printed if either satistfy a cutoff, and and means\nthat both energy components will be written only if both satisfy the cutoffs.\nThecutoutkeywordcanbeusedtowriteoutMOL2fileseachframenamed\n’<cutmol2prefix>.evdw.mol2.X’and’<cutmol2prefix>.eelec.mol2.X’(where\nX is the frame number) containing only atoms with energies that satisfy the\ncutoffs. Similarly, the pdbout keyword can be used to write out a PDB file\n(with 1 MODEL per frame). The occupancy and B-factor columns will contain\nthe total van der Waals and electrostatic energy for each atom if cutoffs are\nsatisfied, or 0.0 otherwise.\n11.62 principal\nprincipal [<mask>] [dorotation] [out <filename>] [name <dsname>]\n[<mask>] Mask of atoms used to determine principal\naxes (default all).\n[dorotation] Align coordinates along principal axes.\n[out <filename>] Write resulting\neigenvalues/eigenvectors to <filename>.\n[name <dsname>] Data set name (3x3 matrices).\nData Sets Created (name keyword only):\n<dsname>[evec] Eigenvectors (3x3 matrix, row-major).\n<dsname>[eval] Eigenvalues (vector).\nDetermine principal axes of each frame determined by diagonalization of the\ninertial matrix from the coordinates of the specified atoms. At least one of\ndorotation, out, or name must be specified. The resulting eigenvectors are\nsorted from largest eigenvalue to smallest, and the corresponding axes labelled\n186"}, {"page": 187, "text": "using the cpptraj convention of X > Y > Z (similar to ’vector principal’). If\nout is specified the eigenvectors and eigenvalues will be written for each frame\nN with format:\n<N> EIGENVALUES: <EX> <EY> <EZ>\n<N> EIGENVECTOR 0: <Xx> <Xy> <Xz>\n<N> EIGENVECTOR 1: <Yx> <Yy> <Yz>\n<N> EIGENVECTOR 2: <Zx> <Zy> <Zz>\nNOTE: The eigenvector 3x3 matrix data set could subsequently be used e.g.\nwith the rotate action.\nExample: Align system (residues 1-76) along principle axes:\nparm myparm.parm7\ntrajin protein.nc\nprincipal :1-76 dorotation out principal.dat\n11.63 projection\nprojection [<name>] evecs <dataset name> [out <outfile>] [beg <beg>] [end <end>]\n{[<mask>] | [dihedrals <dataset arg>] | [data <dataset arg> ...]}\n[start <start>] [stop <stop>] [offset <offset>]\n[<name>] Output data set name.\nevecs <dataset name> Data set containing eigenvectors\n(modes).\n[out <outfile>] Write projections to <outfile>.\n[beg <beg>] First eigenvector/mode to use (default 1).\n[end <end>] Final eigenvector/mode to use (default 2).\n[<mask>] (Not dihedral covariance) Mask of atoms to\nuse in projection; MUST CORRESPOND TO HOW\nEIGENVECTORS WERE GENERATED.\n[dihedrals <dataset arg>] (Dihedral covariance only)\nDihedral data sets to use in projection; MUST\nCORRESPOND TO HOW EIGENVECTORS WERE GENERATED.\n[data <dataset arg>] (Data covariance only, e.g. from\nTICA). 1D data sets to use in projection; MUST\nCORRESPOND TO HOW EIGENVECTORS WERE GENERATED.\n[start <start>] Frame to start calculating projection.\n[stop <stop>] Frame to stop calculating projection.\n[offset <offset>] Frames to skip between projection\ncalculations.\n187"}, {"page": 188, "text": "Data Sets Created:\nDataSet indices correspond to mode #.\n<name> (All execpt IDEA) Projection data set.\n<name>[X] X component of mode (IDEA modes only).\n<name>[Y] Y component of mode (IDEA modes only).\n<name>[Z] Z component of mode (IDEA modes only).\n<name>[R] Magnitude of mode (IDEA modes only).\nProjects snapshots onto eigenvectors obtained by diagonalizing covariance or\nmass-weightedcovariancematrices. Eigenvectorsaretakenfrompreviouslygen-\nerated(e.g. withdiagmatrix/tica)orpreviouslyread-in(e.g. withreaddata)\neigenvectors with name <dataset name>. The user has to make sure that the\natoms selected by <mask> agree with the ones used to calculate the modes\n(i.e.,ifmask=’@CA’wasusedinthe“matrix” command,mask=’@CA’needs\nto be set here as well). See 13 on page 286 for examples using the projection\ncommand. If only 1D data sets need to be projected, see the projectdata\nanalysis command, 12.28 on page 266.\n11.64 pucker\npucker [<name>] <mask1> <mask2> <mask3> <mask4> <mask5> [<mask6>] [geom]\n[out <filename>] [altona | cremer] [amplitude] [theta]\n[range360] [offset <offset>]\n<name> Output data set name.\n<maskX> Five (optionally six) atom masks selecting\natom(s) to calculate pucker for.\n[geom] Use geometric center of atoms in <maskX> (default\nis center of mass).\n[out <filename>] Output file name.\n[altona] Use method of Altona & Sundaralingam (5 masks\nonly).\n[cremer] Use method of Cremer and Pople (5 or 6 masks).\nThis is the default when 6 masks are specified.\n[amplitude] Also calculate amplitude.\n[theta] (6 masks only) Also calculate theta.\n[range360] Wrap pucker values from 0.0 to 360.0 (default\nis -180.0 to 180.0).\n[offset <offset>] Add <offset> to pucker values.\nData Sets Created:\n188"}, {"page": 189, "text": "<name> Pucker in degrees.\n<name>[Amp] Amplitude (if amplitude was specified).\n<name>[Theta] Theta (if theta and 6 masks were\nspecified).\nCalculatethepucker(indegrees)foratomsin<mask1>,<mask2>,<mask3>,\n<mask4>, <mask5> using the method of Altona & Sundarlingam[24, 25] (de-\nfault for 5 masks, or if altona specified), or the method of Cremer & Pople[26]\n(default for 6 masks, or if cremer is specified). If the amplitude or theta\nkeywords are given, amplitudes/thetas (also in degrees) will be calculated in\naddition to pucker. The results from pucker can be further analyzed with the\nstatistics analysis.\nBy default, pucker values are wrapped to range from -180 to 180 degrees. If\nthe range360 keyword is specified values will be wrapped to range from 0 to\n360 degrees. Note that the Cremer & Pople convention is offset from Altona\n& Sundarlingam convention (with nucleic acids) by +90.0 degrees; the offset\nkeyword will add an offset to the final value and so can be used to convert\nbetweenthetwo. Forexample,toconvertfromCremertoAltonaspecify“offset\n90”.\nTo calculate nucleic acid pucker specify C1’ first, followed by C2’, C3’, C4’\nand O4’. For example, to calculate the sugar pucker for nucleic acid residues 1\nand 2 using the method of Altona & Sundarlingam, with final pseudorotation\nvalues ranging from 0 to 360:\npucker p1 :1@C1’ :1@C2’ :1@C3’ :1@C4’ :1@O4’ range360 out pucker.dat\npucker p2 :2@C1’ :2@C2’ :2@C3’ :2@C4’ :2@O4’ range360 out pucker.dat\n11.65 radgyr | rog\nradgyr [name>] [<mask>] [out <filename>] [mass] [nomax] [tensor]\n[<name>] Data set name.\n[<mask>] Atoms to calculate radius of gyration for;\ndefault all atoms.\n[out <filename>] Write data to <filename>.\n[mass] Mass-weight radius of gyration.\n[nomax] Do not calculate maximum radius of gyration.\n[tensor] Calculate radius of gyration tensor, output\nformat ’XX YY ZZ XY XZ YZ’.\nData Sets Created:\n<name> Radius of gyration in Ang.\n<name>[Max] Max radius of gyration in Ang.\n189"}, {"page": 190, "text": "<name>[Tensor] Radius of gyration tensor; format ’XX\nYY ZZ XY XZ YZ’.\nCalculate the radius of gyration of specified atoms. For example, to calcu-\nlate only the mass-weighted radius of gyration (not the maximum) of the non-\nhydrogen atoms of residues 4 to 10 and print the results to “RoG.dat”:\nradgyr :4-10&!(@H=) out RoG.dat mass nomax\n11.66 radial | rdf\nradial [out <outfilename>] <spacing> <maximum> <solvent mask1> [<solute mask2>]\n[noimage] [mass]\n[density <density> | volume] [<dataset name>] [intrdf <file>] [rawrdf <file>]\n[{{center1|center2|nointramol|toxyz <x>,<y>,<z>} |\n[byres1] [byres2] [bymol1] [bymol2]}]\n[out <outfilename>] File to write RDF to.\n<spacing> Bin spacing, required.\n<maximum> Max bin value, required.\n<solvent mask1> Atoms to calculate RDF for, required.\n[<solute mask2>] (Optional) If specified calculate RDF\nof all atoms in <solvent mask1> to each atom in\n<solute mask2>.\n[noimage] Do not image distances.\n[mass] Use center of mass for centerX/byresX/bymolX\nkeywords, otherwise use geometric center.\n[density <density>] Use density value of <density> for\nnormalization (default 0.033456 molecules Å−3).\n[volume] Determine density for normalization from\naverage volume of input frames.\n[<dataset name>] Name of output data sets.\n[intrdf <file>] Calculate integral of RDF bin values\n(averaged over # of frames but otherwise not\nnormalized) and write to <file> (can be same as\n<output_filename>).\n[rawrdf <file>] Write raw (non-normalized) RDF values to\n<file>.\n[center1] Calculate RDF from center of atoms in <solvent\nmask1> to all atoms in <solute mask2>.\n[center2] Calculate RDF from center of atoms in <solute\nmask2> to all atoms in <solvent mask1>.\n190"}, {"page": 191, "text": "[nointramol] Ignore intra-molecular distances.\n[toxyz <x>,<y>,<z>] Calculate RDF from atoms selected\nby <solvent mask1> to point specified by <x> <y> and\n<z> (in Ang.).\n[byres1] Calculate using the centers of each residue in\nthe first mask.\n[bymol1] Calculate using the centers of each molecule in\nthe first mask.\n[byres2] Calculate using the centers of each residue in\nthe second mask.\n[bymol2] Calculate using the centers of each molecule in\nthe second mask.\nDataSet Aspects:\n<setname> The radial distribution function.\n<setname>[int] (intrdf only) Integral of RDF bin\nvalues.\n<setname>[raw] (rawrdf only) Raw (non-normalized) RDF\nvalues.\nCalculatetheradialdistributionfunction(RDF,akapaircorrelationfunction)of\natoms in <solvent mask1> (note that this mask does not need to be solvent,\nbutthisnomenclatureisusedforclarity). Ifanoptionalsecondmask(<solute\nmask2>) is given, calculate the RDF of ALL atoms in <solvent mask1> to\nEACH atom in <solute mask2>. If desired, the geometric center of atoms\nin <solvent mask1> or <solute mask2> can be used by specifying the\ncenter1orcenter2keywordsrespectively, oralternativelyintra-moleculardis-\ntancescanbeignoredbyspecifyingthenointramolkeyword. If byresXorby-\nmolX are specified, distances will be between the centers of residues/molecules\nselected by <solvent mask1> or <solute mask2>. If mass is specified use\nthe centers of mass for centerX/byresX/bymolX, otherwise use geometric\ncenter.\nThe RDF is calculated from the histogram of the number of particles found\nas a function of distance R, normalized by the expected number of particles at\nthat distance. The normalization is calculated from:\n(cid:16) (cid:17)\nDensity∗ 4π (R+dR)3−R3\n3\nwhere dR is equal to the bin spacing. Some care is required by the user in\norder to normalize the RDF correctly. The default density value is 0.033456\nmolecules Å−3, which corresponds to a density of water approximately equal to\n1.0 g mL−1. To convert a standard density in g mL−1, multiply the density by\n0.6022,whereM isthemassofthemoleculeinatomicmassunits. Alternatively,\nif M thre volume k r eyword is specified the density is determined from the average\nvolume of the system over all Frames.\n191"}, {"page": 192, "text": "NotethatcorrectnormalizationoftheRDFdependsonthenumberofatoms\nineachmask;ifmultipletopologyfilesarebeingprocessedthatresultinchanges\nin the number of atoms in each mask, the normalization will be off.\nThebasic(i.e. nocenter1/center2/byres1/byres2/bymol1/bymol2)RDFcal-\nculationsarenowCUDAparallelized. However,thecalculationisdoneinsingle-\nprecisiononGPUssotheresultinghistogramsmaydifferslightlyfromtheCPU\n(on the order of 0.0002 - 0.0004).\n11.67 randomizeions\nrandomizeions <mask> [around <ardoundmask> by <distance>] [{allowoverlap|overlap <value>}]\n[noimage] [seed <value>] [originalalgorithm]\n<mask> Mask of ions to randomize.\naround <mask> by <distance> Ensure ions come no\ncloser than <distance> Ang. to atoms in\n<aroundmask>.\nallowoverlap No restrictions on how close ions can be to\neach other.\noverlap <value> Ions in <mask> can be no closer than\n<value> Ang. to each other.\n[noimage] Do not image distances.\n[seed <value>] Seed for the random number generator.\n[originalalgorithm] Use the original, slower algorithm\n(from versions before 5.1.0).\nThiscanbeusedtorandomlyswapthepositionsofsolventandsingleatomions.\nThe “overlap” specifies the minimum distance between ions, and the “around”\nkeywordcanbeusedtospecifyasolute(orsetofatoms)aroundwhichtheions\ncangetnocloserthanthedistancespecified. Theoptionalkeywords“noimage”\ndisableimagingand“seed” updatetherandomnumberseed. Anexampleusage\nis\nrandomizeions @NA around :1-20 by 5.0 overlap 3.0\nThe above will swap Na+ ions with water getting no closer than 5.0 Å from\nresidues 1 – 20 and no closer than 3.0 Å from any other Na+ ion.\n11.68 remap\nremap data <setname>\n[outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\n192"}, {"page": 193, "text": "data <setname> Data set to use for remapping; should\nbe a 1D integer data set with X= reference (old)\natom index, Y = target (new) atom index.\noutprefix <prefix> Write remapped topology to\n<prefix>.<originalname>\n[nobox] Remove any box information from the remapped\ntopology.\nparmout <filename> Write remapped topology to\n<filename>\nparmopts <list> Options for writing topology file\nRe-map atoms according to the given reference data set which is of the format:\nReference[Target]\nwith atom numbering starting from 1. E.g. Reference[1] = 10 would mean\nremap atom 10 in target to position 1.\n11.69 replicatecell\nreplicatecell [out <traj filename>] [name <dsname>]\n{ all | dir <XYZ> [dir <XYZ> ...] } [<mask>]\n[outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\nout <traj filename> Write replicated cell to output\ntrajectory file.\nname <dsname> If specified save replicated cell to\nCOORDS data set.\nall Replicate cell once in all possible directions.\ndir <XYZ> Repicate cell once in specified directions.\n<XYZ> should consist of 3 numbers with no spaces in\nbetween them and are restricted to values of -1, 1,\nand 0. May be specified more than once.\n<mask> Mask of atoms to replicate.\noutprefix <prefix> Write replicated topology to\n<prefix>.<originalname>\n[nobox] Remove any box information from the replicated\ntopology.\nparmout <filename> Write replicated topology to\n<filename>\nparmopts <list> Options for writing topology file\n193"}, {"page": 194, "text": "Create a trajectory where the unit cell is replicated in 1 or more directions\n(up to 27). The resulting coordinates and topology can be written to a trajec-\ntory/topology file. They can also be saved as a COORDS data set for subse-\nquent processing. Currently replication is only allowed 1 axis length in either\ndirection. The all keyword will replicate the cell once in all directions. The\ndir keyword can be used to restrict replication to specific directions, e.g. ’dir\n10-1’ would replicate the cell once in the +X, -Z directions.\nFor example, to replicate a cell in all directions, writing out to NetCDF\ntrajectory cell.nc:\nparm ../tz2.truncoct.parm7\ntrajin ../tz2.truncoct.nc\nreplicatecell out cell.nc parmout cell.parm7 all\n11.70 rms | rmsd\nrmsd [<name>] <mask> [<refmask>] [out <filename>] [mass]\n[nofit | norotate | nomod]\n[savematrices [matricesout <file>]]\n[savevectors {combined|separate} [vecsout <file>]]\n[ first | reference | ref <name> | refindex <#> | previous |\nreftraj <name> [parm <name> | parmindex <#>] ]\n[perres perresout <filename> [perresavg <avgfile>]\n[range <resRange>] [refrange <refRange>]\n[perresmask <additional mask>] [perrescenter] [perresinvert]\n[<name>] Output data set name.\n[<mask>] Mask of atoms to calculate RMSD for; if not\nspecified, calculate for all atoms.\n[<refmask>] Reference mask; if not specified, use\n<mask>.\n[out <filename>] Output data file name.\n[mass] Mass-weight the RMSD calculation.\n[nofit] Do not perform best-fit RMSD.\n[norotate] If calculating best-fit RMSD, translate but\ndo not rotate coordinates.\n[nomod] If calculating best-fit RMSD, do not modify\ncoordinates.\n[savematrices] If specified save rotation matrices to\ndata set with aspect [RM].\nmatricesout <file> Write rotation matrices to\nspecified file.\n194"}, {"page": 195, "text": "[savevectors {combined|separate}] If specified save\ntranslation vectors: combined means save\ntarget-to-origin plus the origin-to-reference\ntranslation vectors, separate means save\ntarget-to-origin as Vx, Vy, Vz and save\norigin-to-reference as Ox Oy Oz in the output vector\ndata set.\nvecsout <file> Output translation vector data set\nto <file>.\nReference keywords:\nfirst Use the first trajectory frame processed as\nreference.\nreference Use the first previously read in reference\nstructure (refindex 0).\nref <name> Use previously read in reference structure\nspecified by filename/tag.\nrefindex <#> Use previously read in reference\nstructure specified by <#> (based on order read in).\nprevious Use frame prior to current frame as reference.\nreftraj <name> Use frames from COORDS set <name> or\nread in from trajectory file <name> as references.\nEach frame from <name> is used in turn, so that\nframe 1 is compared to frame 1 from <name>, frame 2\nis compared to frame 2 from <name> and so on. If\n<trajname> runs out of frames before processing is\ncomplete, the last frame of <trajname> continues to\nbe used as the reference.\nparm <parmname> | parmindex <#> If reftraj\nspecifies a trajectory file, associate it with\nspecified topology; if not specified the first\ntopology is used.\nPer-residue RMSD keywords:\nperres Activate per-residue no-fit RMSD calculation.\nperresout <perresfile> Write per-residue RMSD to\n<perresfile>.\nperresavg <avgfile> Write average per-residue RMSDs to\n<avgfile>.\nrange <res range> Calculate per-residue RMSDs for\nresidues in <res range> (default all solute\nresidues).\n195"}, {"page": 196, "text": "refrange <ref range> Calculate per-residue RMSDs to\nreference residues in <ref range> (use <res range>\nif not specified).\nperresmask <additional mask> By default residues are\nselected using the mask ’:X’ where X is residue\nnumber; this appends <additional mask> to the mask\nexpression.\nperrescenter Translate residues to a common center of\nmass prior to calculating RMSD.\nperresinvert Make X-axis residue number instead of frame\nnumber.\nData Sets Created:\n<name> RMSD of atoms in mask to reference.\n<name>[RM] (savematrices only) Rotation matrices of\ntarget to reference.\n<name>[TV] (savevectors only) Translation vector.\n<name>[res] (perres only) Per-residue RMSDs; index is\nresidue number.\n<name>[Avg] (perres only) Average per-residue RMSD\nfor each residue.\n<name>[Stdev] (perres only) Standard deviation of RMSD\nfor each residue.\nNote that perres data sets are not generated until run is called.\nCalculate the coordinate RMSD of input frames to a reference frame (or\nreference trajectory). Both <mask> and <refmask> must specify the same\nnumber of atoms, otherwise an error will occur.\nFor example, say you have a trajectory and you want to calculate RMSD to\ntwoseparatereferencestructures. Tocalculatethebest-fitRMSDoftheC,CA,\nand N atoms of residues 1 to 20 in each frame to the C, CA, and N atoms of\nresidues3to23inStructX.crd,andthencalculatetheno-fitRMSDofresidue7\nto residue 7 in another structure named Struct-begin.rst7, writing both results\nto Grace-format file “rmsd1.agr”:\nreference StructX.crd [structX]\nreference md_begin.rst7 [struct0]\nrmsd BB :1-20@C,CA,N ref [structX] :3-23@C,CA,N out rmsd1.agr\nrmsd Res7 :7 ref [struct0] out rmsd1.agr nofit\nPer-residue RMSD calculation\nIf the perres keyword is specified, after the initial RMSD calculation the no-fit\nRMSD of specified residues is also calculated. So for example:\n196"}, {"page": 197, "text": "rmsd :10-260 reference perres perresout PRMS.dat range 190-211 perresmask &!(@H=)\nwill first perform a best-fit RMSD calculation to the first specified reference\nstructure using residues 10 to 260, then calculate the no-fit RMSD of residues\n190 to 211 (excluding any hydrogen atoms), writing the results to PRMS.dat.\nTwo additional recommendations for the ’perres’ option: 1) try not including\nbackboneatomsbyusingthe’perresmask’keyword,e.g. \"perresmask&!@H,N,CA,HA,C,O\",\nand 2) try using the ’perrescenter’ keyword, which centers each residue prior to\nthe’nofit’calculation;thisisusefulforisolatingchangesinresidueconformation.\n11.71 rms2d | 2drms\nAlthough the ’rms2d’ command can still be specified as an action, it is now\nconsidered an analysis. See 12.31 on page 268.\n11.72 rmsavgcorr\nAlthough the ’rmsavgcorr’ command can still be specified as an action, it is\nnow considered an analysis. See 12.32 on page 269.\n11.73 rmsf | atomicfluct\nSee 11.6 on page 104.\n11.74 rotate\nrotate [<mask>] { [x <xdeg>] [y <ydeg>] [z <zdeg>] |\naxis0 <mask0> axis1 <mask1> <deg> |\nusedata <set name> [inverse] |\ncalcfrom <set name> [name <output set name>] [out <file>]\n}\n[<mask>] Rotate atoms in <mask> (default all).\n[x <xdeg] Degrees to rotate around the X axis.\n[y <xdeg] Degrees to rotate around the Y axis.\n[z <xdeg] Degrees to rotate around the Z axis.\naxis0 <mask0> Mask defining the beginning of a\nuser-defined axis.\naxis1 <mask1> Mask defining the end of a\nuser-defined axis.\n<deg> Value in degrees to rotate around user\ndefined axis.\nusedata <set name> If specified, use 3x3 rotation\nmatrices in specified data set to rotate\ncoordinates.\n197"}, {"page": 198, "text": "[inverse] Perform inverse rotation from input\nrotation matrices.\ncalcfrom <set name> Instead of rotating coordinates,\ncalculate rotations around the X Y and Z axes (as\nwell as total rotation) in degrees from existing\nrotation matrices specified by <set name>.\n[name <output set name>] Output set name.\n[out <file>} File to write output sets to.\nDataSets Created:\n<output set name>[TX] (caclfrom only) Rotation around\nthe X axis in degrees.\n<output set name>[TY] (caclfrom only) Rotation around\nthe Y axis in degrees.\n<output set name>[TZ] (caclfrom only) Rotation around\nthe Z axis in degrees.\n<output set name>[T] (caclfrom only) Total rotation in\ndegrees.\nRotatespecifiedatomsaroundtheX,Y,and/orZaxesbythespecifiedamounts,\naround a user-defined axis (specified by <mask0> and <mask1>), or use a\npreviously read in or generated data set of 3x3 matrices to perform rotations.\nFor example, to rotate the entire system 90 degrees around the X axis:\nrotate x 90\nTorotateresidue27090degreesaroundtheaxisdefinedbetweenatomsC1,C2,\nC3, C4, C5, and C6 in residue 270 and atoms C7, C8, C9, C10, C11, and C12\nin residue 270:\nrotate :270 axis0 :270@C1,C2,C3,C4,C5,C6 axis1 :270@C7,C8,C9,C10,C11,C12 90.0\nTo rotate the system with rotation matrices read in from rmatrices.dat:\ntrajin tz2.norotate.crd\nreaddata rmatrices.dat name RM mat3x3\nrotate usedata RM\nTo calculate rotations from rotation matrices generated by a previous RMSD\ncalculation:\nparm ../tz2.parm7\nreference tz2.separate.rotate.rst7.save name REF\ntrajin ../tz2.nc\nrms R0 reference savematrices matricesout matrices.dat\nrotate calcfrom R0[RM] name Rot out rotations.dat\n198"}, {"page": 199, "text": "11.75 rotdif\nThe’rotdif’ commandisnowananalysis(see12.33onpage271),andrequires\nthat rotation matrices be generated via an rmsd action. For example:\nreference avgstruct.pdb\ntrajin tz2.nc\nrms R0 reference @CA,C,N,O savematrices\nrotdif rmatrix R0[RM] rseed 1 nvecs 10 dt 0.002 tf 0.190 \\\nitmax 500 tol 0.000001 d0 0.03 order 2 rvecout rvecs.dat \\\nrmout matrices.dat deffout deffs.dat outfile rotdif.out\n11.76 runavg | runningaverage\nrunavg [window <window_size>]\nNote that for backwards compatibility with ptraj “runningaverage” is also ac-\ncepted.\nReplaces the current frame with a running average over a number of frames\nspecified by window <window_size> (5 if not specified). This means that in\nordertobuildupthecorrectnumberofframestocalculatetheaverage,thefirst\n<window_size>minusoneframeswillnotbeprocessedbysubsequentactions.\nSo for example given the input:\nrunavg window 3\nrms first out rmsd.dat\nthe rms command will not take effect until frame 3 since that is the first time\n3 frames are available for averaging (1, 2, and 3). The next frame processed\nwould be an average of frames 2, 3, and 4, etc.\n11.77 scale\nscale x <sx> y <sy> z <sz> <mask>\nScale the X|Y|Z coordinates of atoms in <mask> by <sx>|<sy>|<sz>.\n11.78 secstruct\nsecstruct [<name>] [out <filename>] [<mask>] [sumout <filename>]\n[assignout <filename>] [totalout <filename> [ptrajformat]\n[betadetail]\n[namen <N name>] [nameh <H name>] [nameca <CA name>]\n[namec <C name>] [nameo <O name>] [namesg <sulfur name>]\n[<name>] Output data set name.\n199"}, {"page": 200, "text": "[out <filename>] Output file name for secondary\nstructure vs time.\n[<mask>] Atom mask in which residues should be looked\nfor.\n[sumout <sumfilename>] Write average secondary\nstructure values for each residue to <sumfilename>;\nif not specified <filename>.sum is used.\n[assignout <filename>] Write overall secondary structure\nassignment (based on dominant secondary structure\ntype for each residue) to file.\n[ptrajformat] Write secondary structure as a string of\ncharacters for each frame, similar to ptraj output.\n[betadetail] Record anti-parallel beta and parallel beta\nin place of extended and bridge secondary structure.\nIf a residue could be both only anti-parallel is\nreported.\n[namen <N name>] Backbone amide nitrogen atom name\n(default ’N’).\n[nameh <H name>] Backbone amide hydrogen atom name\n(default ’H’).\n[nameca <CA name>] Backbone alpha carbon atom name\n(default ’CA’).\n[namec <C name>] Backbone carbonyl carbon atom name\n(default ’C’).\n[nameo <O name>] Backbone carbonyl oxygen atom name\n(default ’O’).\n[namesg <SG name>] Cysteine sulfur atom name, used to\nignore disulfide connectivity (default ’SG’).\nData Sets Created:\n<name>[res] Residue secondary structure per frame;\nindex corresponds to residue number. If ptrajformat\nspecified these will be characters, otherwise\nintegers (see table below).\n<name>[avgss] Average of each type of secondary\nstructure; index corresponds to secondary structure\ntype (see table below; no index for “None”).\n<name>[None] Total fraction of residues with no\nstructure vs time.\n<name>[Para] Total fraction of residues with parallel\nbeta structure vs time.\n200"}, {"page": 201, "text": "<name>[Anti] Total fraction of residues with\nanti-parallel beta structure vs time.\n<name>[3-10] Total fraction of 3-10 helical structure\nvs time.\n<name>[Alpha] Total fraction of alpha helical\nstructure vs time.\n<name>[Pi] Total fraction of Pi helical structure vs\ntime.\n<name>[Turn] Total fraction of turn structure vs\ntime.\n<name>[Bend] Total fraction of bend structure vs\ntime.\nAs of version 4.18.0, this command now produces output that better\nconforms with the original definitions in Kabsch and Sander 1983;\nnamely that Extended beta (i.e. 2 or more consecutive beta bridges\nof the same type) and beta Bridge (i.e. an isolated beta bridge) are\nnow reported instead of anti-parallel and parallel beta. To restore\nthe original behavior the ’betadetail’ keyword must be specified.\nNote that the residue and [avgss] data sets are not generated until run is\ncalled.\nCalculate secondary structural propensities for residues in <mask> (or\nall solute residues if no mask given) using the DSSP method of Kabsch and\nSander[27], which assigns secondary structure types for residues based on back-\nbone amide (N-H) and carbonyl (C=O) atom positions. By default cpptraj\nassumes these atoms are named “N”, “H”, “C”, and “O” respectively. If a differ-\nentnamingschemeisused(e.g. amidehydrogensarenamed“HN”)thebackbone\natom names can be customized with the nameX keywords (e.g. ’nameH HN’).\nNote that it is expected that some residues will not have all of these atoms\n(such as proline); in this case cpptraj will print an informational message but\nthe calculation will proceed normally. If a residue has no atoms selected it will\nbe skipped. When determining residue connecivity, disulfide bonds will be ig-\nnored; cpptraj identifies such bonds based on the namesg atom name (default\n“SG”).\nResults will be written to filename specified by out with format:\n<#Frame> <ResX SS> <ResX+1 SS> ... <ResN SS>\nwhere <#Frame> is the frame number and <ResX SS> is an integer repre-\nsenting the calculated secondary structure type for residue X. If the keyword\nptrajformat is specified, the output format will instead be:\n<#Frame> STRING\nwhereSTRINGisastringofcharacters(oneforeachresidue)whereeachcharac-\nterrepresentsadifferentstructuraltype(thisformatissimilartowhatptraj had\n201"}, {"page": 202, "text": "outputed and is retained for backwards compatibility). The various secondary\nstructure types and their corresponding integer/character are listed below. If\n’betadetail’isspecifiedwhatisreportedandthecharactersusedchangeslightly.\nSTRING (betadetail) Integer DSSP SS type (betadetail)\n0 0 ’ ’ None\nE (b) 1 ’E’ Extended beta (parallel beta)\nB 2 ’B’ Isolated beta (anti-parallel beta)\nG 3 ’G’ 3-10 helix\nH 4 ’H’ Alpha helix\nI 5 ’I’ Pi (3-14) helix\nT 6 ’T’ Turn\nS 7 ’S’ Bend\nAverage structural propensities over all frames for each residue will be writ-\nten to the file specified by sumout (or “<filename>.sum” if sumout is not\nspecified). The total structural propensity over all residues for each secondary\nstructure type will be written to the file specified by totalout. If assignout\nis specified, the overall secondary structure assignment for each residue will be\nprintedintwolinechunksof50residues,withthefirstlinecontainingtheresidue\nnumberthelinestartswithandonecharacterresiduenames,andthesecondline\ncontainingsecondarystructureassignmentusingDSSP-stylecharacters,likeso:\n1 KCNTATCATQ RLANFLVHSS NNFGAILSST NVGSNTRn\nSSS TH HHHTTSEEEE TTTEEEE SS S\nTheoutputofsecstructcommandisamenabletovisualizationwithgnuplot. To\ngenerate a 2D map-style plot of secondary structure vs time, with each residue\non the Y axis simply give the output file a “.gnu” extension. For example,\nto generate a 2D map of secondary structure vs time, with different colors\nrepresenting different secondary structure types for residues 1-22:\nsecstruct :1-22 out dssp.gnu\nThe resulting file can be visualized with gnuplot:\ngnuplot dssp.gnu\nSimilarly, the sumout file can be nicely visualized using xmgrace (use “.agr”\nextension).\nsecstruct :1-22 out dssp.gnu sumout dssp.agr\nxmgrace dssp.agr\nC <X> <Y> <Z> <Density>\nValuesof dgbulkanddhbulkfordifferentwatermodelscanbecalculatedfrom\npure water simulations with the purewater keyword.\n202"}, {"page": 203, "text": "11.79 setvelocity\nsetvelocity [<mask>]\n[{ tempi <temperature> |\nscale [factor <fac>] [sx <xfac>] [sy <yfac>] [sz <zfac>] |\nadd [value <val>] [vx <xval>] [vy <yval>] [vz <zval>] |\nnone |\nmodify}]\n[[ntc <#>]] [[dt <time>] [epsilon <eps>]]\n[zeromomentum] [ig <random seed>]\n<mask> Mask of atoms to assign velocities to.\ntempi <temperature> Assign velocities at specified\ntemperature (default 300.0 K).\nscale Scale existing velocities\n[factor <fac>] Factor to scale velocities by.\n[sx <xfac>] Factor to scale X component of\nvelocities by.\n[sy <yfac>] Factor to scale Y component of\nvelocities by.\n[sz <zfac>] Factor to scale Z component of\nvelocities by.\nadd Add to existing velocities\n[value <val>] Value to add to velocities.\n[vx <xval>] Value to add to X component of\nvelocities.\n[vy <yval>] Value to add to Y component of\nvelocities.\n[vz <zval>] Value to add to Z component of\nvelocities.\nnone Remove any velocities.\nmodify If specified, do not set, just modify any\nexisting velocities (via ’ntc’ or ’zeromomentum’).\nig <random seed> Random seed to use to generate\nvelocity distribution.\nntc <#> Correct set velocities for SHAKE constraints.\nNumbers match sander/pmemd: 1 = no SHAKE, 2 = SHAKE\non hydrogens, 3 = SHAKE on all atoms.\ndt <time> Time step for SHAKE correction.\nepsilon <eps> Epsilon for SHAKE correction\nzeromomentum If specified adjust velocities so the\ntotal momentum of atoms in <mask> is zero.\n203"}, {"page": 204, "text": "Setvelocitiesinframeforatomsin<mask>usingMaxwelliandistributionbased\non given temperature, optionally adjusted for SHAKE constraints. Can also be\nused to modify existing velocity information or remove it entirely. The total\nmomentum of the system can be set to zero as well, which can be useful for\nNVE simulations.\n11.80 spam\nspam [name <name>] [out <datafile>] [cut <cut>] [solv <solvname>]\n{ purewater |\n<peaksname> [reorder] [info <infofile>] [summary <summary>]\n[site_size <size>] [sphere] [temperature <T>]\n[dgbulk <dgbulk>] [dhbulk <dhbulk>] }\nname <name> Output data sets name.\nout <datafile> Data file with all SPAM energies for\neach snapshot.\ncut <cut> Non-bonded cutoff for energy evaluation\nsolv <solvname> Name of the solvent residues.\n[purewater] The system is pure water. Used to\nparametrize the bulk values. If this is specified,\nnone of the below options are relevant.\n<peaksname> Data set or file (XYZ- format: see\nbelow) with the peak locations present .\n[reorder] The solvent should be re-ordered so the same\nsolvent molecule is always in the same site.\ninfo <infofile> File with stats about which sites are\noccupied when.\nsummary <summary> File with the summary of all SPAM\nresults. If not specified, no SPAM energies will be\ncalculated.\nsite_size <size> Size of the water site around each\ndensity peak (sphere diameter/box edge length) in\nAng.\n[sphere] Treat each site like a sphere.\ntemperature <T> Temperature at which SPAM calculation\nwas run.\ndgbulk <dgbulk> SPAM free energy of the bulk solvent\nin kcal/mol; default is -30.3 kcal/mol (SPC/E\nwater).\ndhbulk <dhbulk> SPAM enthalpy of the bulk solvent in\nkcal/mol; default is -22.2 kcal/mol (SPC/E water).\n204"}, {"page": 205, "text": "Data Sets Created for ’purewater’:\n<name> Energies for each water at each frame.\nData Sets Created otherwise:\n<name>:<#> SPAM energies for peak <#> starting from\n1.\n<name>[DG] SPAM delta G values for valid peaks.\n<name>[DH] SPAM delta H values for valid peaks.\n<name>[-TDS] SPAM -T * delta S values for valid\npeaks.\nPerformprofilingofboundwatermoleculesviaSPAManalysis[28]. Briefly, this\nmethod identifies and estimates the free energy profiles of bound waters via\ncalculationofthedistributionofinteractionenergiesbetweenthewaterandit’s\nenvironmentfromexplicitsolventMDtrajectories. Theinteractionenergiesare\ncalculatedusingaforce-andenergy-shiftedelectrostatictermwithahardcutoff.\nFor a given peak, SPAM energies will only be calculated for peaks where the\npeak is singly-occupied (i.e. a multiple-occupied peak is not considered valid).\nPriortothiscommand,thevolmap commandshouldberunwiththepeak-\nfilekeyword(see11.93onpage217)togeneratethepeaksfile. Ifnotusingpeaks\nfrom the volmap command, the peaks file should have one line per peak with\nformat:\n<# of peaks>\nC <X> <Y> <Z> <Peak Density>\n...\nWith a ’C’ line for each peak.\n11.81 stfcdiffusion\nstfcdiffusion mask <mask> [out <file>] [time <time per frame>]\n[mask2 <mask> [lower <distance>] [upper <distance>]]\n[nwout <file>]) [avout <file>] [distances] [com]\n[x|y|z|xy|xz|yz|xyz]\nmask Atoms for which MSDs will be computed.\nout Output file: time vs. MSD.\ntime Time step in the trajectory. (1.0 ps)\nmask2 Compute MSDs only within the lower and upper\nlimit of mask2. IMPORTANT: may be very slow!!!\nlower Smaller distance from reference point(s). (0.01\nÅ)\nupper Larger distance from reference point(s). (3.5 Å)\n205"}, {"page": 206, "text": "nwout Output file containing number of water molecules\nin the chosen region, see mask2. (off)\navout Output file containing average distances. (off)\nx|y|z|xy|xz|yz|xyz Computation of the mean square\ndisplacement in the chosen dimension. (xyz)\ndistances Dump un-imaged distances. By default only\naverages are output. (off)\ncom Calculate MSD for centre of mass. (off)\nCalculatediffusionforselectedatomsusingcodebasedonthe’diffusion’routine\ndeveloped by Hannes Loeffler at STFC (http://www.stfc.ac.uk/CSE).\n11.82 strip\nstrip <mask> [charge <new charge>]\n[outprefix <prefix>] [nobox] [parmout <filename>]\n[parmopts <comma-separated-list>]\n<mask> Remove atoms specified by mask from the\nsystem.\ncharge <new charge> Scale charges so total charge of\nremaining atoms matches the specified <new charge>.\n[outprefix <prefix>] Write out stripped topology file\nwith name ’<prefix>.<Original Topology Name>’.\n[nobox] Remove any box information from the stripped\ntopology.\n[parmout <file>] Write stripped topology to file with\nname <file>.\n[parmopts <list>] Options for writing topology file.\nStrip all atoms specified by <mask> from the frame and modify the topology\nto match for any subsequent Actions. The outprefix keyword can be used\nto write stripped topologies; stripped Amber topologies are fully-functional.\nAvailable options for <parmopts> can be determined by running the help\nFormats parmwrite command.\nNote that stripping a system renumbers all atoms and residues, so for ex-\nample after this command:\nstrip :1\nresidue 1 will be gone, and the former second residue will now be the first, and\nso on.\nForexample,tostripallresiduesnamedWATfromeachtopology/coordinate\nframe:\n206"}, {"page": 207, "text": "strip :WAT\nThe next example uses a distance-based mask to strip atoms in a single frame.\nNote that with the exception of the mask command, distance-based masks do\nnot update on a per-frame basis. To strip all residues outside of 6.0 from any\natom in residues 1 to 14 and write out the stripped topology and coordinates,\nboth with no box information:\nparm parm7\ntrajin frame_1000.rst.1\nreference frame_1000.rst.1\nstrip !(:1-14<:6.0) outprefix f1.1 nobox\ntrajout f1.1.x restart nobox\n11.83 surf\nsurf [<name>] [<mask1>] [out <filename>] [solutemask <mask>]\n[offset <offset>] [nbrcut <cut>]\n<name> Output data set name.\n<mask1> Atoms to calculate surface area for.\nout <filename> File to write surface area to.\nsolutemask <mask> If specified, calculate the\ncontribution of <mask1> to <mask>.\noffset <offset> Increment van der Waals radii by\n<offset>; 1.4 Ang. is the default (as used by\nAmber).\nnbrcut <cut> Only atoms with van der Waals radii\ngreater than <cut> are considered to have neighbors\n(2.5 Ang Amber default).\nCalculate the surface area in Å2 of atoms in <mask> (if no mask specified, all\natoms not marked as ’solvent’ that are part of a molecule > 1 atom in size)\nusing the LCPO algorithm of Weiser et al.[29]. In order for this to work, the\ntopology needs to have bond information and atom type information.\nNote that even if <mask> does not include all solute atoms, the neighbor\nlist is still calculated for all solute atoms so the surface area calculated reflects\nthecontributionofatomsin<mask>totheoverallsurfacearea,notthesurface\nareaof<mask>asanisolatedsystem. Asaresult,itmaybepossibletoobtain\na negative surface area if only a small fraction of the solute is selected.\nForexample, to calculatethe overall surfacearea ofall soluteatoms, as well\nas the contribution of residue 1 to the overall surface area, writing both results\nto “surf.dat”:\nsurf out surf.dat\nsurf :1 out surf.dat\n207"}, {"page": 208, "text": "11.84 symmrmsd\nsymmrmsd [<name>] [<mask>] [<refmask>] [out <filename>] [nofit] [mass] [remap]\n[ first | reference | ref <name> | refindex <#> | previous |\nreftraj <name> [parm <parmname> | parmindex <#>] ]\n[<name>] Output data set name.\n[<mask>] Mask of atoms to calculate RMSD for; if not\nspecified, calculate for all atoms.\n[<refmask>] Reference mask; if not specified, use\n<mask>.\n[out <filename>] Output data file name.\n[nofit] Do not perform best-fit RMSD (not recommended).\n[mass] Mass-weight the RMSD calculation.\n[remap] Re-arrange atoms according to symmetry. See\nbelow for more details.\nReference keywords:\nfirst Use the first trajectory frame processed as\nreference.\nreference Use the first previously read in reference\nstructure (refindex 0).\nref <name> Use previously read in reference structure\nspecified by filename/tag.\nrefindex <#> Use previously read in reference\nstructure specified by <#> (based on order read in).\nprevious Use frame prior to current frame as reference.\nreftraj <name> Use frames from COORDS set <name> or\nread in from trajectory file <name> as references.\nEach frame from <name> is used in turn, so that\nframe 1 is compared to frame 1 from <name>, frame 2\nis compared to frame 2 from <name> and so on. If\n<trajname> runs out of frames before processing is\ncomplete, the last frame of <trajname> continues to\nbe used as the reference.\nparm <parmname> | parmindex <#> If reftraj\nspecifies a file associate trajectory <name>\nwith specified topology; if not specified the\nfirst topology is used.\nPerform symmetry-corrected RMSD calculation. This is done by identifying\npotential symmetric atoms in each residue, performing an initial best-fit, then\n208"}, {"page": 209, "text": "determiningwhichconfigurationofsymmetricatomswillgivethelowestRMSD\nusing atomic distance to reference atoms.\nNote that when re-mapping, all atoms in the residues of interest\nshould be selected to prevent cases where selected symmetric atoms\nare swapped but the atoms they are bonded to are not. Also, occasion-\nally larger symmetric structures (e.g. 6 membered rings) may become distorted\ndue to only part of the residue being corrected for symmetry. This appears\nto happen about 4% of the time but does not overly inflate the RMSD. The\n’check’ command can be used after symmrmsd to look for such distortions.\nWarning: thesymmetrycorrectionisgenerallyrobustenoughtoaccountfor\nsymmetries in the standard amino and nucleic acid residues, but has not been\nextensively tested on residues with more extended types of symmetry.\n11.85 temperature\ntemperature [<name>] [out <filename>]\n{ frame |\n[<mask>] [ntc <#>] [update] [remove {trans|rot|both}]\n}\n[<name>] Data set name.\n[out <filename>] File to write values to.\nframe Do not calculate temperature; use existing frame\ntemperature.\n[<mask>] Atoms to calculate temperature for.\n[ntc <#>] Value of SHAKE bond constraint: 1 - none, 2\n- bonds to H, 3- all bonds (equivalent to\nSANDER/PMEMD).\n[update] Update temperature in Frames with calculated\ntemperatures.\n[remove {trans|rot|both}] Correct for removed\ntranslational, rotational, or both kinds of degrees\nof freedom.\nCalculate temperature in frame based on velocity information. If ’update’ is\nspecified, update frame temperature too. If ’frame’ is specified just use frame\ntemperature (e.g. read in from a REMD trajectory).\nThe ’ntc’ keyword can be used to correct for lost degrees of freedom due\nto SHAKE constraints (2 = bonds to hydrogen, 3 = all bonds). The ’remove’\nkeyword can be used to account for removed translational and/or rotational\ndegrees of freedom.\nFor example, if using a trajectory that has been generated with SHAKE\non hydrogens, no periodic boundary conditions (i.e. no box), and has had the\ncenter of mass periodically removed:\n209"}, {"page": 210, "text": "temperature T1 ntc 2 remove both out T1.dat\nIf using a trajectory that has been generated with SHAKE on hydrogens, pe-\nriodic boundary conditions (i.e. with a box), and has had the center of mass\nperiodically removed:\ntemperature T1 ntc 2 remove trans out T1.dat\nIfusingatrajectorythathasbeengeneratedwithSHAKEonallbonds,periodic\nboundary conditions, and no center of mass motion removal:\ntemperature T1 ntc 3 out T1.dat\n11.86 time\ntime {time0 <initial time> dt <step> [update] | remove}\ntime0 <initial time> Time of the first frame (ps).\ndt <step> Time step between frames (ps).\n[update] If specified, modify any existing time info.\nremove Remove any time info from frame.\nEither add time information to frames, modify existing time information in\nframes, or remove existing time information from frames. Note that currently\nCOORDS data sets do not store time information, so using this command with\nthe crdaction command will have no effect.\n11.87 tordiff\ntordiff [<set name>] [<mask>] [mass] [out <file>] [diffout <file>] [time <dt>]\n[<set name>] Output mean-squared-displacement data sets\nname.\n[<mask>] Mask of molecules to calculate diffusion for.\n[mass] Use center of mass of molecules instead of\ngeometric center.\n[out <file>] File to write mean-squared-displacement\nsets to.\n[diffout <file>] Write diffusion contants calculated from\nfits of mean-squared-displacent data sets to\n<filename>.\n[time <dt>] Time between frames in ps.\nDataSet Aspects:\n[X] MSD(s) in the X direction.\n210"}, {"page": 211, "text": "[Y] MSD(s) in the Y direction.\n[Z] MSD(s) in the Z direction.\n[R] Overall MSD(s).\n[A] Overall displacement(s) in Å.\n[D] Diffusion constants (1x10-5 cm2/s).\n[Label] Diffusion constant lablels.\n[Slope] Linear regression slopes.\n[Intercept] Linear refression Y-intercepts.\n[Corr] Linear regression correlation coefficients.\nCalculate the diffusion via mean-squared displacement (in Å2/ps) of specified\nmolecules using the toroidal-view-preserving (TOR) scheme of Hummer et al.\nNote that currently this only works for orthogonal boxes. (https:\n//arxiv.org/abs/2303.09418). Unlikethediffusion (11.24onpage120)and\nunwrap (11.90onthenextpage)commandswhichcorrectforboxfluctuations\nvia fractional coordinates, the TOR scheme corrects for box fluctuations by\ntracking the displacement of each molecule with respect to its position in the\nprevious frame in Cartesian space.\nDiffusion constants are calculated in the same manner as the diffusion\ncommand; for more details see 11.24 on page 120.\nFor example, to calculate the diffusion (with data set name TOR) of waters\n(name WAT) in an orthogonal box using the toroidal scheme:\nparm tz2.ortho.parm7\ntrajin tz2.ortho.nc\ntordiff TOR :WAT@O out tor.msd.dat diffout tor.diff.dat\n11.88 trans | translate\ntranslate [<mask>] {[x <dx>] [y <dy>] [z <dz>] | topoint <x>,<y>,<z> [mass]}\n<mask> Mask of atoms to translate (all atoms if not\nspecified).\nx <dx> Translation (delta) in the X direction (Å).\ny <dx> Translation (delta) in the Y direction (Å).\nz <dx> Translation (delta) in the Z direction (Å).\ntopoint <x>,<y>,<z> If specified, translate center\nof specified atoms to a specific point defined by\n<x>, <y>, and <z> in the given comma-separated list\ninstead of by deltas.\nmass If specified, translate center of mass of\nspecified atoms (topoint only).\n211"}, {"page": 212, "text": "Translate atoms in <mask> (all atoms if no mask specified) <dx> Å in the\nX direction, <dy> Å in the Y direction, and <dz> Å in the Z direction. If\n’topoint’ is specified, translate atoms in <mask> to the specified coordinates\n(also in Å).\n11.89 unstrip\nunstrip\nRequests that the original topology and frame be used for all following actions.\nThis has the effect of undoing any command that modifies the state (such as\nstrip). For example, the following code takes a solvated complex and uses a\ncombination of strip, unstrip, and outtraj commands to write out separate dry\ncomplex, receptor, and ligand files:\nparm Complex.WAT.pdb\ntrajin Complex.WAT.pdb\n# Remove water, write complex\nstrip :WAT\nouttraj Complex.pdb pdb\n# Reset to solvated Complex\nunstrip\n# Remove water and ligand, write receptor\nstrip :WAT,LIG\nouttraj Receptor.pdb pdb\n# Reset to solvated Complex\nunstrip\n# Remove water and receptor, write ligand\nstrip :WAT\nstrip !(:LIG)\nouttraj Ligand.pdb pdb\n11.90 unwrap\nunwrap [center] [{byatom | byres | bymol}] [avgucell <avg ucell set>]\n[ reference | ref <name> | refindex <#> ] [<mask>]\n[scheme {frac|tor}]\n[center] Unwrap by center of mass; otherwise unwrap by\nfirst atom position (byres and bymol).\nbyatom Unwrap by atom (default).\nbyres Unwrap by residue.\nbymol Unwrap by molecule.\n[avgucell <avg ucell set>] Average unit cell data set;\nuseful when unwrapping NPT trajectories.\n212"}, {"page": 213, "text": "[ reference | ref <name> | refindex <#> ] Reference\nstructure to use in unwrapping.\n[<mask>] Selection to unwrap.\n[scheme frac|tor}] Unwrap using either fractional\ncoordinates (default) or toroidal-view-preserving\nscheme (orthogonal boxes only).\nUnder periodic boundary conditions, MD trajectories are not continuous if\nmolecules are wrapped(imaged) into the central unit cell. Especially, in sander,\nwith iwrap=1, molecular trajectories become discontinuous when a molecule\ncrossestheboundaryoftheunitcell. Thiscommand,unwrap processesthetra-\njectories to force the masked molecules continuous by translating the molecules\ninto the neighboring unit cells. It is the opposite function of image, but this\ncommand can also be used to place molecules side by side, for example, two\nstrands of a DNA duplex. However, this command may fail if the masked en-\ntities travel more than half of the box size within a single frame. Note that\nuwrappingbyatom(thedefaultbehavior)isslowerthanunwrappingbyresidue\nor molecule, but is usually the safer method, especially if it is unknown if the\noriginal imaging was done by atom, by residue, or by molecule. If the optional\nreference arguments are specified, then the first frame is unwrapped according\nto the reference structure. Otherwise, the first frame is not modified.\nTo remove the “noise” caused by box fluctuations in NPT trajectories, the\naverage unit cell vectors describing the average box can be provided with the\navgucell keyword; see the avgbox command (11.10 on page 110).\nBy default coordinates are unwrapped using fractional coordinates. To un-\nwrapviathetoroidal-view-preservingscheme(https://arxiv.org/abs/2303.\n09418)specifyscheme tor. However,notethatthiscurrentlyonly works for\northogonal boxes and may result in unrealistic trajectories (particularly at\nlongtimeswhenmoleculeshavehadthechancetotraversemultipleboxlengths)\nand is only currently recommended for calculating diffusion.\nAs an example, assume that :1-10 is the first strand of a DNA duplex and\n:11-20 is the other strand of the duplex. Then the following commands could\nbe used to create system where the two strands are not separated artificially:\nunwrap :1-20\ncenter :1-20 mass origin\nimage origin center familiar\nTounwrapanNPTtrajectorybyusingtheaverageunitcell(box),thecalculate\ndiffusion from the unwrapped trajectory:\nparm tz2.ortho.parm7\ntrajin tz2.ortho.nc\n# Create average box data\navgbox MyBox\nrun\n213"}, {"page": 214, "text": "# Unwrap using average box data\nunwrap bymol avgucell MyBox[avg]\n# Calculate diffusion\ndiffusion Water :WAT@O out tz2.ortho.wato.dat\nrun\n11.91 vector\nvector [<name>] <Type> [out <filename> [ptrajoutput]] [<mask1>] [<mask2>]\n[magnitude] [geom] [ired] [gridset <grid>] [debye]\n<Type> = { mask | minimage | dipole | center | corrplane |\nbox | boxcenter | ucellx | ucelly | ucellz\nmomentum | principal [x|y|z] | velocity | force }\n[<name>] Vector data set name.\n<Type> Vector type; see below.\n[out <filename>] Write vector data to <filename> with\nformat ’Vx Vy Vz Ox Oy Oz’ where V denotes vector\ncoordinates and ’O’ denotes origin coordinates.\n[ptrajoutput] Write vector data in ptraj style (Vx Vy Vz\nOx Oy Oz Vx+Ox Vy+Oy Vz+Oz). This prevents\nadditional formatting of <filename> and is not\ncompatible with ’magnitude’.\n[<mask1>] Atom mask, required for all types except\n’box’.\n[<mask2>] Second atom mask, only required for type\n’mask’.\n[magnitude] Store the magnitude of the vector with\naspect [Mag].\n[geom] If specified, use geometric centers instead of\ncenters of mass.\n[ired] Mark this vector for subsequent IRED analysis with\ncommands ’matrix ired’ and ’ired’.\n[gridset <grid>] Name of grid data set to get box info\nfrom instead of frame for box, boxcenter, and\nucell[x|y|z].\n[debye] (’dipole’ vector only) Report dipole vector in\nunits of Debye instead of e-*Ang.\nData Sets Created:\n<name> Vector data set.\n<name>[Mag] (magnitude only) Vector magnitude.\n214"}, {"page": 215, "text": "This command will keep track of a vector value (and its origin) over the trajec-\ntory; the data can be referenced for later use based on the name (which must\nbe unique). The types of vectors that can be calculated are:\nmask (Default) Store vector from center of mass of atoms in <mask1> to\natoms in <mask2>.\nminimage Store minimum-imaged vector from center of mass of atoms in\n<mask1> to atoms in <mask2>.\ndipole Storethedipoleandcenterofmassoftheatomsspecifiedin<mask1>.\nThedipolevectorhasunitsofe-*Angunless’debye’isspecifiedforunits\nof Debye. The center is always stored as simply Ang (since it is just\ncoordinates). Note that the value may not be well-defined if the atoms in\nthe mask are not overall charge neutral.\ncenter Store the center of mass of atoms in <mask1>. The reference point\nis the origin (0.0, 0.0, 0.0).\ncorrplane Thisdefinesavectorperpendiculartothe(least-squaresbest)plane\nthroughtheatomsin<mask1>. Thereferencepointisthecenterofmass\nof atoms in <mask1>.\nbox (No mask needed) Store the box lengths of the trajectory. The reference\npoint is the origin (0.0, 0.0, 0.0).\nboxcenter (No mask needed) Store the center of the box as a vector.\nucell{x|y|z}: (No mask needed) Store specified unit cell (i.e. box) vector.\nmomentum Store momentum of atoms selected by <mask1> (requires veloc-\nities).\nprincipal [x|y|z] Store one of the principal axis vectors determined by diago-\nnalizationoftheinertialmatrixfromthecoordinatesoftheatomsspecified\nby <mask1>. The eigenvector with the largest eigenvalue is considered\n“x” (i.e., the hardest axis to rotate around) and the eigenvector with the\nsmallest eigenvalue is considered “z”. If none of x or y or z are specified,\nthen the “x” principal axis is stored. The reference point is the center of\nmass of atoms in <mask1>.\nvelocity Store velocity of atoms in <mask1> (requires velocities).\nforce Store force of atoms in <mask1> (requires forces).\nCpptrajsupportswritingoutvectordatainapseudo-trajectoryformatforeasy\nvisualization. Onceavectordatasethasbeengeneratedthewritedatacommand\ncanbeusedwiththevectrajkeyword(see6onpage27formoredetails)towrite\na pseudo trajectory consisting of two atoms, one for the vector origin and one\nfor the vector from the origin (i.e. V+O). For example, to create a MOL2\ncontainingapseudo-trajectoryoftheminimum-imagedvectorfromresidue4to\nresidue 11:\n215"}, {"page": 216, "text": "trajin tz2.nc\nvector v8 minimage out v8.dat :4 :11\nrun\nwritedata v8.mol2 vectraj v8 trajfmt mol2\nAuto-correlation or cross-correlation functions can be calculated subsequently\nfor vectors using either the corr analysis command or the timecorr analysis\ncommand (to calculate via spherical harmonic theory).\n11.92 velocityautocorr\nvelocityautocorr [<set name>] [<mask>] [usevelocity] [out <filename>] [diffout <file>]\n[maxlag <frames>] [tstep <timestep>] [direct] [norm]\n[<set name>] Data set name.\n[<mask>] Atoms(s) to calculate velocity\nautocorrelation (VAC) function for.\n[usevelocity] Use velocity information in frame if\npresent. This will only give sensible results if\nthe velocities are recorded close to the order of\nthe simulation time step.\n[out <filename>] Write VAC function to <filename>.\n[diffout <file>] File to write diffusion constants to.\n[maxlag <frames>] Maximum lag in frames to calculate\nVAC function for. Default is half the total number\nof frames.\n[tstep <timestep>] Time between frames in ps (default\n1.0).\n[direct] Calculate VAC function directly instead of via\nFFT (will be much slower).\n[norm] Normalize resulting VAC function to 1.0.\nDataSet Aspects:\n[D] Diffusion constant calculated from integral over VAC\nfunction in 1x10-5 cm2/s.\nCalculate the velocity autocorrelation (VAC) function averaged over the atoms\nin <mask>. Pseudo-velocities are calculated using coordinates and the speci-\nfied time step. As with all time correlation functions the statistical noise will\nincrease if the maximum lag is greater than half the total number of frames. In\naddition to calculating the velocity autocorrelation function, the self-diffusion\ncoefficient will be reported in the output, calculated from the integral over the\nVAC function.\n216"}, {"page": 217, "text": "11.93 volmap\nvolmap [out <filename>] <mask> [radscale <factor>] [stepfac <fac>]\n[sphere] [radii {vdw | element}] [splinedx <spacing>]\n[calcpeaks] [peakcut <cutoff>] [peakfile <xyzfile>]\n{ data <existing set> |\nname <setname> <dx> [<dy> <dz>]\n{ size <x,y,z> [center <x,y,z>] |\ncentermask <mask> [buffer <buffer>] |\nboxref <reference> } }\nout <filename> The name of the output file with the\ngrid density.\n<mask> The atom selection from which to calculate the\nnumber density.\nradscale <factor> Factor by which to scale radii (by\ndivision). To match the atomic radius of Oxygen\nused by the VMD volmap tool, a scaling factor of\n1.36 should be used. Default 1.0.\nstepfac <factor> Factor for determining how many voxels\nto smear Gaussian (default 4.1, 1.0 for sphere).\nsphere When smearing Gaussian, skip voxels farther than\nradii/2.\nradii {vdw|element} Specify either van der Waals radii\n(default) or elemental radii.\nsplinedx <spacing> Spacing to use for cubic spline\ninterpolation (default 0.01 Ang.).\ncalcpeaks If specified, peaks in the grid density will\nbe calculated and saved to set <setname> with aspect\n“peaks”.\npeakcut <cutoff> The minimum density required to\nconsider a local maximum a ’density peak’ in the\noutputted peak file (default 0.05).\npeakfile <xyzfile> A file in XYZ-format that contains a\ncarbon atom centered at the grid point of every\nlocal density maximum. This file is necessary input\nto the spam action command.\ndata <setname> Name of existing grid data set to use.\nname <setname> Name of grid set that will be created\n(size/center or centermask/buffer keywords).\ndx, dy, dz The grid spacing (Angstroms) in the X-, Y-,\nand Z-dimensions, respectively.\n217"}, {"page": 218, "text": "size <x,y,z> Specify the size of the grid in the X-, Y-,\nand Z-dimensions. Must be used alongside the center\nargument.\ncenter <x,y,z> Specify the grid center explicitly.\nNote, the size argument must be present in this\ncase. Default is the origin.\ncentermask <mask> The mask around which the grid\nshould be centered (via geometric center). If this\nis omitted and the center and size are not\nspecified, the default <mask> entered (see above) is\nused in its place.\nbuffer <buffer> A buffer distance, in Angstroms, by\nwhich the edges of the grid should clear every\natom of the centermask (or default mask if\ncentermask is omitted) in every direction. The\ndefault value is 3. The buffer is ignored if\nthe center and size are specified (see below).\nboxref <reference> Set up the grid using the unit cell\ninfo in the specified reference.\nData Sets Created:\n<setname> The 3D grid.\n<setname>[peaks] The density peaks if calcpeaks\nspecified.\nGrid data as a volumetric map, similar to the ’volmap’ command in VMD.\nThe density is calculated by treating each atom as a 3-dimensional Gaussian\nfunction whose standard deviation is equal to the van der Waals radius. The\ndensity calculated is the number density averaged over the entire simulation.\nThe grid can be specified in one of three ways:\n1. An existing grid data set (from e.g. bounds), specified with the data\nkeyword.\n2. Viathesizesandcenterspecifiedbythesizeandcenterkeywords(comma-\nseparated strings, e.g. ’20,20,20’).\n3. Centered on the atoms in the mask given by centermask with an addi-\ntional buffer in each direction specified by buffer.\nThe calculation is sped up by using cubic splines to interpolate the exponential\nfunction when calculating the Gaussians.[30]\n11.94 volume\nvolume [<name>] [out <filename>]\n218"}, {"page": 219, "text": "<name> Data set name.\nout <filename> Output file name.\nCalculate unit cell volume.\n11.95 watershell\nwatershell <solutemask> [out <filename>] [lower <lower cut>] [upper <upper cut>]\n[noimage] [<solventmask>]\n<solutemask> Atom mask corresponding to solute of\ninterest (required).\n[out <filename>] Output file name.\n[lower <lower cut>] Cutoff for the first water shell\n(default 3.4 Angstroms).\n[upper <upper cut>] Cutoff for the second water shell\n(default 5.0 Angstroms).\n[noimage] Do not image distances.\n[<solventmask>] Optional atom mask corresponding to\nsolvent.\nDataSet Aspects:\n[lower] Number of solvent molecules in first solvent\nshell.\n[upper] Number of solvent molecules in second solvent\nshell.\nThis option will count the number of waters within a certain distance of the\natoms in the <solutemask> in order to represent the first and second solvation\nshells. Theoptional<solventmask>canbeusedtoconsiderotheratomsasthe\nsolvent; the default is “:WAT”.\nThisactionisoftenusedpriortotheclosest commandinordertodetermine\nhowmanywatersaroundasoluteshouldberetainedtomaintainthefirstand/or\nsecond water shells.\nAsofversion17thiscommandisCUDA-enabledinCUDAversionsofCPP-\nTRAJ.\n11.96 xtalsymm\nxtalsymm <mask> group <space group> [collect [centroid]]\n[ first | reference | ref <name> | refindex <#> ]\n[na <na>] [nb <nb>] [nc <nc>]\n<mask> Atom mask defining the asymmetric unit within\nthe larger system (required).\n219"}, {"page": 220, "text": "group <space group> The space group to which the\nsystem belongs. Omit spaces in the name. Example:\n“P22(1)2(1)”.\n[collect] Optional flag to have all solvent particles,\nnot just the asymmetric units, re-imaged. This will\ntrigger cpptraj to compute the unit cell volume that\nconstitutes the aymmetric unit and thereby classify\nall particles for re-imaging.\n[centroid] If specified along with collect, re-image\nsolvent molecules by centroids, not individual\natom coordinates. This is useful for keeping\nwater molecules intact.\n[first | reference | ref <name> | refindex <#>] Reference\nstructure to use for determining crystal symmetry.\n[na <na>] [nb <nb>] [nc <nc>] The number of times the\ncrystal unit cell is replicated along the “a,” “b,”\nor “c” axes (for orthorhombic unti cells, these are\nthe x, y, and z axes) of the simulation; default is\n1. Many crystal unit cells are too small in one or\nmore dimensions for our simulation cutoffs, and\nreplicating the unit cell is an effective way to\ncounter imaging artifacts even for larger unit\ncells.\nCalculate the optimal approach for superimposing symmetry-related subunits\nof the simulation back onto one another. The calculation assumes that the\nsystem is a simulation of an X-ray structure in its native crystal lattice, finds\nall copies of the asymmetric unit among the entire system, and devises plans\nfor re-imagining their coordinates to superimpose them back on the original\nasymmetric unit. The space group information can be found in a PDB X-\nray structure used as the initial coordinates for a simulation. All 230 space\ngroupsaresupported, andascanofthePDBwasmadetoensurethatcommon\nvariants of the names are included (P2(1)22(1) is the same as P22(1)2(1), but\nwith different axis conventions). If your space group is not understood, contact\nthe Amber mailing list. This command is compute intensive, especially for\nsimulations that are “supercells” containing many crystallographic unit cells.\nThiscommandwillcausecpptrajtolocateallasymmetricunitsfromwithin\nthe topology, then determine what wrapping, if any, has occurred in order to\nbring about an optimal re-alignment based on the space group symmetry op-\nerations. The user need not worry about wrapping or drift of the simulation\nover time–the asymmetric units will be re-imaged frame by frame. Coordinate\nmodifications due to this action are permanent and will affect the results of\nsubsequent actions and analyses.\n220"}, {"page": 221, "text": "12 Analysis Commands\nAnalyses in cpptraj operate on data sets which have been generated by Ac-\ntions in a prior Run or read in with a readdata command (8.22 on page 64).\nUnlike ptraj, Analysis commands in cpptraj do not need to be prefaced with\n’analysis’. The exception to this is ’analyze matrix’ in order to differenti-\nate it from the matrix Action command; users are encouraged to use the new\ncommand diagmatrix instead.\nLike Actions, when an Analysis command is issued it is by default added\nto the Analysis queue and is not executed until after trajectory processing is\ncompleted; a complete list of data sets available for analysis is shown after\ntrajectory processing (prefaced by ’DATASETS’) or can be shown with the\n’list dataset’ command. Analyses can also be executed immediately via the\nrunanalysis command (8.27 on page 66).\nNote that for Analysis commands that use COORDS data sets, if no CO-\nORDSdatasetisspecifiedthenadefaultonewillbeautomaticallycreatedfrom\nframes read in by trajin commands.\nCommand Description Set Type(s)\nautocorr Calculate autocorrelation function for N 1D sclar\nmultiple data sets.\navg Calculate average, standard deviation, N 1D scalar\nmin, and max for (or over) data sets.\ncalcdiffusion Calculate diffusion using multiple time COORDS\norigins.\ncalcstate Calculate states based on given data N 1D scalar\nsets and criteria.\ncluster Perform cluster analysis. COORDS, N 1D\nscalar\ncorr, Calculate auto or cross correlation for 1D scalar, vector\n1 or 2 data sets.\ncorrelationcoe\ncphstats Calculate statistics for constant pH pH data sets\ndata sets.\ncrank, Calculate crankshaft motion between 2 1D scalar\ntwo data sets.\ncrankshaft\ncrdfluct Calculate atomic fluctuations (RMSF) COORDS\nfor atoms over time blocks.\ncrosscorr Calculate a matrix of Pearson N 1D scalar\nproduct-moment\ncoefficients between given data sets.\ncurvefit Perform non-linear curve fitting on 1D scalar\ngiven data set.\n221"}, {"page": 222, "text": "diagmatrix Calculate eigenvectors and eigenvalues symmetric matrix\nfrom given symmetric matrix.\ndivergence Calculate Kullback-Leibler divergence 2 1D scalar\nbetween two data sets.\nevalplateau Evaluate whether the data in a 1D set\nhas reached a single exponential\nplateau.\nFFT Perform a fast Fourier transform on N 1D scalar\ndata sets.\nhausdorff Calculate the Hausdorff distance for N 2D matrices\ngiven matrix data set(s).\nhist, histogram Calculate N-dimensional histogram for N 1D scalar\nN given data sets.\nintegrate Perform integration on each of the N 1D scalar\ngiven data sets.\nired Perform isotropic reorientational N IRED vectors\neigenmode dynamics\nanalysis using given IRED vectors.\nkde Calculate 1D histogram from given 1 or 2 1D scalar\ndata set using a kernel density\nestimator.\nAlso time-dependent Kullback-Leibler\ndivergence analysis with another set.\nlifetime Perform lifetime analysis on given data N 1D scalar\nsets.\nlowestcurve For each given data set, calculate a N 1D scalar\ncurve that traces\nthe lowest N points over specified bins.\nmeltcurve Calculate a melting curve from given N 1D scalar\ndata sets assuming simple 2 state\nkinetics.\nmodes Perform various analyses on eigenmodes\neigenmodes (from e.g. diagmatrix).\nmulticurve Perform non-linear curve fitting for N 1D scalar\nmultiple input data sets.\nmultihist Calculate 1D histograms (optionally N 1D scalar\nwith a kernel\ndensity estimator) from multiple input\ndata sets.\nphipsi Calculate and plot the average phi and N phi/psi dihedrals\npsi values from input dihedral data\nsets.\nprojectdata Project data along eigenmodes. eigenmodes, N 1D\nscalar\n222"}, {"page": 223, "text": "regress Perform linear regression on multiple N 1D scalar\ninput data sets.\nremlog Calculate various statistics from a replica log\nreplica log data set.\nrms2d, 2drms Calculate 2D RMSD between frames in 1 or 2 COORDS\n1 or 2 COORDS data sets.\nrmsavgcorr Calculate RMS average correlation COORDS\ncurve for a COORDS data set.\nrotdif Calculate rotational diffusion using rotation matrices\ngiven rotation matrices (from e.g.\nrms).\nrunningavg Calculate running average for given N 1D scalar\ndata sets using given window size.\nspline Calculate cubic splines for given data N 1D scalar\nsets.\nstat, statistics Calculate various statistics for given N 1D scalar\ndata sets.\nti Peform Gaussian quadrature N 1D scalar\nintegration for given DV/DL data sets.\ntica Perform time-independent correlation COORDS, N 1D\nanalysis. scalar\ntimecorr Calculate auto/cross-correlation 1 or 2 vector\nfunctions for given\nvector(s) using spherical harmonics.\nvectormath Perform math on given vector data 2 vector\nsets.\nwavelet Perform wavelet analysis on COORDS\ncoordinates from given COORDS set.\n12.1 autocorr\nautocorr [name <dsetname>] <dsetarg0> [<dsetarg1> ...] [out <filename>]\n[lagmax <lag>] [nocovar] [direct]\n<dsetarg0> [dsetarg1> ...] Argument(s) specifying\ndatasets to be used.\n[name <dsetname>] Store results in dataset(s) named\n<dsetname>:X.\n[out <filename>] Write results to file named\n<filename>.\n[lagmax] Maximum lag to calculate for. If not specified\nall frames are used.\n223"}, {"page": 224, "text": "[nocovar] Do not calculate covariance.\n[direct] Do not use FFTs to calculate correlation; this\nwill be much slower.\nThis is for integer/double/float datasets only; for vectors see the ’timecorr’\ncommand.\nCalculateauto-correlation(actuallyauto-covariancebydefault)functionfor\ndatasets specified by one or more dataset arguments. The datasets must have\nthe same # of data points.\n12.2 avg\navg <dset0> [<dset1> ...] [torsion] [out <file>] [oversets]\n[name <name>] [nostdout]\n<dsetX> Data set(s) to calculate the average for.\n[torsion] If the data sets are not already marked\nperiodic (e.g. if read in via ’readdata’), treat\nthem as periodic torsion.\n[out <file>] File to write results to.\n[oversets] If specified, calculate the average over all\ninpout sets instead of each input set.\n[name <name>] Output data set name.\n[nostdout] If ’nostdout’ specified do not write averages\nto STDOUT when ’out’ not specified.\nDataSets Created (not oversets):\n<name>[avg] Average of each set.\n<name>[sd] Standard deviation of each set.\n<name>[ymin] Y minimum of each set.\n<name>[ymax] Y maximum of each set.\n<name>[yminidx] Index of minimum Y value.\n<name>[ymaxidx] Index of maximum Y value.\n<name>[names] Name of each set.\nDataSets Created (oversets)\n<name> Average over all input sets for each frame.\n<name>[SD] Standard deviation over all input sets for\neach frame.\n224"}, {"page": 225, "text": "Calculate the average, standard deviation, min, and max of given 1D data sets.\nAlternatively, if oversets is specified the average over each set for each point\nis calculated; this requires all input sets be the same size.\nFor example, to read in data from a file named perres.peptide.dat and cal-\nculate the averages etc for all the input sets:\nreaddata perres.peptide.dat\navg perres.peptide.dat out output.dat name V\n12.3 calcdiffusion\ncalcdiffusion [crdset <coords set>] [maxlag <maxlag>] [<mask>] [time <dt>]\n[<name>] [out <file>] [diffout <file>]\n[crdset <coords set>] Input COORDS set to calculate\ndiffusion for.\n[maxlag <maxlag>] Maximum lag to calculate diffusion\nfor in frames. Defaults to half the number of input\nframes if not specified.\n[<mask>] Mask of atoms to calculate diffusion for.\n[time <dt>] Time between frames in ps.\n[<name>] Output MSD data sets name.\n[out <file>] File to write MSD data sets to.\n[diffout <file>] Write diffusion contants calculated from\nfits of MSD data sets to <filename>.\nDataSet Aspects:\n[X] MSD(s) in the X direction.\n[Y] MSD(s) in the Y direction.\n[Z] MSD(s) in the Z direction.\n[R] Overall MSD(s).\n[A] Overall displacement(s) in Å.\n[D] Diffusion constants (1x10-5 cm2/s).\n[Label] Diffusion constant lablels.\n[Slope] Linear regression slopes.\n[Intercept] Linear refression Y-intercepts.\n[Corr] Linear regression correlation coefficients.\nCalculate diffusion via mean-squared-displacements (MSD) of specified atoms.\nUnlike the diffusion Action (11.24 on page 120), which calculates MSD from\nasingletimeorigin, thecalcdiffusion commandcalculatesdiffusionfrommul-\ntiple time origins up to a user-specified lag (in frames). Note that no imaging\n225"}, {"page": 226, "text": "is performed for this command, so any unwrapping should be performed prior\nto this command (see 11.90 on page 212). Diffusion constants are calculated\nin the same manner as the diffusion command; for more details see 11.24 on\npage 120.\nThis command is both OpenMP- and MPI-parallelized; either or both can\nbe active. In order to keep per-process memory requirements low, it is recom-\nmended that TRAJ (i.e. on-disk) data sets be used with MPI instead of CRD\n(in-memory) sets (see 7.11 on page 42).\nFor example, to calculate diffusion from multiple time origins (with a maxi-\nmum lag of half the number of trajectrory frames) from an input trajectory in\nmemory (unwrapping first):\nparm tz2.ortho.parm7\nloadcrd tz2.ortho.nc name TZ2\ncrdaction TZ2 unwrap bymol\nrunanalysis calcdiffusion crdset TZ2 out tz2.diff.dat :WAT@O\nTo do the same calculation using MPI parallelism, it will first be necessary\nto unwrap the trajectory (non-MPI parallel, but OpenMP parallelism can be\nused):\nparm tz2.ortho.parm7\ntrajin tz2.ortho.nc\nunwrap bymol\ntrajout unwrap.dcd\nThen diffusion can be calculated using MPI (and/or OpenMP):\nparm tz2.ortho.parm7\nloadtraj unwrap.dcd name TZ2\nrunanalysis calcdiffusion crdset TZ2 out tz2.traj.diff.dat :WAT@O\n12.4 calcstate\ncalcstate {state <ID>,<dataset>,<min>,<max>[,<dataset1>,<min1>,<max1>]} ...\n[out <state v time file>] [name <setname>]\n[curveout <curve file>] [stateout <states file>]\n[transout <transitions file>] [countout <count file>]\nstate <ID>,<dataset>,<min>,<max> Define a state\naccording to given data set and criteria. Multiple\nstates can be given, and each state can have\nmultiple criteria. If multiple criteria are\nspecified, each one must be satisfied in order to\nassign the state. If the same state is defined\nmultiple times, the state will be assigned if either\ncriteria match.\n226"}, {"page": 227, "text": "<ID> Name to give each state index. State indices\nstart at 0. -1 means “undefined state”.\n<dataset> Data set to use.\n<min>,<max> Frames with data set value above\n<min> and below <max> will be assigned <ID>.\n[out <state v time file>] File to write state index vs\nframe to.\n[name <setname>] Data set name.\n[curveout <curve file>] File to write state lifetime and\ntransition curves to.\n[stateout <states file>] File to write state lifetime data\nto.\n[transout <transitions file>] File to write state\ntransition data to.\n[countout <state count file>] File to write state counts\n(i.e. how many frames each state was observed) to.\nDataSets Created:\n<setname> State index vs frame.\n<setname>[Count] Number of frames each state was\nobserved.\n<setname>[Frac] Fraction of time each state was\nobserved\n<setname>[Nlifetimes] Number of times each state was\nreached.\n<setname>[Avglife] Average lifetime length for each\nstate.\n<setname>[Maxlife] Maximum lifetime of each state.\n<setname>[Name] Name (<ID>) of each state.\n<setname>[Xlifetimes] Number of times each state\ntransitioned to each other state.\n<setname>[Xavglife] Average lifetime of each state\nbefore transitioning to each other state.\n<setname>[Xmaxlife] Maximum lifetime of each state\nbefore transitioning to each other state.\n<setname>[Xname] Name of each transition, format\n“StateA->StateB”.\n<setname>[sCurve]:X State curves; lifetime curve for\ntransitions from given state to any other state.\n227"}, {"page": 228, "text": "<setname>[tCurve]:X Transition curves; lifetime curve\nfor transitions from given state to other specific\nstate.\nDataforthespecifieddataset(s)thatmatchesthegivencriteriawillbeassigned\na state index. State indices start from 0 and match the order in which state\nkeywords were given. The -1 state index is reserved for “undefined state”. For\nexample, the following input:\nparm DPDP.parm7\ntrajin DPDP.nc\ndistance d1 :19@O :12@N\nangle a1 :19@O :12@H :12@N\ncalcstate state D,d1,3.0,4.0 state A,a1,100,120 out state.dat curveout curve.agr \\\nstateout States.dat transout States.dat name d1_a1\nrun\nDefines two states. State index 0 is defined as a state named “D” based on\nthe distance from ’:19@O’ to ’:12@N’ (data set d1) being between 3 and 4\nAngstroms. State index 1 is defined as a state named “A” based on the angle\nbetween’:19@O’,’:12@H’,and’:12@N’(dataseta1)beingbetween100and120\ndegrees. The output in state.dat might look like:\n#Frame d1_a1\n1 -1\n2 0\n3 0\n4 0\n5 -1\n6 1\n7 -1\n8 -1\n9 0\n10 -1\nwhere the values in column d1_a1 refer to state index: -1 is undefined, 0 is\nstate “D”, and 1 is state “A”.\nTo define a state State1 as having a distance named “dist” between 2.5 and\n5.0 Ang. and an angle named “ang” between 30 and 60 degrees OR having a\ndistance named “distA” between 0.0 and 3.0 Ang.:\ncalcstate state State1,dist,2.5,5.0,ang,30,60 \\\nstate State1,distA,0.0,3.0\nLifetime curves (see 12.21 on page 257 for further explanation) are calculated\nfor transitions from each state to any other state (aspect [sCurve]) and each\nstate to each other state (aspect [tCurve]). In this case there will be 3 sCurves\nand 4 tCurves:\n228"}, {"page": 229, "text": "d1_a1[sCurve]:0 \"Undefined\" (double), size is 10\nd1_a1[sCurve]:1 \"D\" (double), size is 3\nd1_a1[sCurve]:2 \"A\" (double), size is 1\nd1_a1[tCurve]:0 \"Undefined->D\" (double), size is 10\nd1_a1[tCurve]:1 \"D->Undefined\" (double), size is 3\nd1_a1[tCurve]:2 \"Undefined->A\" (double), size is 1\nd1_a1[tCurve]:3 \"A->Undefined\" (double), size is 1\nLifetimeanalysisfromeachstatetoanyotherstateisdirectedtothefilespecified\nby stateout and has format:\n#Index N Average Max State\nWhere #Index is the state index, N is the number of lifetimes in that state,\nAverage is the average lifetime while in that state (in frames), Max is the max-\nimum lifetime while in that state (in frames) and State is the name of the\nstate.\nFinally, lifetime analysis of transitions from each state to each other state is\ndirectory to the file specified by transout and has format:\n#N Average Max Transition\nWhere #N is the number of transitions, Average is the average lifetime (in\nframes)inthefirststatebeforetransitioningtothesecondstate,Maxisthemax\nlifetime (in frames) before transitioning to the second state, and Transition is\nthe name of the transition.\n12.5 cluster\ncluster [crdset <crd set>] [data <dset0>[,<dset1>...]] [nocoords]\n[<name>] [<Algorithm>] [<Metric>] [<Pairwise>] [<Sieve>] [<BestRep>]\n[<Output>] [<Coord. Output>] [<Graph>]\n[readinfo {infofile <info file> | cnvtset <dataset>}]\n[useframesincache]\nAlgorithm Args: [{hieragglo|dbscan|kmeans|dpeaks}]\n[hieragglo [epsilon <e>] [clusters <n>] [linkage|averagelinkage|complete]\n[epsilonplot <file>]]\n[dbscan minpoints <n> epsilon <e> [kdist <k> [kfile <prefix>]]]\n[kmeans clusters <n> [randompoint [kseed <seed>]] [maxit <iterations>]]\n[dpeaks epsilon <e> [noise] [dvdfile <density_vs_dist_file>]\n[choosepoints {manual | auto}]\n[distancecut <distcut>] [densitycut <densitycut>]\n[runavg <runavg_file>] [deltafile <file>] [gauss]]\nMetric Args:\n[{dme|rms|srmsd|qrmsd} [mass] [nofit] [<mask>]] [{euclid|manhattan}] [wgt <list>]\nPairwise Args:\n229"}, {"page": 230, "text": "[pairdist <name> [pairdistfile <file>]] [pwrecalc]\n[loadpairdist] [savepairdist] [pairwisecache {mem|disk|none}]\nSieve Args:\n[sieve <#> [sieveseed <#>] [random] [includesieveincalc] [includesieved_cdist]\n[{sievetoframe|sievetocentroid|closestcentroid}] [repsilon <restore epsilon>]]\nBestRep Args:\n[bestrep {cumulative|centroid|cumulative_nosieve}] [savenreps <#>]\nOutput Args:\n[out <cnumvtime> [gracecolor]] [noinfo|info <file>] [summary <file>]\n[summarysplit <splitfile>] [splitframe <comma-separated frame list>]\n[clustersvtime <file> [cvtwindow <#>]] [sil <prefix> [silidx {idx|frm}]]\n[metricstats <file>] [cpopvtime <file> [{normpop|normframe}]] [lifetime]\nCoordinate Output Args:\n[clusterout <trajfileprefix> [clusterfmt <trajformat>]]\n[singlerepout <trajfilename> [singlerepfmt <trajformat>]]\n[repout <repprefix> [repfmt <trajformat>] [repframe]]\n[avgout <avgprefix> [avgfmt <trajformat>]]\n[assignrefs [refcut <rms>] [refmask <mask>]]\nGraph Args:\n[{drawgraph|drawgraph3d} [draw_tol <tolerance>] [draw_maxit <iterations]]\n[crdset <crd set>] Name of COORDS data set to cluster on\nand/or use for coordinate output. If not specified\nthe default COORDS set will be generated and used\nunless nocoords has been specified.\n[data <dset0>[,<dset1>,...] Distance between frames\ncalculated using specified data set(s). Currently\n1D scalar sets and COORDS sets are supported.\n[nocoords] Do not use a COORDS data set; distance\nmetrics that require coordinates and coordinate\noutput will be disabled.\n[<name>] Data set Name for generated cluster data\nsets.\n[readinfo] Use previous cluster results to set up initial\nclusters. Clustering will continue if possible\n(i.e. this can be used to restart clustering).\ninfofile <file> Cluster info file to read clusters\nfrom.\ncnvtset <dataset> Cluster number vs time data set\nto use to generate initial clusters.\n[useframesincache] If a pairwise cache is specified,\ncluster on the frames stored in the cache.\nAlgorithms:\n230"}, {"page": 231, "text": "hieragglo (Default) Use hierarchical agglomerative\n(bottom-up) approach.\n[epsilon <e>] Finish clustering when minimum\ndistance between clusters is greater than <e>.\n[clusters <n>] Finish clustering when <n> clusters\nremain.\n[linkage] Single-linkage; use the shortest distance\nbetween members of two clusters.\n[averagelinkage] Average-linkage (default); use the\naverage distance between members of two\nclusters.\n[complete] Complete-linkage; use the maximum\ndistance between members of two clusters.\n[epsilonplot <file>] Write number of clusters vs\nepsilon to <file>.\ndbscan Use DBSCAN clustering algorithm of Ester et\nal.[31]\nminpoints <n> Minimum number of points required to\nform a cluster.\nepsilon <e> Distance cutoff between points for\nforming a cluster.\n[kdist <k>] Generate K-dist plot for help in\ndetermining DBSCAN parameters (see below).\n[kfile <prefix>] Prefix for K-dist plot file.\ndpeaks Use the density peaks algorithm of Rodriguez and\nLaio[32]\nepsilon <e> Cutoff for determining local density in\nAngstroms.\n[noise] If specified, treat all points within epsilon\nof another cluster as noise.\n[dvdfile <density_vs_dist_file>] File to write\ndensity versus minimum distance to point with\nnext highest density. This can be used to\ndetermine appropriate cutoffs for distance and\ndensity in a subsequent step with choosepoints\nmanual.\n[choosepoints {manual | auto}] Specify whether\nclusters will be chosen based on specified\ndistance/density cutoffs, or automatically. If\nnot specified only the density vs distance file\nwill be written and no clustering will be\nperformed. Currently manual is recommended.\n231"}, {"page": 232, "text": "[distancecut <distcut>] [densitycut <densitycut>] If\nchoosepoints manual, points with minimum\ndistance greather than or equal to <distcut> and\ndensity greater than or equal to <densitycut>\nwill be chosen.\n[runavg <runavg file>] If choosepoints automatic,\nthe calculated running average of density versus\ndistance will be written to <runavg file>.\n[deltafile <file>] If choosepoints automatic, distance\nminus the running average for each point will be\nwritten to this file.\n[gauss] Calculate density with Gaussian kernels\ninstead of using discrete density.\nkmeans Use K-means clustering algorithm.\nclusters <n> Finish clustering when number of\nclusters is <n>.\n[randompoint] Randomize initial set of points used\n(recommended).\n[kseed <seed>] Random number generator seed for\nrandompoint.\n[maxit <iteration>] Algorithm will run until frames\nno longer change clusters of <iteration>\niterations are reached (default 100).\nDistance Metric Options:\n[{rms | srmsd} [<mask>]] (Default rms) For COORDS data,\ndistance between coordinate frames calculated via\nbest-fit coordinate RMSD using atoms in <mask>. If\nsrmsd specified use symmetry-corrected RMSD (see\n11.84 on page 208).\n[mass] Mass-weight the RMSD.\n[nofit] Do not fit structures onto each other prior\nto calculating RMSD.\nqrmsd [<mask>] For COORDS data, distance between\ncoordinate frames calculated using best-fit\nquaternion RMSD (can be 15-20% faster than regular\nRMSD) using atoms in <mask>.\n[mass] Mass-weight the RMSD.\ndme [<mask>] For COORDS data, distance between\ncoordinate frames calculated using distance-RMSD\n(aka DME, distrmsd) using atoms in <mask>.\neuclid Use Euclidean distance (sqrt(SUM(distance^2)))\nwhen more than one data set has been specified\n(default).\n232"}, {"page": 233, "text": "manhatttan Use Manhattan distance (SUM(distance)) when\nmore than one data set has been specified.\nwgt <list> Factor to multiply distances from each\nmetric by in a comma-separated list. Can be used to\nadjust the contribution from each metric. Default\nis 1 for each metric. Output from the metricstats\nkeyword can be used to determine the relative\ncontribution of each metric to the distance.\nPairwise Distance Matrix Options:\n[pairdist <name>] Pairwise cache DataSet/File name to\nuse for loading/saving pairwise distances.\n[pairdistfile <file>] File name to use for pairwise\ncache; if not specified and ’pairdist’\nspecified, uses ’pairdist’.\n[pwrecalc] If the loaded pairwise distance matrix does\nnot match the current setup, force recalculation.\n[loadpairdist] Load pairwise distances from file\nspecified by pairdist (CpptrajPairDist if pairdist\nnot specified).\n[savepairdist] Save pairwise distances to file specified\nby pairdist (CpptrajPairDist if pairdist not\nspecified). NOTE: If sieving was performed only the\ncalculated distances are saved.\n[pairwisecache {mem | disk | none}] Cache pairwise\ndistance data in memory (default), to disk, or\ndisable pairwise caching. No caching will save\nmemory but be extremely slow. Caching to disk will\nlikely be slow unless writing to a fast storage\ndevice (e.g. SSD) - data is saved to a file named\n’CpptrajPairwiseCache’.\nSieving Options:\n[sieve <#>] Perform clustering only for every <#>\nframe. After clustering, all other frames will be\nadded to clusters.\n[random] When sieve is specified, select initial frames\nto cluster randomly.\n[sieveseed <#>] Seed for random sieving; if not set the\nwallclock time will be used.\n[includesieved_cdist] Include sieved frames in final\ncluster distance calculation (may be very slow).\n233"}, {"page": 234, "text": "[includesieveincalc] Include sieved frames when\ncalculating within-cluster average (may be very\nslow).\n[sievetoframe] When restoring sieved frames, compare\nframe to every frame in a cluster using a cutoff of\n<restore epsilon> (default is algorithm epsilon when\nusing DPeaks/DBscan) instead of the centroid; slower\nbut more accurate.\n[sievetocentroid] When restoring sieved frames, compare\nframe to cluster centroid using a cutoff of <restore\nepsilon> (default is algorithm epsilon when using\nDPeaks/DBscan). Default method for DPeaks/DBSCAN.\n[closestcentroid] When restoring sieved frames, add each\nframe to its closest centroid. Default method for\nhieragglo/kmeans.\n[repsilon <restore epsilon>] Epsilon to use for\nsievetoframe/sievetocentroid (default is algorithm\nepsilon when using DPeaks/DBscan).\nBest Representative Options:\n[bestrep {cumulative|centroid|cumulative_nosieve}]\nMethod for choosing cluster representative frames.\ncumulative Choose by lowest cumulative distance to\nall other frames in cluster. Default when not\nsieving.\ncentroid Choose by lowest distance to cluster\ncentroid. Default when sieving.\ncumulative_nosieve Choose by lowest cumulative\ndistance to all other frames, ignoring sieved\nframes.\n[savenreps <#>] Number of best representative frames\nto choose (default 1).\nOutput Options:\n[out <cnumvtime>] Write cluster # vs frame to\n<cnumvtime>. Algorithms that calculate noise (e.g.\nDBSCAN) will assign noise points a value of -1.\n[gracecolor] Instead of cluster # vs frame, write\ncluster# + 1 (corresponding to colors used by\nXMGRACE) vs frame. Cluster #s larger than 15\nare given the same color. Algorithms that\ncalculate noise (e.g. DBSCAN) will assign noise\npoints a color of 0 (blank).\n234"}, {"page": 235, "text": "[summary <summaryfile>] Summarize each cluster with\nformat ’#Cluster Frames Frac AvgDist Stdev Centroid\nAvgCDist’:\n#Cluster Cluster number starting from 0 (0 is most\npopulated).\nFrames # of frames in cluster.\nFrac Size of cluster as fraction of total\ntrajectory.\nAvgDist Average distance between points in the\ncluster.\nStdev Standard deviation of points in the cluster.\nCentroid Frame # of structure in cluster that has\nthe lowest cumulative distance to every other\npoint. If multiple representatives are being\nsaved this column is replaced with two columns\nfor each representative, ’Rep’ (representative\nframe #) and ’RepScore’ (score according to\ncurrent best representative metric).\nAvgCDist Average distance of this cluster to every\nother cluster.\n[info <infofile>] Write ptraj-like cluster information to\n<infofile>. This file has format:\n#Clustering: <X> clusters <N> frames\n#Cluster <I> has average-distance-to-centroid <AVG>\n...\n#DBI: <DBI>\n#pSF: <PSF>\n#SSR/SST: <SSR/SST>\n#Algorithm: <algorithm-specific info>\n<Line for cluster 0>\n...\n#Representative frames: <representative frame list>\nWhere <X> is the number of clusters, <N> is the\nnumber of frames clustered, <I> ranges from 0 to\n<X>-1, <AVG> is the average distance of all frames\nin that cluster to the centroid, <DBI> is the\nDavies-Bouldin Index, <pSF> is the pseudo-F\nstatistic, <SSR/SST> is the SSR/SST ratio, and\n<representative frame list> contains the frame # of\nthe representative frame (i.e. closest to the\ncentroid) for each cluster. Each cluster has a line\nmade up of characters (one for each frame) where ’.’\nmeans ’not in cluster’ and ’X’ means ’in cluster’.\n[noinfo] Suppress printing of cluster info.\n235"}, {"page": 236, "text": "[summarysplit <splitfile>] Summarize each cluster based\non which of its frames fall in portions of the\ntrajectory specified by splitframe with format\n’#Cluster Total Frac C# Color NumInX ... FracX ...\nFirstX ... RepX’:\n#Cluster Cluster number starting from 0 (0 is most\npopulated).\nTotal # of frames in cluster.\nFrac Size of cluster as a fraction of the total\ntrajectory.\nC# Grace color number.\nColor Text description of the color (based on\nstandard XMGRACE coloring).\nNumInX Number of frames in Xth portion of the\ntrajectory.\nFracX Fraction of frames in Xth portion of the\ntrajectory.\nFirstX Frame in the Xth portion of the trajectory\nwhere the cluster is first observed.\nRepX Best representative frame in the Xth portion\nof the trajectory for that cluster.\n[splitframe <frame>] For summarysplit, frame or\ncomma-separated list of frames to split the\ntrajectory at, e.g. ’100,200,300’.\n[clustersvtime <filename>] Write number of unique\nclusters observed in a given time window to\n<filename>.\n[cvtwindow <windowsize>] Window size for\nclustersvtime output.\n[sil <prefix>] Write average cluster silhouette value for\neach cluster to ’<prefix>.cluster.dat’ and cluster\nsilhouette value for each individual frame to\n’<prefix>.frame.dat’.\n[silidx {idx|frm}] Choose what indices to write to\nthe cluster silhouette frame file: idx (the\ndefault) specifies the sorted index (starting\nfrom 0), frm specifies the actual frame number.\n[metricstats <file>] When more than one metric in use,\nprint the fraction contribution of each metric to\nthe total distance. This information can be used in\nconjunction with the wgt keyword to adjust the\ncontribution of each metric to the total distance.\nIt is written to <file> with format:\n236"}, {"page": 237, "text": "#Metric FracAv FracSD Avg SD Min Max Description\nWhere #Metric is the metric number, FracAv and\nFracSD are the average and standard devation of the\nfraction contribution of that metric to the total\ndistance (taking into account distance type and\nweights), Avg, SD, Min, and Max are the average,\nstandard devation, minimum, and maximum of the\nunmodified distance contribution from that metric,\nand Description is the metric description. This may\nbe slow for large numbers of frames, so it is\nadvisable to run this on a smaller (potentially\nsieved) number of frames.\n[cpopvtime <file> [normpop | normframe]] Write cluster\npopulation vs time to <file>; if normpop specified\nnormalize each cluster to 1.0; if normframe\nspecified normalize cluster populations by number of\nframes.\n[lifetime] Create a DataSet with aspect [Lifetime] for\neach cluster; for each frame, have 1 if the cluster\nis present and 0 otherwise. Can be used with\nlifetime analysis (12.21 on page 257).\nCoordinate Output Options:\nclusterout <trajfileprefix> Write frames in each cluster\nto files named <trajfileprefix>.cX, where X is the\ncluster number.\nclusterfmt <trajformat> Format keyword for\nclusterout (default Amber Trajectory).\nsinglerepout <trajfilename> Write all representative\nframes to single trajectory named <trajfilename>.\nsinglerepfmt <trajformat> Format keyword for\nsinglerepout (default Amber Trajectory).\nrepout <repprefix> Write representative frames to\nseparate files named <repprefix>.X.<ext>, where X is\nthe cluster number and <ext> is a format-specific\nfilename extension.\nrepfmt <trajformat> Format keyword for repout\n(default Amber Trajectory).\nrepframe Include representative frame number in\nrepout filename.\navgout <avgprefix> Write average structure for each\ncluster to separate files named <avgprefix>.X.<ext>,\nwhere X is the cluster number and <ext> is a\nformat-specific filename extension.\n237"}, {"page": 238, "text": "avgfmt <trajformat> Format keyword for avgout.\nassignrefs In summary/summarysplit, assign clusters to\nloaded reference structures if RMSD to that\nreference is less than specified cutoff. This will\nbe printed in summary and summarysplit files as 2\nextra columns: ’Name’ (reference name) and ’RMS’\n(RMS to cluster centroid).\n[refcut <rms>] RMSD cutoff in Angstroms.\n[refmask <mask>] Mask to use for RMSD calculation.\nIf not specified the default mask is all heavy\natoms.\nDataSets Created:\n<name> Cluster number vs time (color number if\ngracecolor specified).\n<name>[DBI] Hold final Davies-Bouldin index.\n<name>[PSF] Hold final pseudo-F value.\n<name>[SSRSST] Hold final SSR/SST value.\n<name>[NCVT] (clustersvtime only). Number of unique\nclusters observed over time.\n<name>[Pop]:<X> Cluster X population vs time; index\nX corresponds to cluster number.\n<name>[Lifetime]:<X> (lifetime only). For each\ncluster X, contain 1 if cluster present that frame,\n0 otherwise.\nNote cluster population vs time data sets are not generated until the analysis\nhas been run.\nCluster input frames using the specified input data sets (can be any com-\nbination of coordinates/COORDS and/or 1D scalar data) with the specified\nclustering algorithm. For COORDS sets, the distance metric can be RMSD,\nsymmetry-corrected RMSD, or DME. When multiple data sets are present, the\ntotal distance can be determined either via the Euclidean (default) or Manhat-\ntan method.\nIn order to speed up clustering of large trajectories, the sieve keyword can\nbe used. In addition, subsequent clustering calculations can be sped up by\nwriting/reading calculated pair distances between each frame to/from a file\nspecified by pairdist (or “CpptrajPairDist” if pairdist not specified).\nExample: cluster on a specific distance:\ndistance endToEnd :1 :255\ncluster data endToEnd clusters 10 epsilon 3.0 summary summary.dat info info.dat\n238"}, {"page": 239, "text": "Example: two clustering commands on the CA atoms of residues 2-10 using\naverage-linkage, stopping when either 3 clusters are reached or the minimum\ndistancebetweenclustersis4.0forthefirst,and8clustersorminimumdistance\n2.0 for the second. The first command will write the cluster number vs time\nto “cnumvtime.dat” and a summary of each cluster to “avg.summary.dat”. The\nsecond clustering command will use the pairwise distance matrix from the first\nto speed things up:\ncluster C1 :2-10 clusters 3 epsilon 4.0 info C1.info out cnumvtime.dat summary avg.summary.dat pairdist PW\ncluster C2 :2-10 clusters 8 epsilon 2.0 info C2.info pairdist PW\nClustering Success Metrics\nThe Davies-Bouldin Index (DBI, reported in the info file) measures sum over\nall clusters of the within cluster scatter to the between cluster separation; the\nsmaller the DBI, the better. The DBI is defined as the average, for all\nclusters X, of fred, where fred(X) = max, across other clusters Y, of (Cx +\nCy)/dXY. Here Cx is the average distance from points in X to the centroid,\nsimilarly Cy, and dXY is the distance between cluster centroids.\nThepseudo-Fstatistic(pSF,reportedintheinfofile)isanothermeasureof\nclustering goodness. It is intended to capture the ’tightness’ of clusters, and is\nin essence a ratio of the mean sum of squares between groups to the mean sum\nof squares within group. Higher values of pseudo-F are good. Generally,\none selects a cluster-count that gives a peak in the pseudo-f statistic. Formula:\nA/B, where A = (T - P)/(G-1), and B = P / (n-G). Here n is the number of\npoints, G is the number of clusters, T is the total distance from the all-data\ncentroid, and P is the sum (for all clusters) of the distances from the cluster\ncentroid.\nThe SSR/SST (reported in the info file) is the ratio of the sum of squares\nregression(SSRorbetweensumofsquares)andthetotalsumofsquares(SST).\nThe SSR is calculated via the sum of the squared distances of all points within\na given cluster to its centroid, and summed together for all clusters. The total\nsumofsquaresisthesumofsquareddistancesforallframestotheoverallmean.\nThe ratio lies between 0 and 1 and is supposed to give the fraction of explained\nvariance by the data. The ratio should increase with cluster count. There\nshouldbeapointatwhichaddingmoreclustersdoesnotsubstantially\nincrease SSR/SST, i.e. the point where increasing the cluster count does not\nadd new information and should not increase further.\nThe cluster silhouette (sil/silidx keywords) is a measure of how well each\npointfitswithinacluster. Valuesof1indicatethepointisverysimilartoother\npoints in the cluster, i.e. it is well-clustered. Values of -1 indicate the point\nis dissimilar and may fit better in a neighboring cluster. Values of 0 indicate\nthe point is on a border between two clusters. The sil <prefix> keyword will\nwrite two files. The first, <prefix>.cluster.dat, which has the format:\n#Cluster <Si> StdDev\n239"}, {"page": 240, "text": "Where #Cluster is the cluster number, <Si> is the average silhouette value for\nall frames in the cluster, and StdDev is the standard deviation of the silhouette\nvalueforallframesinthecluster. Thesecond,<prefix>.frame.dat,willcontain\nthe silhouette value for each frame grouped by cluster, with indices controlled\nby the silidx keyword (default sorted by ascending silhouette value), e.g.:\n#C0 Silhouette\n#C0 Silhouette\n0 -0.135988\n1 -0.0266746\n2 -0.0167628\n3 0.0609673\n4 0.0649603\n5 0.0835595\n#C1 Silhouette\n6 0.319039\n7 0.319785\n8 0.348833\n9 0.358286\n0 0.1376\n9 0.1376\nThe last two lines will contain the overall average silhouette value twice, one at\nthe lowest index and one at the highest. The file is formatted in this way to\nmake it easy to visualize each cluster silhouette relative to the average value in\ne.g. the XMGRACE plotting program. If the clustering results in one or more\ncluster with silhouette values completely below the average line, the clustering\nis likely poor.\nHints for setting DBSCAN parameters with ’kdist’\nIt is not always obvious what parameters to set for DBSCAN. You can get a\nrough idea of what to set ’mindist’ and ’epsilon’ to by generating a so-called\n\"K-dist\"plotwiththe’kidst<k>’option. TheK-distplotshowsforeachpoint\n(X axis) the Kth farthest distance (Y axis), sorted by decreasing distance. You\nsupply the same distance metric and sieve parameters you want to use for the\nactual clustering, but nothing else. For example:\ncluster C0 dbscan kdist 4 rms :1-4@CA sieve 10 loadpairdist pairdist CpptrajPairDist\nTheK-distplotwillbenamed<prefix>.<k>.dat, withthedefaultprefixbeing\n’Kdist’(inthiscasethefilenamewouldbeKdist.4.dat). TheK-distplotusually\nlookslikeacurvewithaninitiallysteepslopethatgraduallydecreases. Around\nwhere the initial part of the curve starts to flatten out (indicating an increase\nin density) is around where epsilon should be set; minpoints is set to whatever\n<k> was. It has been suggested that the shape of the K-dist curve doesn’t\nchange too much after Kdist=4, but users are encouraged to experiment.\n240"}, {"page": 241, "text": "Using ’dpeaks’ clustering\nThe ’dpeaks’ (density peaks) algorithm attempts to find clusters by identifying\npointsinhighdensityregionswhicharefarfromotherpointsofhighdensity[32].\nThere are two ways these points can be chosen. The first and recommended\nway is manually. In this method, clustering if first run with choosepoints\nnot specified to generate a plot containing density versus minimum distance\nto point with next highest density (the decision graph). Appropriate cut offs\nfor distance and density can then be chosen based on visual inspection; cutoffs\nshould be chosen so that they select points that have both a high density and\na high distance to point with next highest density. Clustering can then be run\nagain with distancecut and densitycut set.\nThe second way is automatically; cpptraj will attempt to identify outliers\nin the density vs distance plot based on distance from the running average.\nAlthough this only requires a single pass, this method of choosing points is not\nwell-tested and currently not recommended.\nThe Binary pairwise matrix file format\nWhenNetCDF isnotpresent, thepairwisematrixfile willbewritteninbinary.\nThe exact format depends on what version of cpptraj generated the file (since\nearlier versions had no concept of ’sieve’). The CpptrajPairDist file starts with\na 4 byte header containing the characters ’C’ ’T’ ’M’ followed by the version\nnumber. Aquickwaytofigureouttheversionistousethelinux’od’command\nto output the first 4 bytes as hexadecimal, e.g.:\n$ od -t x1 -N 4 CpptrajPairDist 0000000 43 54 4d 02\nSo the CpptrajPairDist file version in the above example is 2.\nThe next few numbers describe the matrix size and depend on the version.\nVersion 0: Two 4-byte integers: # of rows and # of elements.\nVersion 1: Two 8-byte unsigned integers (equivalent to size_t on most sys-\ntems): # of rows and # of elements.\nVersion 2: Three 8 byte unsigned integers: original # of rows, actual # of\nrows, and sieve value.\nThis is followed by the actual matrix data, stored as a single array of floats (4\nbytes). For versions 1 and 2 the number of elements is explicitly stored. For\nversion 2, to calculate the number of matrix elements you need to read:\nElements = (actual_rows * (actual_rows - 1)) / 2\nThe cluster pair-distance matrix is an upper-right triangle matrix without the\ndiagonal (in row-major order), so the first element is the distance between ele-\nments 0 and 1, the second is between elements 0 and 2, etc.\n241"}, {"page": 242, "text": "Inversion2files,ifthesievevalueisgreaterthan1thatmeansoriginal_rows\n> actual_rows and there is an additional array of characters original_nrows\nlong, with ’T’ if the row is being ignored (i.e. it was sieved out) and ’F’ if the\nrow is active (i.e. is active in the actual pairwise-distance matrix).\nThe code that cpptraj uses to read in CpptrajPairDist files is in ClusterMa-\ntrix::LoadFile() (ClusterMatrix.cpp).\nThe NetCDF pairwise matrix format\nThedefaultwaytowritepairwisematrixfilesasofversion6.0.0iswithNetCDF.\nThis will be set up with the following parameters:\nAttributes:\nConventions “CPPTRAJ_CMATRIX”\nVersion <version string>\nMetricDescription <description of the distance metric\nused to create matrix>\nDimensions:\nn_original_frames Number of frames originally in the\nset (i.e. before sieving).\nn_rows Number of rows in the upper-triangle pairwise\nmatrix.\nmsize Actual size of the matrix; should be (nRows_ *\n(nRows_-1)) / 2;\nVariables:\nsieve (integer, no dimension). Sieve value.\nmatrix (float, dimension msize). The pairwise matrix,\nflattened to 1 dimension. Index calc is: i1 = i +\n1; index = (((nRows_ * i) - ((i1 * i) / 2)) + j -\ni1);\nactual frames (integer, dimension n_rows). The actual\nframe numbers for which pairwise distances were\ncalculated.\n12.6 cphstats\ncphstats <pH sets> [name <name>] [statsout <statsfile>] [deprot]\n[fracplot [fracplotout <file>]]\n<pH sets> Previously read in pH data sets.\nname <name> Output set name.\nstatsout <statsfile> Write pH statistics to <statsfile>\n242"}, {"page": 243, "text": "deprot If specified, calculate fraction deprotonated\ninstead of protonated.\nfracplot If specified, calculate fraction\nprotonated/deprotonated vs pH.\nfracplotout <file> File to write fraction plots to.\nData Sets Generated\n<name>[Frac]:<idx> Fraction protonated/deprotonated\nfor residue <idx>.\nCalculate statistics for constant pH simulation data previously read in with\nreaddata (see 6.11 on page 34). Statistics are calculated for each residue at\neach input pH. Output format is as follows:\nSolvent pH is <pH>\n<res name> <res num> : Offset <off> Pred <pred> Frac Prot <frac> Transitions <#trans>\n...\nAverage total molecular protonation: <avg>\nWhere<off>isoffsetfrompredicted,<pred>ispredictedpH,and<#trans>\nisthenumberoftransitions. Alineisprintedforeachresidue. Thisfunctionality\nis similar to the cphstats utility that comes with Amber (see ?? on page ??).\nNote that data from constant pH REMD must be sorted prior to use with\ncphstats. See the readensembledata (8.23 on page 65) and sortensemble-\ndata (8.30 on page 67) commands for more details.\nFor example, to read in constant pH data from constant pH REMD, sort\nand analyze:\nreadensembledata ExplicitRemd/cpout.001 cpin ExplicitRemd/cpin name PH\nsortensembledata PH\nrunanalysis cphstats PH[*] statsout stats.dat fracplot fracplotout frac.agr deprot\n12.7 corr | correlationcoe\ncorr out <outfilename> <dataset1> [<dataset2>]\n[lagmax <lag>] [nocovar] [direct]\nout <outfilename> Write results to file named\n<outfilename>. The datasets must have the same # of\ndata points.\n<dataset1> [<dataset2>] Data set(s) to calculate\ncorrelation for. If one dataset or the same dataset\nis given twice, the auto-correlation will be\ncalculated, otherwise cross-correlation.\n[lagmax] Maximum lag to calculate for. If not specified\nall frames are used.\n243"}, {"page": 244, "text": "[nocovar] Do not calculate covariance.\n[direct] Do not use FFTs to calculate correlation; this\nwill be much slower.\nDataSet Aspects:\n[<dataset1>] (Auto-correlation) The aspect will be the\nname of each of the input data set.\n[<dataset1>-<dataset2>] (Cross-correlation) The aspect\nwill be the names of each of the input data sets\njoined by a dash (’-’).\nDataSet Aspects:\n[coeff] Correlation coefficient.\nCalculate the auto-correlation function for data set named <dataset1> or the\ncross-correlation function for data sets named <dataset1> and <dataset2> up\nto <lagmax> frames (all if lagmax not specified), writing the result to file\nspecified by out. The two datasets must have the same # of datapoints.\n12.8 crank | crankshaft\ncrank {angle | distance} <dsetname1> <dsetname2> info <string>\n[out <filename>] [results <resultsfile>]\nangle Analyze angle data sets.\ndistance Analyze distance data sets.\n<dsetname1> Data set to analyze.\n<dsetname2> Data set to analyze.\ninfo <string> Title the analysis <string>.\n[out <filename>] Write frame-vs-bin to <filename>.\n[results <resultsfile>] Write results to <resultsfile>.\nCalculate crankshaft motion between two data sets.\n12.9 crdfluct\n[crdset <crd set>] [<mask>] [out <filename>] [window <size>] [bfactor]\nCalculateatomicpositionalfluctuationsforatomsin<mask>overwindowsof\nsize<size>. Ifbfactorisspecified,thefluctuationsareweightedby 8π2(similar\n3\nbut not necessarily equivalent to crystallographic B-factor calculation). Units\nare Å, or Å2x8π2 if bfactor specified.\n3\n244"}, {"page": 245, "text": "12.10 crosscorr\ncrosscorr [name <dsetname>] <dsetarg0> [<dsetarg1> ...] [out <filename>]\n[name <dsetname>] The resulting upper-triangle matrix\nis stored with name <dsetname>.\n<dsetarg0> [<dsetarg1> ...] Argument(s) specifying\ndatasets to be used.\n[out <filename>] Write results to file named\n<filename>.\nCalculatethePearsonproduct-momentcorrelationcoefficientsbetweenallspec-\nified datasets.\n12.11 curvefit\ncurvefit <dset> { <equation> |\nname <dsname> {gauss | nexp <m> [form {mexp|mexpk|mexpk_penalty}} }\n[AX=<value> ...] [out <outfile>] [resultsout <results>]\n[maxit <max iterations>] [tol <tolerance>]\n[outxbins <NX> outxmin <xmin> outxmax <xmax>]\n<dset> Data set to fit.\n<equation> Equation to fit of form <Variable> =\n<Equation>. See 5.2 on page 25 for more details on\nequations cpptraj understands.\nname <dsname> Final data set name (required if using\nnexp or gauss).\ngauss Fit to Gaussian of form A0 * exp( -((X - A1)^2) /\n(2 * A2^2) )\nnexp <m> Fit to specified number of exponentials.\nform <type> Fit to specified exponential form:\nmexp Multi-exponential, SUM(m)[ An * exp(An+1 *\nX)]\nmexpk Multi-exponential plus constant, A0 +\nSUM(m)[An * exp(An+1 * X)]\nmexpk_penalty Same as mexpk except sum of\nprefactors constrained to 1.0 and exponential\nconstants constrained to < 0.0.\nAX=<value> Value of any constants in specified\nequation with X starting from 0 (can specify more\nthan one).\nout <outfile> Write resulting fit curve to <outfile>.\n245"}, {"page": 246, "text": "resultsout <results> Write details of the fit to\n<results> (default STDOUT).\nmaxit <max iterations> Number of iterations to run\ncurve fitting algorithm (default 50).\ntol <tolerance> Curve-fitting tolerance (default 1E-4).\noutxbins <NX> Number of points to use when generating\nfinal curve (default same number of points as input\ndata set).\noutxmin <xmin> Minimum X value to use for final curve\n(default same number of points as input data set).\noutxmax <xmax> Maximum X value to use for final\ncurve (default same number of points as input data\nset).\nPerform non-linear curve fitting for the specified data set using the Levenberg-\nMarquardt algorithm. Any equation form that cpptraj understands (see 5.2\non page 25) can be used, or several preset forms can be used. Similar to Grace\n(http://plasma-gate.weizmann.ac.il/Grace/),anequationcancontainconstants\nfor curve fitting termed AX (with X being a numerical digit, one for each con-\nstant),andisassignedtoavariablewhichthenbecomesadataset. Forexample,\nto fit a curve to data from a file named Data.dat to a data set named ’FitY’:\nreaddata Data.dat\nrunanalysis curvefit Data.dat \\\n\"FitY = (A0 * exp(X * A1)) + (A2 * exp(X * A3))\" \\\nA0=1 A1=-1 A2=1 A3=-1 \\\nout curve.dat tol 0.0001 maxit 50\nToperformthesamefitbuttoamulti-exponentialcurvewithtwoexponentials:\nreaddata Data.dat\nrunanalysis curvefit Data.dat nexp 2 name FitY \\\nA0=1 A1=-1 A2=1 A3=-1 \\\nout curve1.dat tol 0.0001 maxit 50\n12.12 diagmatrix\ndiagmatrix <matrix name> [out <filename>] [thermo [outthermo <filename>]]\n[vecs <#>] [name <modesname>] [reduce]\n[nmwiz [nmwizvecs <#>] [nmwizfile <filename>]]\n<matrix name> Name of symmetric matrix to\ndiagonalize.\n[out <filename>] Write results to <filename>.\n246"}, {"page": 247, "text": "[thermo [outthermo <filename>]] Mass-weighted\ncovariance (mwcovar) matrix only. Calculate\nentropy, heat capacity, and internal energy from the\nstructure of a molecule (average coordinates, see\nabove) and its vibrational frequencies using\nstandard statistical mechanical formulas for an\nideal gas. Results are written to <filename> if\nspecified, otherwise results are written to STDOUT.\nNote that this converts the units of the calculated\neigenvalues to frequencies (cm-1).\n[vecs <#>] Number of eigenvectors to calculate.\nDefault is 0, which is only allowed when ’thermo’ is\nspecified.\n[name <modesname>] Store resulting modes data set\nwith name <modesname>.\n[reduce] Covariance (covar/mwcovar/distcovar) matrices\nonly. For coordinate covariance (covar/mwcovar)\nmatrices, each eigenvector element is reduced via Ei\n= Eix^2 + Eiy^2 + Eiz^2. For distance covariance\n(distcovar) the eigenvectors are reduced by taking\nthe sum of the squares of each row. See Abseher &\nNilges, JMB 1998, 279, 911-920 for further details.\nThey may be used to compare results from PCA in\ndistance space with those from PCA in\ncartesian-coordinate space.\n[nmwiz] Generate output in .nmd format file for viewing\nwith NMWiz[33]. See\nhttp://prody.csb.pitt.edu/tutorials/nmwiz_tutorial/\nfor further details.\n[nmwizvecs <#>] Number of vectors to write out\nfor nmwiz output, starting with the lowest\nfrequency mode (default 20).\n[nmwizfile <filename>] Name of nmwiz file to write\nto (default ’out.nmd’).\n[nmwizmask <mask>] Mask of atoms corresponding\nto eigenvectors - should be the same one used to\ngenerate the matrix.\nCalculateeigenvectorsandeigenvaluesforthespecifiedsymmetricmatrix. This\nis followed by Principal Component Analysis (in cartesian coordinate space in\nthe case of a covariance matrix or in distance space in the case of a distance-\ncovariance matrix), or Quasiharmonic Analysis (in the case of a mass-weighted\ncovariance matrix). Diagonalization of distance, correlation, idea, and ired ma-\ntrices are also possible. Eigenvalues are given in cm−1 in the case of a mass-\nweighted covariance matrix and in the units of the matrix elements in all other\n247"}, {"page": 248, "text": "cases. In the case of a mass-weighted covariance matrix, the eigenvectors are\nmass-weighted.\nFor quasi-harmonic analysis the input must be a mass-weighted covariance\nmatrix. Thermodynamic quantities are calculated based on statistical mechan-\nicalformulaethatassumetheinputsystemisoscillatinginasingleenergywell:\nsee Statistical Thermodynamics by D. A. McQuarrie, particularly chapters 4,\n5, and 6 for more details.[34] For an in-depth discussion of the accuracy of\nthermodynamic parameters obtained via quasi-harmonic analysis see Chang et\nal..[35]\nNotethatthemaximumnumberofnon-zeroeigenvaluesobtainabledepends\non the number of frames used to generate the input matrix; the number of\nframes should be equal to or greater than the number of columns in the matrix\nin order to obtain all eigenmodes.\nResults may include average coordinates (in the case of covar, mwcovar,\ncorrel), average distances (in the case of distcovar), main diagonal elements (in\nthe case of idea and ired), eigenvalues, and eigenvectors.\nForexample,inthefollowingamass-weightedcovariancematrixofallatoms\nis generated and stored internally with the name mwcvmat; the matrix itself is\nwritten to mwcvmat.dat. Subsequently, the first 20 eigenmodes of the matrix\narecalculatedandwrittentoevecs.dat,andquasiharmonicanalysisisperformed\nat 300.0 K, with the results written to thermo.dat.\nmatrix mwcovar name mwcvmat out mwcvmat.dat\ndiagmatrix mwcvmat out evecs.dat vecs 20 \\\nthermo outthermo thermo.dat temp 300.0\nOutput Format\nThe “modes” or “evecs” output file is a text file with the following format:\n[Reduced] Eigenvector file: <Type> nmodes <#> width <width>\n<# Avg Coords> <Eigenvector Size>\n<Average Coordinates>\nWhere<Type>isastringidentifyingwhatkindofmatrixtheeigenvectors/eigenvalues\nwere determined from, nmodes is how many eigenvectors are in the file, and\n<Average Coordinates> are in lines 7 columns wide, with each element having\nwidth specified by <width>. Then for each eigenvector:\n****\n<Eigenvector#> <Eigenvalue>\n<Eigenvector Coordinates>\n...\nWhere <Eigenvector Coordinates> are in lines 7 columns wide, with each ele-\nment having width specified by <width>.\n248"}, {"page": 249, "text": "12.13 divergence\ndivergence ds1 <ds1> ds2 <ds2>\nCalculate Kullback-Leibler divergence between specified data sets.\n12.14 evalplateau\nevalplateau [name <set out name>] [tol <tol>] [valacut <valacut>]\n[initpct <initial pct>] [finalpct <final pct>]\n[chisqcut <chisqcut>] [slopecut <slopecut>] [maxit <maxit>]\n[out <outfile>] [resultsout <resultsfile>] [statsout <statsfile>]\n<input set args> ...\nname <set out name>] Name for output data sets.\ntol <tol> Curve fitting tolerance. Default 0.00001.\nvalacut <valacut> (“Value A cutoff”) Cutoff for last\nhalf average vs estimated long term value. Default\n0.01.\ninitpct <initial pct> The initial percentage of data to\nuse for the initial density guess. Default 1%.\nfinalpct <final pct> The final percentatge of data to\nuse for the final density guess. Default 50%.\nchisqcut <chisqcut> Curve fit chi-squared cutoff.\nDefualt 0.5.\nslopecut <slopecut> Final slope of fitted curve cutoff.\nDefault 0.000001.\nmaxit <maxit> Maximum number of iterations to perform\nduring curve fit.\nout <outfile> File to write data and fitted curve to.\nresultsout <resultsfile> File to write plateau results\nto.\nstatsout <statsfile> File to write curve fitting stats\nto.\n<input set args> Data sets to evaluate plateau for.\nData Sets Created\n<name>[A0] The A0 (initial density) values.\n<name>[A1] The A1 (rate constant) values.\n<name>[A2] The A2 (final density) values.\n<name>[OneA1] One over the A1 (rate constant)\nvalues.\n249"}, {"page": 250, "text": "<name>[corr] Curve fit correlation.\n<name>[vala] Difference between last half average of\ndata vs final density (A2).\n<name>[chisq] Chi-squared of the curve fit.\n<name>[pltime] Plateau time (time at which all cutoffs\nsatisfied).\n<name>[fslope] Final slope of fitted curve.\n<name>[name] Input set legend.\n<name>[result] Final result: yes, no, err (error).\nAttempttodetermineifdatahas“plateaued”,i.e. stoppedchangingsignificantly\nbytheendoftheset. Currentlythedefaultsaresetuptoevaluatedensitydata\nas part of the system preparation protocol described by Roe & Brooks.[36]\n12.15 fft\nfft <dset0> [<dset1> ...] [out <outfile>] [name <outsetname>] [dt <samp_int>]\n<dset0> [<dset1 ...] Argument(s) specifying datasets to\nbe used.\n[out <outfile>] Write results to file named <outfile>.\n[name <outsetname>] The resulting transform will be\nstored with name <outsetname>.\n[dt <samp_int>] Set the sampling interval (default is\n1.0).\nPerform fast Fourier transform (FFT) on specified data set(s). If more than 1\ndata set, they must all have the same size.\n12.16 hausdorff\nhausdorff <set arg0> [<set arg1> ...]\n[outtype {basic|trimatrix nrows <#>|fullmatrix nrows <#> [ncols <#>]}]\n[name <output set name>] [out <file>] [outab <file>] [outba <file>]\n<set arg0> ... Input matrix data set(s) to calculate\nHausdorff distance(s) for.\n[outtype] Specify the output type.\nbasic Output the Hausdorff distance for each input\nmatrix as scalar 1D data.\ntrimatrix nrows <#> Output Hausdorff distances\nfor each input matrix as a 2D upper-triangular\nmatrix with the given number of rows. Must have\n(nrows * (nrows-1)) / 2 input sets.\n250"}, {"page": 251, "text": "fullmatrix nrows <#> ncols <#> Output Hausdorff\ndistances for each input matrix as a full matrix\nwith the given number of columns and rows. If\nncols is not given, use nrows. Must have nrows\n* ncols input sets.\n[name <output set name>] Name of output data sets.\n[out <file>] File to write Hausdorff distances to.\n[outab <file>] File to write directed A->B Hausdorff\ndistances to.\n[outba <file>] File to write directed B->A Hausdorff\ndistances to.\nCalculate the symmetric Hausdorff distance for one or more matrices. The\nresultscanbesavedasanarrayorasafullorupper-triangularmatrixwiththe\nspecified dimensions. The Hausdorff distance H is determined from:\nH = max{dH(A,B), dH(B,A)]\nWhere dH(A,B) is the directed Hausdorff distance between sets A and B, etc.\nColloquially speaking, the directed Hausdorff distance between A and B is de-\ntermined as follows:\n1. Whatistheclosestapproach(distance)ofeachpointinAtoanypointin\nB?\n2. Choose the largest distance from among those distances.\nIf desired, the output can be formed into a matrix, which can be useful e.g.\nwhen doing multiple 2D rms calculations on different regions of a trajectory.\nFor example, the following input divides a 100 frame trajectory into 10 frame\nchunks, calcultes the 2D RMS matrix for each chunk, then performs Hausdorff\nanalysis on the resulting matrices and forms a full output matrix.\nparm ../DPDP.parm7\nfor beg=1;beg<100;beg+=10 end=10;end+=10 i=1;i++\nloadcrd ../DPDP.nc \\$beg \\$end name Chunk\\$i\ndone\n# Do the 2drms in chunks\nfor i=1;i<11;i++\nfor j=1;j<11;j++\n2drms crdset Chunk\\$i reftraj Chunk\\$j M\\$i.\\$j\ndone\ndone\nhausdorff M* out hausdorff.fullmatrix.gnu title hausdorff.matrix.gnu \\\nouttype fullmatrix nrows 10\nrunanalysis\n251"}, {"page": 252, "text": "Thistypeofcalculationlendsitselfwelltoparallelization. Theparallelanalysis\ncommand can be used to run all the 2drms calculations in parallel with MPI-\nenabled cpptraj:\nparm ../DPDP.parm7\nfor beg=1;beg<100;beg+=10 end=10;end+=10 i=1;i++\nloadcrd ../DPDP.nc \\$beg \\$end name Chunk\\$i\ndone\n# Do the 2drms in chunks\nfor i=1;i<11;i++\nfor j=1;j<11;j++\n2drms crdset Chunk\\$i reftraj Chunk\\$j M\\$i.\\$j\ndone\ndone\nparallelanalysis sync\nrunanalysis hausdorff M* out hausdorff.fullmatrix.gnu title hausdorff.matrix.gnu \\\nouttype fullmatrix nrows 10\n12.17 hist | histogram\nhist <dataset_name>[,<min>,<max>,<step>,<bins>] ...\n[free <temperature>] [norm | normint] [gnu] [circular] out <filename>\n[amd <amdboost_data>] [name <outputset name>]\n[traj3d <file> [trajfmt <format>] [parmout <file>]]\n[min <min>] [max <max>] [step <step>] [bins <bins>] [nativeout]\n<dataset_name>[,<min>,<max>,<step>,<bins>]\nDataset(s) to be histogrammed. Optionally, the min,\nmax, step, and/or number of bins can be specified\nfor this dimension after the dataset name separated\nby commas. It is only necessary to specify the step\nor number of bins, an asterisk ’*’ indicates the\nvalue should be calculated from available data.\n[free <temperature>] If specified, estimate free energy\n(cid:16) (cid:17)\nfrom bin populations using G i =−k B T ln N N M i ax ,\nwhere K is Boltzmann’s constant, T is the\nB\ntemperature specified by <temperature>, N is the\ni\npopulation of bin i and N is the population of\nMax\nthe most populated bin. Bins with no population are\ngiven an artificial barrier equivalent to a\npopulation of 0.5.\n[norm] If specified, normalize bin populations so the\nsum over all bins equals 1.0.\n[normint] Normalize bin populations so the integral over\nthem is 1.0.\n252"}, {"page": 253, "text": "[gnu] Internal output only; data will be\ngnuplot-readable, i.e. a space will be printed\nafter the highest order coordinate cycles.\n[circular] Internal output only; data will wrap, i.e. an\nextra bin will be printed before min and after max\nin each direction. Useful for e.g. dihedral\nangles.\nout <filename> Write results to file named <filename>.\n[amd <amdboost_data>] Reweight bins using AMD boost\nenergies in data set <amdboost_data> (in KT).\n[name <outputset name>] Output histogram data set\nname.\n[traj3d <file> [trajfmt <format>]] (3D histograms only)\nWrite a pseudo-trajectory of the 3 data sets (1\natom) to <file> with format <format>.\n[parmout <file>] (3D histograms only) Write a topology\ncorresponding to the pseudo-trajectory to <file>.\n[min <min>] Default minimum to bin if not specified.\n[max <max>] Default max to use if not specified.\n[step <step>] Default step size to use if not specified.\n[bins <bins>] Default bin size to use if not specified.\n[nativeout] Do not use cpptraj data file framework; only\nnecessary for writing out histograms with > 3\ndimensions.\nCreate an N-dimensional histrogram, where N is the number of datasets spec-\nified. For 1-dimensional histograms the xmgrace ’.agr’ file format is recom-\nmended; for 2-dimensional hisograms the gnuplot ’.gnu’ file format is recom-\nmended; for all other dimensions plot formatting is disabled and the routine\nuses its own internal output format; this is also enabled if gnu or circular is\nspecified.\nFor example, to create a two dimensional histogram of two datasets ’phi’\nand ’psi’:\ndihedral phi :2@C :3@N :3@CA :3@C\ndihedral psi :3@N :3@CA :3@C :4@N\nhist phi,-180,180,*,72 psi,-180,180,*,72 out hist.gnu\nIn this case the number of bins (72) has been specified for each dimension and\n’*’ has been given for the step size, indicating it should be calculated based on\nmin/max/bins. The following ’hist’ command is equivalent:\nhist phi psi min -180 max 180 bins 72 out hist.gnu\n253"}, {"page": 254, "text": "12.18 integrate\nintegrate <dset0> [<dset1> ...] [out <outfile>] [intout <intfile>]\n[name <name>]\n<dset0> [<dset1> ...] Data set(s) to integrate.\n[out <outfile>] If specified, write cumulative sum\ncurves to <outfile>.\n[intout <intfile>] If specified, write final integral\nvalues to <intfile>.\n[name <name>] Output data set(s) name.\nDataSets Created:\n<name> Final integral values, 1 for each input data\nset (indexed from 0).\n<name>[Sum]:<idx> Cumulative sum curves if out was\nspecified, 1 for each input data set (indexed from\n0).\nIntegrate specified data set(s) using trapezoid integration. If ’out’ is specified\nwrite cumulative sum curves to <outfile>. If ’intout’ is specified write final\nintegral values for each set to <intfile>.\n12.19 ired\nired [relax freq <MHz> [NHdist <distnh>] [noefile <noefilename>]]\n[order <order>] [orderparamfile <orderfilename>]\ntstep <tstep> tcorr <tcorr> out <filename> [norm] [drct]\nmodes <modesname> [name <output sets name>] [ds2matrix <file>]\n[relax freq <MHz>] Should only be used when ired\nvectors represent N-H bonds; calculate correlation\ntimes τ for each eigenmode and relaxation rates\nm\nand NOEs for each N-H vector. ’freq <MHz>’\n(required) is the Lamor frequency of the\nmeasurement.\n[NHdist <distnh>] Specifies the length of the NH\nbond in Angstroms (default is 1.02).\n[noefile <noefilename>] File to write the T1, T2,\nand NOE data to.\n[order <order>] Order of the Legendre polynomials to\nuse when calculating spherical harmonics (default\n2).\n254"}, {"page": 255, "text": "[orderparamfile <orderfilename>] File to write the S2\ndata to.\n[tstep <tstep>] Time between snapshots in ps (default\n1.0).\n[tcorr <tcorr>] Maximum time to calculate correlation\nfunctions for in ps (default 10000.0).\n[out <filename> Name of file to write plateau and TauM\ndata. Also the prefix for the .cmt and .cjt files\n(see below).\n[norm] Normalize all correlation functions, i.e.,\nC (t=0)=P (t=0)=1.0.\nl l\n[drct] Use the direct method to calculate correlations\ninstead of FFT; this will be much slower.\nmodes <modesname> Name of previously calculated\neigenmodes corresponding to IRED vectors.\n[name <name>] Output data set name.\n[ds2matrix <file>] If specified, write full delta*S^2\nmatrix (# IRED vector rows by # eigenmodes columns)\nto <file>.\nDataSets Created:\n<name>[S2] S2 order parameters for each vector.\n<name>[Plateau] Plateau values for each vector.\n<name>[TauM] TauM values for each vector.\n<name>[dS2] Full delta*S^2 matrix.\n<name>[T1] T1 relaxation values for each vector.\n<name>[T2] T2 relaxation values for each vector.\n<name>[NOE] NOEs for each vector.\n<name>[Cm(t)]:X Cm(t) function for vector X.\n<name>[Cj(t)]:X Cj(t) function for vector X.\nPeform IRED[17] analysis on previously defined IRED vectors (see vector ired)\nusing eigenmodes calculated from those vectors with a previous ’diagmatrix’\ncommand. The number of defined IRED vectors should match the number\nof eigenmodes calculated. Autocorrelation functions for each mode and the\ncorrespondingcorrelationtimeτ willbewrittento<filename>.cmt. Autocor-\nm\nrelation functions for each vector will be written to <filename>.cjt. Relaxation\nrates and NOEs for each N-H vector will be written to <filename> or added to\nthe the end of the standard output. For the calculation of τ the normalized\nm\ncorrelation functions and only the first third of the analyzed time steps will be\nused. For further information on the convergence of correlation functions see\n[Schneider, Brünger, Nilges, J. Mol. Biol. 285, 727 (1999)].\n255"}, {"page": 256, "text": "Example of IRED in Cpptraj\nIn cpptraj, IRED analysis[17] can now be performed in one pass (as opposed to\nthe two passes previously required in ptraj). First, IRED vectors are defined\n(in this case for N-H bonds) and an IRED matrix is calculated and analyzed.\nThe IRED vectors are then projected onto the calculated IRED eigenvectors in\nthe ired analysis command to calculate the time correlation functions. If the\nparameter order is specified, order parameters based on IRED are calculated.\nBy specifying the relax parameter, relaxation rates and NOEs can be obtained\nfor each N-H vector. Note that the order of the IRED matrix should be the\nsame as the one specified for IRED analysis.\n# Define N-H IRED vectors\nvector v0 @5 ired @6\nvector v1 @7 ired @8\n...\nvector v5 @15 ired @16\nvector v6 @17 ired @18‘\n# Define IRED matrix using all previous IRED vectors\nmatrix ired name matired order 2\n# Diagonalize IRED matrix\ndiagmatrix matired vecs 6 out ired.vec name ired.vec\n# Perform IRED analysis\nired relax NHdist 1.02 freq 500.0 tstep 1.0 tcorr 100.0 out v0.out \\\nnoefile noe order 2\n12.20 kde\nkde <dataset> [bandwidth <bw>] [out <file>] [name <dsname>]\n[min <min>] [max <max] [step <step>] [bins <bins>] [free]\n[kldiv <dsname2> [klout <outfile>]] [amd <amdboost_data>]\n[bandwidth <bw>] Bandwidth to use for KDE; if not\nspecified bandwidth will be estimated using the\nnormal distribution approximation.\n[out <file>] Output file name.\n[name <dsname>] Output data set name.\n[min <min>] Minimum bin.\n[max <max>] Maximum bin.\n[step <step>] Bin step.\n[bins <bins>] Number of bins.\n[free] Calculate free energy from bin population.\n256"}, {"page": 257, "text": "[kldiv <dsname2> [klout <outfile>]] Calculate\nKullback-Leibler divergence over time of <dataset>\ndistribution to <dsname2> distribution. Output to\n<outfile> if klout specified.\n[amd <amdboost_data>] Reweight histogram using AMD\nboost data from data set <amdboost_data> (in KT).\nHistogram 1D data set using a Gaussian kernel density estimator.\n12.21 lifetime\nlifetime [out <filename>] <dsetarg0> [ <dsetarg1> ... ]\n[window <windowsize> [name <setname>]] [averageonly]\n[cumulative] [delta] [cut <cutoff>] [greater | less] [rawcurve]\n[fuzz <fuzzcut>] [nosort]\n[out <filename>] Write results to file named\n<filename>, and lifetime curves to ’crv.<filename>’.\nIf performing windowed lifetime analysis, <filename>\ncontains the fraction present over time windows, and\n2 additional files are written: ’max.<filename>’,\ncontaining max lifetime over windows, and\n’avg.<filename>’, containing average lifetime over\nwindows.\n<dsetarg0> [<dsetarg1> ...] Argument(s) specifying\ndatasets to be used.\n[window <windowsize>] Size of window (in frames) over\nwhich to calculate lifetimes/averages. If not\nspecified lifetime/average will be calculated over\nall frames.\n[name <setname>] Store results in data sets with name\n<setname>.\n[averageonly] Just calculate averages (no lifetime\nanalysis).\n[cumulative] Calculate cumulative lifetimes/averages\nover windows.\n[delta] Calculate difference from previous window\naverage.\n[cut <cutoff>] Cutoff to use when determining if data is\n’present’ (default 0.5).\n[greater] Data is considered present when above the\ncutoff (default).\n[less] Data is considered present when below the cutoff.\n257"}, {"page": 258, "text": "[rawcurve] Do not normalize lifetime curves to 1.0.\n[fuzz <fuzzcut>] Ignore changes in lifetime state that\nare less than <fuzzcut> frames.\n[nosort] Do not sort data sets by name.\nData Sets Created:\n<setname> Number of lifetimes for each set, or if\nwindow specified fraction present over time windows.\n<setname>[max] Maximum lifetime for each set, or if\nwindow specified maximum lifetime over time windows.\n<setname>[avg] Average lifetime for each set, or if\nwindow specified average lifetime over time windows.\n<setname>[curve] Lifetime curves.\nThe following are created only if window not specified:\n<setname>[frames] Total number of frames lifetime\npresent for each set.\n<setname>[name] Name of each set.\nPerform lifetime analysis for specified data sets. Lifetime data can either be\ndetermined for the entire set, or for time windows of specified size within the\nset if window specified.\nA “lifetime” is defined as the length of time something remains ’present’;\ndata is considered present when above or below a certain cutoff (the default is\ngreater than 0.5, useful for analysis of hbond time series data). For example,\nin the case of a hydrogen bond ’series’ data set, if a hydrogen bond is present\nduring a frame the value is 1, otherwise it is 0. Given the hbond time series\ndataset{1110100011}, theoverallfractionpresentis0.6. However, there\nare 3 lifetimes of lengths 3, 1, and 2 ({1 1 1}, {1}, and {1 1}). The maximum\nlifetime is 3 and the average lifetime is 2.0, i.e. (3 + 1 + 2) / 3 lifetimes = 2.0.\nOne can also construct a “lifetime curve”, which is constructed as the sum of all\nindividual lifetimes. By default these curves are normalized to 1.0, but the raw\ncurve can be obtained using the rawcurve keyword. For the example data set\nhere the raw lifetime curve would be 3 frames long:\n1 1 1\n1\n1 1\nCurve: 3 2 1\nBydefaultdatasetsaresortedbynameunlessnosortisspecified. Thelifetime\ncommand can calculate lifetimes over specific time windows by using the win-\ndow keyword. This can be particularly useful if one wants to get a sense for\nhowlifetimesarechangingoverthecourseofverylongtimeseriesdata. Inaddi-\ntion,averagescanbecalculatedinsteadoflifetimesbyspecifyingaverageonly.\n258"}, {"page": 259, "text": "Cumulative averages over windows can be obtained using the cumulative key-\nword, or the change from the average value in the previous window can be\nobtained using the delta keyword.\nThefuzzkeywordcanbeusedtotryandsmooththeinputdatabyignoring\nchanges in state that occur for fewer frames than <fuzzcut>. For example,\nin the above example hbond time series data set there is a one frame change\nin state between the first and second lifetimes which could be interpreted as a\ntransient breaking of the hydrogen bond. Using a <fuzzcut> value of 1, this\none frame change in state would be ignored, and the data set would effectively\nappeartolifetimeas{1111100011}. Thestatechangebetweenthesecond\nandthirdlifetimesislongerthan<fuzzcut>(3frames)andsoitwouldremain.\nIfwindowisnotspecified,twofilesareoutput: <filename>andcrv.<filename>.\nThe file <filename> contains overall lifetime stats for each set with format:\n#Set <setname> <setname>[max] <setname>[avg] <setname>[frames] <setname>[name]\nwhere <setname> denotes the total number of lifetimes, <setname>[max] de-\nnotes the maximum lifetime, <setname>[avg] denotes the average lifetime,\n<setname>[frames]denotesthetotalnumberofframespresentinalllifetimes,\nand <setname>[name] is the data set name. The file crv.<filename> contains\nthe lifetime curves for each set.\nIf window isspecified,fourfilesareoutput: <filename>,max.<filename>,\navg.<filename>,andcrv.<filename>. <filename>containsthefraction“present”\nover each time window for each set, max.<filename> contains the maximum\nlifetime in each time window for each set, avg.<filename> contains the aver-\nage lifetime over each window for each set, and crv.<filename> contains the\noverall lifetime curves for each set. For window output, Gnuplot format is rec-\nommended.\nExample: hbond lifetime analysis\nparm DPDP.parm7\ntrajin DPDP.nc\nhbond HB out hbond.dat @N,H,C,O series uuseries solutehb.agr \\\navgout hbavg.dat printatomnum\n# ’run’ is used here to process the trajectory and generate hbond data\nrun\n# Perform lifetime analysis\nrunanalysis lifetime HB[solutehb] out lifehb.dat\nCalculate ion lifetimes from hbond over windows of size 100 frames:\nhbond ION out ion.dat solventdonor :WAT solventacceptor :WAT@O series\nrun\nlifetime HB[solventhb] out ion.lifetime.100.gnu window 100\n259"}, {"page": 260, "text": "12.22 lowestcurve\nlowestcurve points <# lowest> [step <stepsize>] <dset0> [<dset1> ...]\n[out <file>] [name <setname>]\n<# lowest> Number of lowest points in each bin to\naverage over.\n[step <stepsize>] Bin step size\n<dset0> [<dset1> ...] Data set(s) to use.\n[out <file>] File to write lowest curve to.\n[name <setname>] Output lowest curve set name.\nCalculate a curve of the average of the # lowest points in bins of stepsize.\nEssentially each input data set is binned over bins of stepsize, then the lowest\n<#> points are averaged over for each bin.\n12.23 meltcurve\nmeltcurve <dset0> [<dset1> ...] [out <outfile>] [name <outsetname>] cut <cut>\nCalculatemeltingcurvefrominputdatasets(i.e. fraction’folded’foreachdata\nset) assuming a simple 2-state transition model, using data below <cut>as\n’folded’ and data above <cut> as ’unfolded’.\n12.24 modes\nmodes {fluct|displ|corr|eigenval|trajout|rmsip} name <modesname> [name2 <modesname>]\n[beg <beg>] [end <end>] [bose] [factor <factor>] [calcall]\n[out <outfile>] [setname <name>]\nOptions for ’trajout’: (Generate pseudo-trajectory)\n[trajout <name> parm <name> | parmindex <#>\n[trajoutfmt <format>] [trajoutmask <mask>]\n[pcmin <pcmin>] [pcmax <pcmax>] [tmode <mode>]]\nOptions for ’corr’: (Calculate dipole correlation)\n{ maskp <mask1> <mask2> [...] | mask1 <mask> mask2 <mask> }\nparm <name> | parmindex <#>\nTypes of Calculations:\nfluct RMS fluctuations (X, Y, Z, and total) for each\natom across specified normal modes.\ndispl Displacement of cartesian coordinates in the X, Y\nand Z directions for each atom across specified\nnormal modes.\n260"}, {"page": 261, "text": "corr Dipole-dipole correlation functions. Must also\nspecify maskp (see below).\neigenval Calculate eigenvalue fractions.\ntrajout Create a pseudo-trajectory along the given mode\nfrom the average structure.\nrmsip Calculate the root-mean-square inner product\nbetween modes specified by name and name2.\nOptions:\nname <modesname> Previously read-in or generated\nModes data set name.\n[beg <beg>] [end <end>] If modes taken from datafile,\nbeginning and end modes to read. Default for beg\nis 7 (which skips the first 6 zero-frequency modes\nin the case of a normal mode analysis); for end it\nis 50.\n[bose] Use quantum (Bose) statistics in populating the\nmodes.\n[factor <factor>] multiplicative constant on the\namplitude of displacement/pseudo-trajectory, default\n1.0.\n[calcall] If specified use all eigenvectors; otherwise\neigenvectors associated with zero or negative\neigenvalues will be skipped.\n[out <outfile>] File to write data results to. If not\ngiven results are written to STDOUT.\n[setname <name>] Output data set name.\nOptions for ’trajout’:\n<name> Output trajectory file name.\n[parm <parmfile/tag>|parmindex <#>] Topology file to\nuse (default first Topology loaded).\n[trajoutfmt <format>] Output trajectory format.\n[trajoutmask <mask>] Mask of atoms that correspond to\nhow modes were originally generated.\n[pcmin <pcmin>] Lowest principal component projection\nvalue to use for output trajectory.\n[pcmax <pcmax>] Highest principal component\nprojection value to use for output trajectory.\n[tmode <mode>] Mode to generate pseudo-trajectory\nfor.\nOptions for ’corr’:\n261"}, {"page": 262, "text": "[maskp <mask1> <mask2> [...]] If corr, pairs of atom\nmasks (mask1, mask2; each pair preceded by “maskp”\nand each mask defining only a single atom) have to\nbe given that specify the atoms for which the\ncorrelation functions are desired.\nmask1 <mask> mask2 <mask> Instead of maskp,\nspecify two masks; atoms from the first mask will be\npaired up with atoms from the second mask.\nDataSets Created (fluct)\n<name>[rmsX] RMS fluctuations in the X direction.\n<name>[rmsY] RMS fluctuations in the Y direction.\n<name>[rmsZ] RMS fluctuations in the Z direction.\n<name>[rms] Total RMS fluctuations.\nDataSets Created (displ)\n<name>[displX] Displacement in X direction.\n<name>[displY] Displacement in Y direction.\n<name>[displZ] Displacement in Z direction.\nDataSets Created (eigenval)\n<name>[Frac] Fraction eigenvalue contributes to\noverall motion.\n<name>[Cumulative] Cumulative fraction.\n<name>[Eigenval] Value of eigenvlue.\nDataSets Created (rmsip)\n<name> Result of RMSIP calculation.\nAnalyze previously calculated eigenmodes obtained from principal component\nanalyses(ofcovariancematrices)orquasiharmonicanalyses(diagmatrixanalysis\ncommand). Modes are taken from a previously generated data set (i.e. from\ndiagmatrix) or read in from a data file with readdata. By default, classical\n(Boltzmann) statistics are used in populating the modes. A possible series\nof commands would be “matrix covar | mwcovar ...” to generate the matrix,\n“diagmatrix ...” to calculate the modes, and, finally, “modes ...”.\nFor example, to calculate the RMS fluctuations or displacements of the first\n3 eigenmodes caluclated from a mass-weighted covariance matrix:\nmatrix mwcovar name mwcvmat out mwcvmat.dat\ndiagmatrix mwcvmat name evecs vecs 5\nmodes fluct out rmsfluct.dat name evecs beg 1 end 3\nmodes displ out resdispl.dat name evecs beg 1 end 3\nAdditionally, dipole-dipole correlation functions for modes obtained from prin-\nciple component analysis or quasiharmonic analysis can be computed.\n262"}, {"page": 263, "text": "modes corr out cffromvec.dat name evecs beg 1 end 3 \\\nmaskp @1 @2 maskp @3 @4 maskp @5 @6\nor\nmode corr out cffromvec.dat name evecs beg 1 end 3 mask1 @1,3,5 mask2 @2,4,6\nIf eigenvalisspecified,thefractioncontributionofeacheigenvectortothetotal\nmotion is calculated and output with format:\n#Mode Frac. Cumulative Eigenval\nwhere#Modeistheeigenvectornumber,Frac. istheeigenvalueoverthesumof\nalleigenvalues,Cumulativeisthecumulativesumof Frac.,andEigenvalisthe\neigenvalueitself. Notethatinordertogetanideaforhowmucheacheigenvector\ncontributes to all motion, this is best used when all possible eigenvectors have\nbeen determined for a system.\nInordertovisualizeeigenvectors,pseudo-trajectoriesalongeigenvectorscan\nbe created using average coordinates with the trajout keyword. For example,\nto write a pseudo-trajectory of the first principal component from principal\ncomponent value of -100 to 100 for a previously calculated Modes data set\ncorresponding to heavy atoms (no hydrogens) for residues 1 to 36:\nparm ../GAAC.nowat.parm7\nreaddata evecs.dat\nrunanalysis modes name evecs.dat trajout test.nc trajoutfmt netcdf \\\ntrajoutmask :1-36&!@H= pcmin -100 pcmax 100 tmode 1\n12.25 multicurve\nmulticurve set <dset> [set <dset> ...]\n<dset> { <equation> |\nname <dsname> nexp <m> [form {mexp|mexpk|mexpk_penalty} }\n[AX=<value> ...] [out <outfile>] [resultsout <results>]\n[maxit <max iterations>] [tol <tolerance>]\n[outxbins <NX> outxmin <xmin> outxmax <xmax>]\nset <dset> [set <dset> ...] Data set(s) to fit.\n<equation> Equation to fit of form <Variable> =\n<Equation>. See 5.2 on page 25 for more details on\nequations cpptraj understands.\nname <dsname> Name of output data sets (required if\nusing nexp).\nnexp <m> Fit to specified number of exponentials.\nform <type> Fit to specified exponential form:\n263"}, {"page": 264, "text": "mexp Multi-exponential, SUM(m)[ An * exp(An+1 *\nX)]\nmexpk Multi-exponential plus constant, A0 +\nSUM(m)[An * exp(An+1 * X)]\nmexpk_penalty Same as mexpk except sum of\nprefactors constrained to 1.0 and exponential\nconstants constrained to < 0.0.\nAX=<value> Value of any constants in specified\nequation with X starting from 0 (can specify more\nthan one).\nout <outfile> Write resulting fit curve to <outfile>.\nresultsout <results> Write details of the fit to\n<results> (default STDOUT).\nmaxit <max iterations> Number of iterations to run\ncurve fitting algotrithm (default 50).\ntol <tolerance> Curve-fitting tolerance (default 1E-4).\noutxbins <NX> Number of points to use when generating\nfinal curve (default same number of points as input\ndata set).\noutxmin <xmin> Minimum X value to use for final curve\n(default same number of points as input data set).\noutxmax <xmax> Maximum X value to use for final\ncurve (default same number of points as input data\nset).\nFit each input data set <dset> to <equation>. See the curvefit command on\npage 245 for more details.\n12.26 multihist\nmultihist [out <filename>] [name <dsname>] [norm | normint] [kde]\n[min <min>] [max <max>] [step <step>] [bins <bins>] [free <T>]\n<dsetarg0> [ <dsetarg1> ... ]\nout <filename> Output file.\nname <dsname> Name for resulting histogram data\nsets.\nnorm (Only used if not kde) Normalize so that max bin\nis 1.0.\nnormint (Default for kde) Normalize integral over\nhistogram to 1.0.\n264"}, {"page": 265, "text": "kde Use kernel density estimator to construct\nhistogram.\nmin <min> Histogram minimum (default data set\nminimum).\nmax <max> Histogram maximum (default data set\nmaximum).\nstep <step> Histogram step.\nbins <bins> Number of histogram bins.\nfree <T> Calculate free energy from bin populations as\nG = -R * <T> * ln( Ni / Nmax ).\n<dsetargX> Data set argument - may specify more than\none.\nHistogram each data set separately in 1D. Must specify at least bins or step.\n12.27 phipsi\nphipsi <dsarg0> [<dsarg1> ...] resrange <range> [out <file>]\n<dsargX> Argument selecting data sets. Can specify\nmore than 1.\nresrange <range> Residue range to use (actually uses\ndata set index).\n[out <file>] Output file.\nCalculate the average and standard deviation of [phi] and [psi] data set pairs,\nwrite to <file> with format:\n#Phi Psi SD(Phi) SD(Psi) Legend\nWhere Phi is the average value of phi, Psi is the average value of psi, SD(Phi)\nis the standard deviation of phi, SD(psi) is the standard deviation of psi, and\nLegendcontainstextdescribingthephiandpsidatasetsusedinthecalculation.\nPeriodicityistakenintoaccountduringaveraging. Thedatasetsmusthavebeen\ninternallylabeledastype’phi’/’psi’andmusthaveadatasetindexset(actions\nlike dihedral and multidihedral do this automatically). For example:\nparm ../DPDP.parm7\ntrajin ../DPDP.nc\nmultidihedral DPDP phi psi\nrun\nphipsi DPDP[phi] DPDP[psi] out phipsi.dat resrange 1-22\n265"}, {"page": 266, "text": "12.28 projectdata\nprojectdata evecs <evecs dataset> [name <name>] [out <outfile>] [beg <beg>] [end <end>]\n{[dihedrals <dataset arg>] | [data <dataset arg> ...]}\nevecs <dataset name> Data set containing eigenvectors\n(modes).\n[name <name>] Output data set name.\n[out <outfile>] Write projections to <outfile>.\n[beg <beg>] First eigenvector/mode to use (default 1).\n[end <end>] Final eigenvector/mode to use (default 2).\n[dihedrals <dataset arg>] (Dihedral covariance only)\nDihedral data sets to use in projection; MUST\nCORRESPOND TO HOW EIGENVECTORS WERE GENERATED.\n[data <dataset arg>] (Data covariance only, e.g. from\nTICA). 1D data sets to use in projection; MUST\nCORRESPOND TO HOW EIGENVECTORS WERE GENERATED.\nData Sets Created:\n<name>:<#> Projection data set for mode <#>.\nProject data along previously generated eigenmodes from e.g. PCA or TICA.\nThis is a faster alternative to the projection command (11.63 on page 187) if\nonly 1D data sets need to be projected.\n12.29 regress\nregress <dset0> [<dset1> ...] [name <name>] [nx <nxvals>]\n[out <filename>] [statsout <filename>]\ndsetX Data set(s) to perform linear regression for.\nname <name> Data set name for resulting linear fits.\nnx <nxvals> Number of X values to use in output data\nset(s) (ranging from input set min to max X). If not\nspecified, input X values used.\nout <filename> File to write fit lines to.\nstatsout <filename> File to write fit statistics to.\nDataSets Generated:\n<name>:<idx> Output fit line(s) (indexed by input\nset order if more than one input set).\n<name>[slope]:<idx> Output fit line slope(s).\n<name>[intercept]:<idx> Output fit line intercept(s).\n266"}, {"page": 267, "text": "Perform linear regression on the specified data set(s). The fit line is calculated\nusingeithertheinputXvaluesor<nxvals>valuesrangingfromtheinputset\nminimum to maximum X. Statistics for the fit(s) are saved to the file specified\nby statsout or reported to STDOUT.\nFor example, to fit data read in from a file and then create a set using the\nfit parameters:\nreaddata esurf_vs_rmsd.dat.txt index 1 name XY\nrunanalysis regress XY name FitXY statsout statsout.dat\ncreateset \"Y = FitXY[slope] * X + FitXY[intercept]\" xstep .2 nx 100\nwritedata Y.dat Y\n12.30 remlog\nremlog {<remlog dataset> | <remlog filename>} [out <filename>] [crdidx | repidx]\n[stats [statsout <file>] [printtrips] [reptime <file>]] [lifetime <file>]\n[reptimeslope <n> reptimeslopeout <file>] [acceptout <file>] [name <setname>]\n[edata [edataout <file>]]\n<remlog dataset> Previously read-in REM log data.\n<remlog filename> REM log file name to read in.\n[out <filename>] Write replica/coordinate index versus\ntime to <filename>.\ncrdidx Print coordinate index vs exchange; output\nsets contain replica indices.\nrepidx Print replica index vs exchange; output sets\ncontain coordinate indices.\nstats [statsout <file>] Calculate round-trip statistics\nand optionally write to <file>.\nprinttrips Print details of each individual round trip.\n[reptime <file>] Write time spent at each replica to\n<file>.\n[lifetime <file>] Print lifetime data at each replica to\n<file>.\n[reptimeslope <n>] Calculate the slope of time spent at\neach replica every <n> exchanges.\n[reptimeslopeout <file>] File to write reptimeslope\noutput to.\n[acceptout <file>] Write overall exchange acceptances to\n<file>.\n[name <setname>] Output data set name.\n267"}, {"page": 268, "text": "[edata [edataout <file>]] Extract energy data from\nreplica log, optionally write to file.\nDataSets created:\n<setname>:<idx> Replica/coordinate index vs exchange.\n<setname>[E]:<idx> If ’edata’ specified, energy data\nfrom replica log.\nAnalyzepreviouslyreadin(viareaddata)M-REMD/T-REMD/H-REMDreplica\nlogdata. Statisticscalculatedincluderound-triptime,whichisthetimeneeded\nforacoordinatesettotravelfromthelowestreplicatothehighestandback,and\nthe number of exchanges each coordinate spent at each replica. For example,\nto read in REM log data from an Amber M-REMD run and analyze it:\nreaddata rem.log.1.save rem.log.2.save dimfile remd.dim as remlog nosearch\nremlog rem.log.1.save stats reptime mremdreptime.dat\nFor an example of remlog analysis applied to actual REMD data, see Roe et\nal.[37].\n12.31 rms2d | 2drms\nrms2d [crdset <crd set>] [<name>] [<mask>] [out <filename>]\n[{dme | nofit | srmsd | qrmsd}] [mass]\n[reftraj <traj> [parm <parmname> | parmindex <parm#>] [<refmask>]]\n[corr <corrfilename>]\n[crdset <crd set>] Name of previously generated COORDS\nDataSet. If not specified the default COORDS set\nwill be used.\n[<mask>] Mask of atoms to calculate 2D-RMSD for.\nDefault is all atoms.\n[out <filename>] Write results to <filename>.\n[dme] Calculate distance RMSD instead of coordinate\nRMSD; this is substantially slower.\n[nofit] Calculate RMSD without fitting.\n[srmsd] Calculate symmetry-corrected RMSD (see 11.84 on\npage 208).\n[qrmsd] Use quaternion RMSD calculation (can be 15-20%\nfaster).\n[mass] Mass-weight RMSD.\n[reftraj <traj>] Calculate 2D RMSD to frames in\ntrajectory <traj> instead (can also be another\nCOORDS set).\n268"}, {"page": 269, "text": "[parm <parmname> | parmindex <#>] Topology to use\nfor <traj>; only useful in conjunction with reftraj.\n[<refmask>] Mask of atoms in reference; only useful in\nconjunction with reftraj.\n[corr <corrfilename>] Calculate pseudo-auto-correlation\n(cid:80)j<N−iexp(−RMSD(j,j+i))\nC for 2D-RMSD as C(i)= j=0 ,\nN−i\nwhere i is the lag, j is the frame #, and N is the\ntotal number of frames. An exponential is used to\nweight the RMSD since 0.0 RMSD is equivalent to\ncorrelation of 1.0. This can only be done if\nreftraj is not used.\nDataSet Aspects:\n[Corr] (corr only) Pseudo-auto-correlation.\nNote: For backwards compatibility with ptraj the command ’2drms’ will also\nwork.\nCalculate the best-fit RMSD of each frame in <crd set> (the default CO-\nORDSsetifnonespecified)toeachotherframe. Thiscreatesanupper-triangle\nmatrixnamed<name>(orafullmatrixif reftrajspecified). Theoutputofthe\nrms2dcommandcanbebest-viewedusinggnuplot; agnuplot-formattedfilecan\nbeproducedbygiving<filename>a’.gnu’extension. Forexample,tocalculate\nthe RMSD of non-hydrogen atoms of each frame in trajectory “test.nc” to each\nother frame, writing to a gnuplot-viewable file “test.2drms.gnu”:\ntrajin test.nc\nrms2d !(@H=) out test.2drms.gnu\nTocalculatetheRMSDofatomsnamedCAofeachframeintrajectory“test.nc”\nto each frame in “ref.nc” (assuming test.nc and ref.nc are using the default\ntopology file):\ntrajin test.nc\nrms2d @CA out test.2drms.gnu reftraj ref.nc\n12.32 rmsavgcorr\nrmsavgcorr [crdset <crd set>] [<name>] [<mask>] [out <filename>] [mass]\n[stop <maxwindow>] [offset <offset>]\n{reference <ref file> parm <parmfile> | first}\n[crdset <crd set>] COORDS data set to use (if not\nspecified the default COORDS set will be used).\n[<name>] Output data set name.\n269"}, {"page": 270, "text": "[<mask>] Atoms to calculate RMS average correlation\nfor.\n[out <filename>] Output filename.\n[mass] Mass weight the RMSD calculation.\n[stop <maxwindow>] Only calculate RMS average\ncorrelation up to <maxwindow>.\n[offset <offset>] Skip every <offset> windows in\ncalculation.\n[first] Use first averaged frame as reference for each\nwindow (default).\n[reference <ref file> [parm <parmfile>] Use reference\nfile (with specified parm) as reference for each\nwindow.\nThe RMS average correlation[1] (RAC) is calculated as the average RMSD of\nrunning-averagedcoordinatesoverincreasingwindowsizes(orlag). Outputhas\nformat:\n<WindowSize> <RAC>\nThe first entry has a window size of 1, and so is just the average RMSD of\nall frames to the specified reference structure. The second entry has a window\nsize of two, so it is the average RMSD of all frames averaged over two adjacent\nwindows to the specified reference, and so on. The RAC will be calculated up\nto the number of frames minus 1 or the value specified by stop, whichever is\nlower. The offset can be used to speed up the calculation by skipping window\nsizes. To calculate mass-weighted RMSD specify mass. Note that to reduce\nmemory costs it can be useful to strip all coordinates not involved in the RMS\nfit from the system prior to specifying ’rmsavgcorr’. For example, to calculate\nthe correlation of C-alpha RMSD of residues 2 to 12:\nstrip !(:2-12@CA)\nrmsavgcorr out rmscorr.dat\nThe curve generated by RAC decays towards zero due to the way RAC is de-\nfined. Bythetimethe\"lag\"isN-1(whereNisthetotalnumberofframes)you\nhave only two averaged coordinates: call them Avg1 (averaged over 1 though\nN-1frames)andAvg2(averagedover2throughNframes). Barringanyextraor-\ndinary circumstances the RMSD between Avg1 and Avg2 will almost certainly\nbe quite low.\nThe RAC is a way to probe the time scales of interesting events. Any\ndeviation from a smoothly decaying curve is an indication that there are some\nsignificant structural differences occurring over that time interval. RAC curves\ncanbeparticularlyusefulwhencomparingindependentsimulationsofthesame\nsystem.\n270"}, {"page": 271, "text": "OnethingtokeepinmindthatsincetheunderlyingmetricisRMSD,itcan\nbe sensitive to the reference frame you choose. It may be useful to try looking\nat both RAC from the first frame, as well as an averaged reference frame. For\nan example of use see Galindo-Murillo et al.[38], in particular Figure 2.\n12.33 rotdif\nrotdif [outfile <outfilename>] [usefft]\nOptions for generating random vectors:\n[nvecs <nvecs>] [rvecin <randvecIn>] [rseed <random seed>]\n[rvecout <randvecOut>] [rmatrix <set name> [rmout <rmOut>]]\nOptions for calculating vector time correlation functions:\n[order <olegendre>] [ncorr <ncorr>] [corrout <corrOut>]\n*** The options below only apply if ’usefft’ IS NOT specified. ***\nOptions for calculating local effective D, small anisotropy:\n[deffout <deffOut>] [itmax <itmax>] [tol <tolerance>] [d0 <d0>]\n[nmesh <NmeshPoints>] dt <tfac> [ti <ti>] tf <tf>\nOptions for calculating D with full anisotropy:\n[amoeba_tol <tolerance>] [amoeba_itmax <iterations>]\n[amoeba_nsearch <n>] [scalesimplex <scale>] [gridsearch]\n*** The options below only apply if ’usefft’ IS specified. ***\nOptions for curve-fitting:\n[fit_tol <tolerance>] [fit_itmax <max # iterations>]\noutfile <outfilename> File to write all output from\nrotdif command to.\nOptions for generating random vectors:\nnvecs <nvecs> Number of random vectors to generate\n(default 1000).\nrvecin <randvecIn> File to read random vectors from\n(format is 1 per line, 4 columns, <#> <VX> <VY>\n<VZ>).\nrseed <random seed> Seed for random number generator\n(default 80531). Specify -1 to use wallclock time.\nrvecout <randvecOut> File to write random vectors to\n(format is 1 per line, 4 columns, <#> <VX> <VY>\n<VZ>).\nrmatrix <set name> Data set to read rotation matrices\nfrom. Rotation matrices will be used to rotate\nrandom vectors.\nrmout <rmOut> Write rotation matrices to file, 1 per\nline, frame # followed by matrix in row-major order.\nOptions for calculating vector time correlation functions:\n271"}, {"page": 272, "text": "order <olegendre> The order of Legendre polynomials to\nuse when calculating vector time correlation\nfunctions (default 2).\nncorr <ncorr> Maximum length of time correlation\nfunctions in frames. If this is not specified it\nwill be set to (tf - ti) / dt (recommended).\ncorrout <corrOut> If specified write vector time\ncorrelation functions to <corrOut>.X with format:\n<Time> <Px>\nOptions for calculating local effective D, small anisotropy:\ndeffout <deffOut> File to write out local effective\ndiffusion constants determined in the limit of small\nanisotropy.\nitmax <itmax> Maximum number of iterations to\ndetermine each local effective diffusion constant\n(small anisotropy) assuming fit to single\nexponential form (default 500).\ntol <tolerance> Tolerance for determining local\neffective diffusion constant (small anisotropy)\nassuming fit to single exponential form (default\n1E-6).\nd0 <d0> Initial guess for small anisotropy diffusion\nconstant in radians^2/ns (default 0.03).\nnmesh <NmeshPoints> Number of points per frame to\nuse when creating cubic-splined-smoothed forms of\nvector time correlation curves (default 2).\ndt <tfac> Time interval between frames (used in\nintegrating vector time correlation curves) in ns.\nti <ti> Initial time value in ns for integrating the\ntime correlation functions (default 0.0).\ntf <tf> Final time value in ns for integrating the time\ncorrelation functions. It is recommended this be\nless than the maximum simulation time since the\ntails of time correlation functions tend to be\nnoisy.\nOptions for calculating D with full anisotropy:\namoeba_tol <tolerance> Tolerance for downhill-simplex\nminimizer (default 1E-7).\namoeba_itmax <iterations> Number of iterations to run\ndownhill-simplex minimizer (default 10000).\n272"}, {"page": 273, "text": "amoeba_nsearch <n> Number of searches to perform\nwith downhill-simplex minimizer (default 1).\nscalesimplex <scale> Factor to use when scaling\nsimplexes (default 0.5).\ngridsearch If specified, perform a brute-force grid\nsearch to attempt to find a better solution for\ndiffusion tensor with full anisotropy (may be\nexpensive).\nEvaluaterotationaldiffusionpropertiesofamoleculeoveratrajectoryaccording\nto an expanded version of the procedure laid out by Wong & Case[39]. Briefly,\nrandom vectors (representing the orientation of the molecule) are rotated ac-\ncording to rotation matrices obtained from an RMS fit to a reference structure\n(typically an averaged structure). For each random vector the time correlation\nfunction of the rotated vector is calculated using Legendre polynomials of the\nspecified order. The integral over this time correlation function (which may be\nsmoothed using cubic splines to improve the integration) is then used to find\ntheeffectivediffusionconstant(D)inthelimitofsmallanisotropy. Then, using\neach calculated D, the diffusion tensor is determined with full anisotropy. Fi-\nnally, a downhill simplex minimizer is used to optimize D with full anisotropy;\n(this last step is not described in the original paper).\nRotationmatricesaregeneratedviaanRMSfittoareferencestructure(see\n11.70 on page 194). It is recommended that the RMS fit be done to an average\nstructure (see 11.9 on page 109). These rotation matrices are used to rotate\neach random vector M times (where M is the total number of frames), which\ncreates a time series for each random vector. The time correlation functions\narecalculatedforeachrandomvectortimeseriesusingLegendrepolynomialsof\nthe specified order (default 2). Calculation of time correlation functions can be\nsped up by using the OpenMP version of CPPTRAJ. The maximum length of\nthecorrelationfunction(orlag)canbespecifiedbyncorr(inframes). If ncorr\nis not specified it will be set internally based on the specified values of ti, tf,\nand dt; this is recommended. Note that if ncorr is specified it should be set to\na number less than the total number of frames since noise in time correlation\nfunctions increases as ncorr approaches the # of frames. The integration over\nthe correlation function is from ti (in whatever units are used of dt, generally\nns; 0.0 ns if not specified) to tf (same units as ti), with the time between\nframes specified by dt; the final time should be less than the total simulation\ntime (see example below). The relative size of the mesh used with cubic spline\ninterpolation for integration is controlled by nmesh (size of the mesh is ncorr\npoints*nmesh); nmesh=1meansnointerpolation, defaultis2. Notethatif\nthe integral of the correlation function for a vector is negative, that vector will\nbeskippedin subsequentcalculations(sinceit wouldimply anegative valuefor\neffective diffusion).\nThe iterative solver for effective value of the diffusion constant from the\ncorrelationfunctionsiscontrolledbyitmax,tol,andd0,whereitmaxspecifies\n273"}, {"page": 274, "text": "the number of iterations to perform (default 500), tol specifies the tolerance\n(default 1E-6), and d0 specifies the initial guess for the diffusion constant in\nradians^2 / ns (default 0.03). Effective diffusion constants for each random\nvector can be written out to a file specified by deffout. Results are printed\nto the file specified by outfile. Details on the Q and D tensors are given, as\nwell as observed and calculated tau for each random vector. First, results are\nprinted for analysis in the limit of small anisotropy. Next, results are printed\nfor analysis with full anisotropy. The results of the full anisotropic calculation\nare first given using results from the small anisotropic analysis as an initial\nguess,followedbythefinalresultsafterminimizationusingthedownhillsimplex\n(amoeba) minimizer.\nExample\nThere are two important things to keep in mind when using rotdif analysis:\n1. When calculating any kind of diffusive property it is best to simulate\nin the microcanonical (NVE) ensemble with a shorter time step and in-\ncreased SHAKE tolerance; thermostats and barostats will effect diffusion\ncalculations.\n2. Time correlation functions become noisier as the length of the function\napproachesthemaximum. Thereforeingeneraloneshouldchooseparam-\netersforthetimecorrelationfunctionthataremuchshorterthanthetotal\nsimulation length.\nForexample,givenatrajectory’mdcrd.nc’containing10000frameswithatotal\nsimulation time of 200 ns (so the time between frames is 0.02 ns), to calculate\nrotational diffusion using 100 vectors using rotation matrices generated via an\nRMSfitto’avgstruct.pdb’,computingandintegratingthetimecorrelationfunc-\ntion for each vector from 0 to 5 ns (1/40th of the simulation), and writing out\ntheeffectivediffusionconstantsandresultsto’deffs.dat’and’rotdif.out’respec-\ntively:\nreference avgstruct.pdb [avg]\nrms R0 @CA,C,N,O ref [avg] savematrices\ntrajin mdcrd.nc\nrotdif nvecs 100 rmatrix R0[RM] \\\nti 0.0 tf 5.0 dt 0.02 deffout deffs.dat \\\noutfile rotdif.out\n12.34 runningavg\nrunningavg <dset1> [<dset2> ...] [name <dsetname>] [out <filename>]\n[ [cumulative] | [window <window>] ]\n<dset1> [<dset2> ...] Data set(s) to calculate running\naverage for.\n274"}, {"page": 275, "text": "[name <dsetname>] Output running average data set\nname.\n[out <filename>] File to write results to.\n[cumulative] Calculate cumulative running average\ninstead.\n[window <window>] Size in frames of window over which\nto calculate running average.\nCalculate running average over windows of given size for data in selected data\nset(s).\n12.35 slope\nslope <dset0> [<dset1> ...] [out <outfile>] [name <name>]\n[type {forward|backward|central}]\n<dset0> [<dset1> ...] Data set(s) to calculate finite\ndifference for.\n[out <outfile>] File to write finite difference curves\nto.\n[name <name>] Output data set(s) name.\n[type {forward|backward|central}] Specify type of finite\ndifference to calculate (default forward).\nDataSets generated:\n<name>:<idx> Output finite difference curves for\neach input data set (indexed from 0).\nCalculate finite differences for each input data set.\n12.36 spline\nspline <dset0> [<dset1> ...] [out <outfile>] [meshsize <n> | meshfactor <x>]\n[meshmin <mmin>] [meshmax <mmax>]\n<dsetX> Data set(s) to perform splining on.\n[out <outfile>] Write splined data to <outfile>.\n[meshsize<n>] Size of the mesh to use for splining.\n[meshfactor <x>] If meshsize is not given, use a mesh\nof data set size * <x>.\n[meshmin <mmin>] Mesh X minimum value.\n[meshmax <mmax>] Mesh X maximum value.\nApply cubic splines to the given input data sets to create new data sets.\n275"}, {"page": 276, "text": "12.37 statistics | stat\nstat {<name> | ALL} [shift <value>] [out <filename>] [noeout <filename>]\n[ignorenv] [name <noe setname>]\n<name> Name of data set to analyze.\nALL analyze all data sets.\nshift <value> Subtract <value> from all elements in\neach data set.\n[out <filename>] Write analysis results to <filename>\n(STDOUT if not specified).\n[noeout <filename>] (Type ’noe’ only) Write summary of\nNOE results to <filename>.\n[ignorenv] (Type ’noe’ only) Ignore negative NOE\nviolations (i.e. shorter-than-expected distances).\n[name <noe setname>] (Type ’noe’ only) Name for\noutput NOE data sets.\nDataSet Aspects for type ’noe’ output:\n[R6] Averaged 1/r6distance for each set.\n[NViolations] Number of violations based on given bounds\nfor each set.\n[AvgViolation] 1/r6 averaged distance minus expected\ndistance for each set.\n[NOEnames] Name of each set.\nAnalyzeangles,dihedrals,distances,and/orpuckersandcalculatevariousprop-\nerties. Morespecificanalysescanbeobtainedbylabellingdistances/dihedrals/puckers\n(from e.g. the distance, dihedral, pucker commands or with the dataset\ncommand) with the ’type <label>’ keyword:\ndihedral type labels: alpha, beta, gamma, delta, epsilon, zeta, chi, c2p h1p,\nphi, psi, omega, pchi\ndistance type labels: noe\npucker type labels: pucker\nForeachinputdataset, theaverage, standarddeviation, initialandfinalvalues\nwill be reported. The cyclic nature of dihedral/pucker data sets is taken into\nconsideration when averaging.\n276"}, {"page": 277, "text": "12.37.1 Torsion Analysis\nA table will be written in ASCII format showing the distribution of torsion\nvaluesforeachdataset. Morespecificinformationmaybeprintedbasedonthe\nsettype. ValuesintheoutputmarkedSNBarefromthosedefinedbySchneider,\nNeidle,andBerman.[40]Formoreinformationonnucleicacidtorsionaspertains\nto RNA see further work by Schneider et al..[41]\nFor example, to perform in-depth analysis on some nucleic acid dihedral\nangles:\ndihedral g0 out dihedrals.dat :1@O5’ :1@C5’ :1@C4’ :1@C3’ type gamma\ndihedral d0 out dihedrals.dat :1@C5’ :1@C4’ :1@C3’ :1@O3’ type delta\ndihedral c0 out dihedrals.dat :1@O4’ :1@C1’ :1@N9 :1@C4 type chi\nanalyze statistics all out stat.dat\n12.37.2 Distance Analysis\nA table will be written in ASCII format showing the distribution of distance\nvalues < 6.5. If a distance is labled as ’type noe’ a compact time series will\nbe printed in ASCII format showing the NOE as strong, medium, or weak.\nIn addition the <r^-6>^(-1/6) averaged value will be reported, as well as the\nnumber of upper/lower bound violations. If ’noeout’ is specified, a summary\nof these results will be written with format:\n<#NOE> <R6> <Nviolation> <AvgViolation> <Name>\nWhere <#NOE> is an index, <R6> is the <r^-6>^(-1/6) averaged distance,\n<Nviolation> is the total number of bounds violations, <AvgViolation> is the\naverage difference from expected distance Rexp when the distance is violated\n(note that if not explicitly set, Rexp is set to the upper bound when the lower\nboundis0.0,ortheaverageofupperandlowerboundsotherwise),and<Name>\nis the data set legend.\nFor example, the following input could be used to check certain distances\nfor NOE violations:\ndistance :3@HB= :10@HG= type noe noe_medium\ndistance :3@HE= :10@HG= type noe noe_strong\ndistance :3@HA :12@HA type noe noe_medium\ndistance :3@HD= :12@HG= type noe noe_medium\ndistance :3@HE= :12@HA type noe noe_strong\nanalyze statistics all out dpdp.noe.dat noeout noe_graph.dat name Res3_NOE\n12.37.3 Pucker Analysis\nA table will be written in ASCII format showing the distribution of pucker\nphases for each data set.\n277"}, {"page": 278, "text": "12.38 ti\nti <dset0> [<dset1> ...] {nq <n quad pts> | xvals <x values>}\n[name <set name>] [out <file>] [curveout <ti curve file>]\n[nskip <#s to skip>]\n[avgincrement <#> [avgmax <#>] [avgskip <#>]]\n[bs_samples <samples> [bs_points <points>] [bs_seed <#>]\n[bs_fac <factor>]]\n<dset0> [<dset1> ...] Data set arguments specifying\ninput DV/DL values.\nnq <n quad pts> Number of points for Gaussian\nquadrature integration. Expect one data set per\npoint.\nxvals <x values> Comma-separated list of X values for\nintegration. Expect one data set per value.\nname <set name> Output data set name.\nout <file> File to write results of integration to.\ncurveout <ti curve file> File to write TI curves to.\nnskip <#s to skip> Comma separated list of number of\npoints to skip. For each number given, the TI\nintegration will be repeated.\navgincrement <#> [avgmax <#>] [avgskip <#>]\nStarting from point ’avgskip’ (default 0), repeat\nthe TI integration calculation in increments of <#>\nup to ’avgmax’ (default all points), so\n’avgincrement 10’ will do points 0-10, 0-20, etc.\nbs_samples <samples> [bs_points <points>] [bs_seed<#>] [bs_fac <factor>]\nEstimate error via bootstrap analysis, repeating the\nTI integration <samples> times using <points> points\nor <factor> times the total number of points.\nRandomize with given seed.\nDataSet Aspects:\n[TIcurve] Raw TI curve. If ’nskip’ index is number of\npoints skipped. If bootstrapping, index is sample\nindex. If ’avgincrement’ the index is the number of\npoints.\n[SD] For bootstrap analysis, standard deviation of\naverage free energy over samples.\nCalculate free energy using DV/DL energies from thermodynamic integration.\nThe results of integration of the DV/DL curve will be written to <file>, while\nthe curves themselves will be written to <ti curve file>. Use nq to specify\n278"}, {"page": 279, "text": "number of Gaussian quadrature points; otherwise the lambda values should be\nspecified by xvals, where <x values> is a comma-separated list.\nFor example, to perform Gaussian quadrature integration using data sets\nnamed ’TIdata’, repeating the calculation for various number of skipped data\npoints:\nti TIdata nq 9 name Curve out skip.agr curveout curve.agr nskip 0,5,10,15,20,30,40,50\n12.39 tica\ntica { crdset <COORDS set name> [mask <mask>] |\ndata <input set arg1> ... }\n[lag <time lag>] [map {kinetic|commute|none}]\n[name <output set name>] [out <file>] [cumvarout <file>]\ncrdset <COORDS set name> Input coordinates (COORDS)\ndata set.\nmask <mask> Selected atoms in input coordinates\n(COORDS) data set.\ndata <input set arg1> Input 1D data set name(s), may\nspecify more than once. If any data set is\nperiodic, all need to be periodic.\nlag <time lag> TICA lag time in frames.\nmap {kinetic|commute|none} How to transform the\nresulting eigenvectors.\nkinetic (default) Scale eigenvectors by eigenvalues\nso that distances in the transformed data\napproximate kinetic distances; particularly\nuseful if using the projections to cluster.\ncommute Scale eigenvectors by regularized time\nscales, sqrt(timescale_i / 2), so that distances\nin the transformed data will approximate commute\ndistances. Timescales smaller than the lag time\nare dampened.[42]\nnone Do not scale eigenvectors.\nname <output set name> Output data set name.\nout <file> File to write TICA modes to.\ncumvarout <file> File to write eigenvalue cumulative\nvariance to.\nPerformtime-independentcorrelationanalysis(TICA).Similartoprincipalcom-\nponent analysis (PCA), TICA calculates eigenvectors/eigenvalues (i.e. eigen-\nmodes) from input data sets (either coordinates or a combination of other 1D\n279"}, {"page": 280, "text": "data sets). Whereas the eigenvectors from PCA describe the variance in the\ninput data, the eigenvectors from TICA describe the maximal autocorrelation\nin the input data at the given lag time.[43] The analysis can be performed on\neither coordinates or 1D data sets; the data sets can either be all periodic (in\nwhich case they will be converted to cos/sin form) or not.\n12.40 timecorr\ntimecorr vec1 <vecname1> [vec2 <vecname2>] out <filename> [name <dsname>]\n[order <order>] [tstep <tstep>] [tcorr <tcorr>]\n[dplr] [norm] [drct] [dplrout <dplrfile>] [ptrajformat]\nvec1 <vecname1> [vec2 <vecname2>] Vector(s) on which\nto operate. By default the auto-correlation\nfunction will be calculated if one vector is\nspecified, and the cross-correlation function will\nbe calculated if two vectors are specified.\nout <filename> Name of file to write output to.\n[name <dsname>] Name of output vector data sets.\n[order <order>] Order of Legendre polynomials to use;\ndefault 2.\n[tstep <tstep>] Time between snapshots (default 1.0).\n[tcorr <tcorr>] Maximum time to calculate correlation\nfunctions for (default 10000.0).\n[dplr] Output correlation functions C ≡<P /(r(0)3r(τ)3)>\nl l\nand <1/(r(0)3r(τ)3)> in addition to the P\nl\ncorrelation function.\n[norm] Normalize all correlation functions, i.e.,\nC (t=0)=P (t=0)=1.0.\nl l\n[drct] Use the direct method to calculate correlations\ninstead of FFT; this will be much slower.\n[dplrout] (dplr only) Write extra information for each\nvector related to dplr option to <dplrfile>.\n[ptrajformat] Write output in ptraj style (prevents use\nof data formatting options).\nDataSet Aspects:\n[P] P<order> correlation function.\n[C] C<order> correlation function (dplr only).\nDataSet Aspects for dplr only:\n[R3R3] <1/(r(0)3r(t)3> correlation function.\n280"}, {"page": 281, "text": "[R] Average magnitude (<R>).\n[RRIG] Sqrt( <R^2> ).\n[R3] <1/R^3>.\n[R6] <1/R^6>.\n[Name] Vector name.\nCalculate time auto/cross-correlation functions for vectors using spherical har-\nmonics theory. NOTE: To calculate direct correlation functions for vectors just\nuse the corr analysis command. The norm keyword will normalize the result-\ning correlation functions. Note that if dplr is specified, several additional data\nsets with aspects [R], [RRIG], [R3], [R6], and [Name] will be created containing\neither 1 or 2 values depending on how many vectors were specified.\nExamples\nVectorsbetweenatoms5and6aswellas7and8arecalculatedbelow,forwhich\nauto and cross time correlation functions are obtained.\nvector v0 @5 @6\nvector v1 @7 @8\ntimecorr vec1 v0 tstep 1.0 tcorr 100.0 out v0.out order 2\ntimecorr vec1 v1 tstep 1.0 tcorr 100.0 out v1.out order 2\ntimecorr vec1 v0 vec2 v1 tstep 1.0 tcorr 100.0 out v0_v1.out order 2\nSimilarly, a vector perpendicular to the plane through atoms 18, 19, and 20 is\nobtained and further analyzed.\nvector v2 @18,@19,@20 corrplane\ntimecorr vec1 v3 tstep 1.0 tcorr 100.0 out v2.out order 2\n12.41 vectormath\nvectormath vec1 <vecname1> vec2 <vecname2> [out <filename>] [norm] [name <setname>]\n[ dotproduct | dotangle | crossproduct | magnitude | average ]\nvec1 <vecname1> vec2 <vecname2> Vector(s) on which\nto operate.\n[out <filename>] Name of file to write output to.\n[norm] Normalize the vectors; this will affect any\nsubsequent calculations with the vectors. This is\nturned on automatically if dotangle/magnitude\nspecified.\n[name <setname>] Output data set name.\n[dotproduct] (Default) Calculate the dot-product of the\ntwo vectors.\n281"}, {"page": 282, "text": "[dotangle] Calculate angle from dot-product between the\ntwo vectors; vectors will be normalized.\n[crossproduct] Calculate cross-product of the two\nvectors.\n[magnitude] Calculate the magnitude of the vectors\nselected by vec1 (no need to specify vec2).\n[average] Calculate the average (over frames) of vectors\nselected by vec1 (no need to specify vec2).\nDataSets created:\n<setname>[Dot] (double) Vector dot product (dotproduct\nonly).\n<setname>[Angle] (double) Angle in degrees from vector\ndot product (dotangle only).\n<setname>[Cross] (vector) Cross product (crossproduct\nonly).\n<setname>[Mag] (double) Magnitude (magnitude only).\n<setname>[Avg] (vector) Average vector (average only).\nCalculate dot product, angle from dot product (degrees), or cross product for\nspecified vectors. Note that norm normalizes the vectors themselves; the vec-\ntors will remain normalized for subsequent calculations or output. Either vec1\nor vec2 can be of size 1; in that case each vector in the set with N frames\noperates on the single vector. For example, if vec1 is size N and vec2 is size 1,\nthen each frame of vec1 is operated on the single vector from vec2.\nFor example, to get the angles between two previously calculated vectors v1\nand v2:\nvectormath vec1 v1 vec2 v2 dotangle out dotproduct.dat name acos(|V1|*|V2|)\n12.42 wavelet\nwavelet [crdset <set name>] nb <n scaling vals> [s0 <s0>] [ds <ds>]\n[correction <correction>] [chival <chival>] [type <wavelet>]\n[out <filename>] [name <setname>]\n[cluster [minpoints <#>] [epsilon <value>] [clusterout <file>]\n[clustermapout <file>] [cmapdetail] [kdist] [cprefix <PDB prefix>]\n[overlay <trajfile>] [overlayparm <parmfile>]]\n[crdset <set name>] COORDS data set to use\nnb <n scaling vals> Number of scales. The smaller the\nnumber the better resolution, but slower to plot.\n282"}, {"page": 283, "text": "[s0 <s0>] The smallest scale of the wavelet function\n(default 2dt where dt is time between snapshots in\nps )\n[ds <ds>] Spacing between discrete scales. (Default is\n0.25. Smaller value of ds gives finer resolution.\nThe largest values that give adequate sampling in\nscale for Morlet and Paul are 0.5 and 1.5,\nrespectively)\n[correction <correction>] The scale-to-wavelength\nparameter (1.01 for Morlet, 1.389 for Paul).\nAutomatically set based on wavelet if not otherwise\nspecified.\n[chival <chival>] The value of χ2at a particular\n2\nconfidence level\n[type <wavelet>] Type of wavelet function to use\n<morlet> or <paul>\n[out <filename>] Write results to file named <filename>\n[name <setname>] Store results in data set with name\n<setname>\n[cluster] Perform wavelet clustering i.e. wavelet\nfeature extraction analysis.\n[minpoints <#>] Minimum number of points necessary\nto form a region of interest.\n[epsilon <value>] Minimum region of interest size.\n[clusterout <file>] Output for clustering (see\nbelow).\n[clustermapout <file>] Output cluster map\n(recommended gnuplot format, see below).\n[cmapdetail] Instead of the map being smoothed to\ncluster regions, show full detail.\n[kdist] Can be used to determine minpoints and\nepsilon - see below.\n[cprefix <PDB prefix>] Output cluster region PDBs\n(only containing from minimum to maximum atom\nand minimum to maximum frame) with given prefix.\n[overlay <trajfile>] Create a trajectory that can be\noverlaid with the original trajectory to\nhighlight atoms of interest. Atoms in cluster\nregions will get their normal coordinates - all\nothers are set to the common center of mass.\n[overlayparm <parmfile>] Topology that can be used\nwith the overlay trajectory.\n283"}, {"page": 284, "text": "<wavelet>: morlet, paul\nPerform the wavelet analysis using fast Fourier transform (FFT) algorithm on\nspecifiedtrajectoryandwriteouttoagnuplot-formattedfilenamed<name.gnu>.\nThe created Wavelet map provides a clear picture of the significant motions\nwhich are characterized both in time and space. Note that typically the tra-\njectory in question should have rotational and translational movement removed\n(via e.g. the rms command); otherwise these will be reflected in the wavelet\nanalysis results.\nWavelet analysis contains two main steps which performs continues wavelet\ntransform (CWT) and statistical significance testing as proposed by Torrence\nandCompo[44]. Analysisisexecutedononedimensional(1-D)coordinatewhich\nis defined as the displacement from the starting position. For each atom, CWT\nis calculated over a specified range of scales from S up to S 2(nb−1)ds. To\n0 0\nobtain the CWT of the trajectory the Fourier transform of atom’s displace-\nment and wavelets which scaled by S ( S is calculated from: S = S 2jds; j =\n0\n0,1,2,...,nb−1) is computed and then the inverse Fourier transform of the\nproduct of Fourier transforms will be calculated as the CWT. After calculating\nthe wavelet coordinates for all atoms, a significance testing is performed to de-\ntermine the significance of each wavelet coordinate. For doing this test we need\ntohaveanappropriatebackgroundspectrumtoconsiderasameanorexpected\nspectrum and compare our wavelet coordinates against this background. In or-\nder to calculate the background spectrum since wavelet spectrum (according to\nthe convolution theorem) follows the Fourier spectrum, the Fourier coefficients\nover every atom’s displacement is calculated using the following formula and a\nmodel (µ ) is constructed on average which Fourier coefficients fit (X ) is the\nk n\ntime series which is the atom’s displacement and k is the frequency index[45].\nN−1 (cid:18) (cid:19)\n(cid:88) −2πikn\nf exp X\nk= N 1 N n\nn=0\nThistestisimplementedbasedonthenullhypothesisthattheassumptionis\nthat Fourier coordinates normally distributed around the expected value, then\nthe wavelet coordinates should also be normally distributed. Assuming the\nexpected background spectrum and since the square of a normally distributed\nvariable is chi-square distributed, then the distribution for the square of the\nabsolute values of wavelet coordinates (|W |2 is as follows (σ2is the variance\ni,k\nof the atom’s displacement).\nσ2µ χ2/2\nk 2\nThenchoosingaconfidencelevelwecandeterminetheminimumacceptable\nvalue for |W |2to be considered as a significant coordinates at that certain\ni,k\nconfidence level. In the final map the scales of only those wavelet coordinates\nwhich are significantly above the expected distribution are stored.\nFor example, to perform wavelet analysis on residues 1 to 17 with 40 scal-\ning values starting from scaling of 0.2 with a spacing of 0.25 using the Morlet\nwavelet:\n284"}, {"page": 285, "text": "parm nowat.withions.parm7\ntrajin nowat.image.nc\nrms :1-17@C*,N*,O*,P* first mass\nwavelet nb 40 s0 0.2 ds 0.25 correction 1.01 chival 1.6094 type morlet \\\n:1-17 out wavelet.gnu usemap\nWavelet Analysis Feature Extraction\nWavelet analysis feature extraction (WAFEX)[46] uses a density-based clus-\ntering algorithm (a modified version of the DBSCAN algorithm) to highlight\nphysical and temporal regions that have significant motions from wavelet map-\nsand can extract the specific atoms and frames involved in these motions for\nfurther analysis. Cluster regions shown in the map will be smoother by de-\nfault for easier visualization (unless cmapdetail is specified). Details of the\nclustering are provided via the clusterout keyword with format:\n#Cluster [points] [minatm] [maxatm] [minfrm] [maxfrm] [avgval]\n#Cluster Cluster region number.\npoints Number of points in the cluster.\nminatm Starting atom of the region.\nmaxatm End atom of the region.\nminfrm Starting frame of the region.\nmaxfrm End frame of the region.\navgval Average value of points in the region.\nFor example, to create a 2D gnuplot map highlight regions of interest called\n’cluster.gnu’ one could use the following input.\nparm ../DPDP.parm7\ntrajin ../DPDP.nc\nrms @C,CA,N first\nwavelet nb 10 s0 2 ds 0.25 type morlet correction 1.01 chival 0.25 \\\n:1-22 name DPDP \\\ncluster clustermapout cluster.gnu clusterout cluster.dat \\\nminpoints 66 epsilon 10.0\ndatafile cluster.gnu usemap palette kbvyw\nSome experimentation with kdist may be required to obtain reasonable values\nfor minpoints and epsilon. See 12.5 on page 240 as well as the Heidari et al\npaper for further discussion.\n285"}, {"page": 286, "text": "13 Analysis Examples\nPleasenotethattypicallyforprincipalcomponentanalysis(PCA)thetrajectory\nneeds to be aligned against a reference structure to remove overall global and\ntranslation motion. Use the rms command for this.\n13.1 Cartesian covariance matrix calculation and projec-\ntion (PCA)\nAftercalculatingmodes,snapshotscanbeprojectedontotheseinanadditional\npass through the trajectory. It is very important that the snapshots used when\nprojectingareexactlythesameasthoseusedtogeneratetheoriginalcovariance\nmatrix. This example takes advantage of the COORDS data set functionality\nin cpptraj to save snapshots for the purposes of projection.\n# Step one. Generate average structure.\n# RMS-Fit to first frame to remove global translation/rotation.\nparm myparm.parm7\ntrajin mytraj.nc\nrms first !@H=\naverage crdset AVG\nrun\n# Step two. RMS-Fit to average structure. Calculate covariance matrix.\n# Save the fit coordinates.\nrms ref AVG !@H=\nmatrix covar name MyMatrix !@H=\ncreatecrd CRD1\nrun\n# Step three. Diagonalize matrix.\nrunanalysis diagmatrix MyMatrix vecs 2 name MyEvecs\n# Step four. Project saved fit coordinates along eigenvectors 1 and 2\ncrdaction CRD1 projection evecs MyEvecs !@H= out project.dat beg 1 end 2\n13.2 Dihedral covariance matrix calculation and projec-\ntion for backbone phi/psi (PCA)\nparm ../1rrb_vac.prmtop\ntrajin ../1rrb_vac.mdcrd\n# Generation of phi/psi dihedral data\nmultidihedral BB phi psi resrange 2\nrun\n# Calculate dihedral covariance matrix and obtain eigenvectors\nmatrix dihcovar dihedrals BB[*] out dihcovar.dat name DIH\ndiagmatrix DIH vecs 4 out modes.dihcovar.dat name DIHMODES\nrun\n# Project along eigenvectors\n286"}, {"page": 287, "text": "projection evecs DIHMODES out dih.project.dat beg 1 end 4 dihedrals BB[*]\nrun\nReferences\n[1] Daniel R. Roe and Thomas E. Cheatham, III. PTRAJ and CPPTRAJ:\nSoftware for Processing and Analysis of Molecular Dynamics Trajectory\nData. J. Chem. Theory Comput., 9:3084–3095, 2013.\n[2] DanielR.RoeandThomasE.CheathamIII. ParallelizationofCPPTRAJ\nenableslargescaleanalysisofmoleculardynamicstrajectorydata. Journal\nof Computational Chemistry, 39(25):2110–2117, 2018.\n[3] Daniel R. Roe and Christina Bergonzo. PrepareForLeap: An Automated\nTool for Fast PDB-to-Parameter Generation. Journal of Computational\nChemistry, 43:930–935, may 2022.\n[4] Melissa E. O’Neill. Pcg: A family of simple fast space-efficient statistically\ngood algorithms for random number generation. Technical Report HMC-\nCS-2014-0905, Harvey Mudd College, Claremont, CA, September 2014.\n[5] Sebastiano Vigna. Further scramblings of Marsaglia’s xorshift generators.\nJournal of Computational and Applied Mathematics, 315:175–181, may\n2017.\n[6] Daniel R. Roe and Bernard R. Brooks. Quantifying the Effects of Lossy\nCompression on Energies Calculated from Molecular Dynamics Trajecto-\nries. Protein Science, 2022.\n[7] S. Chatterjee, P. G. Debenedetti, F. H. Stillinger, and R. M. Lynden-Bell.\nA Computational Investigation of Thermodynamics, Structure, Dynam-\nics and Solvation Behavior in Modified Water Models. J. Chem. Phys.,\n128:124511, 2008.\n[8] T. Lazaridis. Inhomogeneous Fluid Approach to Solvation Thermodynam-\nics. 1 Theory. J. Phys. Chem. B, 102:3531–3541, 1998.\n[9] C. N. Nguyen, T. Kurtzman Young, and M. K. Gilson. Grid Inhomoge-\nneous Solvation Theory: Hydration Structure and Thermodynamics of the\nMiniature Receptor cucurbit[7]uril. J. Chem. Phys., 137:044101, 2012.\n[10] LieyangChen, AnthonyCruz, DanielR.Roe, AndrewC.Simmonett, Lau-\nren Wickstrom, Nanjie Deng, and Tom Kurtzman. Thermodynamic De-\ncompositionofSolvationFreeEnergieswithParticleMeshEwaldandLong-\nRange Lennard-Jones Interactions in Grid Inhomogeneous Solvation The-\nory. Journal of Chemical Theory and Computation, page acs.jctc.0c01185,\napr 2021.\n287"}, {"page": 288, "text": "[11] W.Humphrey,A.Dalke,andK.Schulten. VMDVisualMolecularDynam-\nics. J. Molec. Graph., 14:33–38, 1996.\n[12] D. J. Sindhikara, N. Yoshida, and F. Hirata. Placevent: An Algorithm\nfor Prediction of Explicit Solvent Atom Distribution-Application to HIV-1\nProtease and F-ATP Synthase. J. Comput. Chem., 33:1536–1543, 2012.\n[13] J.J. Chou, D.A. Case, and A. Bax. Insights into the mobility of methyl-\nbearing side chains in proteins. J. Am. Chem. Soc., 125:8959–8966, 2003.\n[14] C.Perez,F.Lohr,H.Ruterjans,andJ.M.Schmidt.Self-ConsistentKarplus\nParameterizationof(3)Jcouplingsdependingonthepolypeptideside-chain\ntorsion chi(1). J. Am. Chem. Soc., 123:7081–7093, 2001.\n[15] P.H. Hunenberger, A.E. Mark, and W.F. van Gunsteren. Fluctuation\nand Cross-correlation Analysis of Protein Motions Observed in Nanosec-\nond Molecular Dynamics Simulations. J. Mol. Biol., 252:492–503, 1995.\n[16] J.J. Prompers and R. Brüschweiler. Dynamic and structural analysis of\nisotropically distributed molecular ensembles. Proteins, 46:177–189, 2002.\n[17] J.J. Prompers and R. Brüschweiler. General framework for studying the\ndynamicsoffoldedandnonfoldedproteinsbyNMRrelaxationspectroscopy\nand MD simulation. J. Am. Chem. Soc., 124:4522–4534, 2002.\n[18] M.L. Connolly. Analytical molecular surface calculation. J. Appl. Cryst.,\n16:548–558, 1983.\n[19] DoreeSitkoff,KimA.Sharp,andBarryHonig.AccurateCalculationofHy-\ndration Free Energies Using Macroscopic Solvent Models. J. Phys. Chem.,\n98:1978–1988, 1994.\n[20] XJLuandWKOlson. 3dna: asoftwarepackagefortheanalysis,rebuilding\nand visualization of three-dimensional nucleic acid structures. NUCLEIC\nACIDS RESEARCH, 31:5108–5121, 2003.\n[21] M.S. Babcock, E.P.D. Pednault, and W.K. Olson. Nucleic Acid Structure\nAnalysis. J. Mol. Biol., 237:125–156, 1994.\n[22] M. A. El Hassan and C. R. Calladine. Two distinct modes of protein-\ninduced bending in dna. J. Mol. Biol., 282:331–343, 1998.\n[23] Wilma K. Olson, Manju Bansal, Stephen K. Burley, Richard E. Dicker-\nson, Mark Gerstein, Stephen C. Harvey, Udo Heinemann, Xiang-Jun Lu,\nStephen Neidle, Zippora Shakked, Heinz Sklenar, Masashi Suzuki, Chang-\nShung Tung, Eric Westhof, Cynthia Wolberger, and Helen M. Berman. A\nstandard reference frame for the description of nucleic acid base-pair ge-\nometry. J. Mol. Biol., 313:229–237, 2001.\n288"}, {"page": 289, "text": "[24] C Altona and M Sundaralingam. Conformational analysis of the sugar\nring in nucleosides and nucleotides. a new description using the concept of\npseudorotation. J Am Chem Soc, 94:8205–8212, 1972.\n[25] SC Harvey and M Prabhakaran. Ribose puckering - structure, dynamics,\nenergetics, and the pseudorotation cycle. J Am Chem Soc, 108:6128–6136,\n1986.\n[26] DCremerandJAPople. Ageneraldefinitionofringpuckeringcoordinates.\nJ Am Chem Soc, 97:1354–1358, 1975.\n[27] W.KabschandC.Sander. Dictionaryofproteinsecondarystructure: pat-\ntern recognition of hydrogen-bonded and geometrical features. Biopoly-\nmers, 22:2577–2637, 1983.\n[28] G. Cui, J. M. Swails, and E. S. Manas. SPAM: A Simple Approach for\nProfilingBoundWaterMolecules. J. Chem. Theory Comput.,9:5539–5549,\n2013.\n[29] J.Weiser,P.S.Shenkin,andW.C.Still. ApproximateAtomicSurfacesfrom\nLinear Combinations of Pairwise Overlaps (LCPO). J. Comput. Chem.,\n20:217–230, 1999.\n[30] Daniel R. Roe and Bernard R. Brooks. Improving the Speed of Volumet-\nric Density Map Generation via Cubic Spline Interpolation. Journal of\nMolecular Graphics and Modelling, 104:107832, 2021.\n[31] M.Ester,H.Kriegel,J.Sander,andX.Xu. ADensity-BasedAlgorithmfor\nDiscovering Clusters in Large Spatial Databases with Noise. Proc. Second\nInt. Conf. Knowledge Disc. Data Mining (KDD-96), pages 226–231, 1996.\n[32] Alex Rodriguez and Alessandro Laio. Clustering by fast search and find of\ndensity peaks. Science, 344:1492–1496, 2014.\n[33] A.Bakan,L.M.Meireles,andI.Bahar. ProDy: ProteinDynamicsInferred\nfrom Theory and Experiments. Bioinformatics, 27(11):1575–1577, 2011.\n[34] DonaldA.McQuarrie. Statistical Thermodynamics. HarperandRow,New\nYork, 1973.\n[35] Chia-En Chang, Wei Chen, and Michael K. Gilson. Evaluating the Accu-\nracy of the Quasiharmonic Approximation. Journal of Chemical Theory\nand Computation, 1(5):1017–1028, 2005. PMID: 26641917.\n[36] Daniel R. Roe and Bernard R. Brooks. A protocol for preparing explicitly\nsolvated systems for stable molecular dynamics simulations. The Journal\nof Chemical Physics, 153(5):054123, aug 2020.\n289"}, {"page": 290, "text": "[37] DanielR.Roe,ChristinaBergonzo,andThomasE.CheathamIII. Evalua-\ntionofenhancedsamplingprovidedbyacceleratedmoleculardynamicswith\nhamiltonian replica exchange methods. J. Phys. Chem. B, 118(13):3543–\n3552, 2014.\n[38] Rodrigo Galindo-Murillo, Daniel R. Roe, and Thomas E. Cheatham, III.\nConvergence and reproducibility in molecular dynamics simulations of the\nDNA duplex d(GCACGAACGAACGAACGC). Biochim. Biophys. Acta,\n1850:1041–1058, 2015.\n[39] V. Wong and D.A. Case. Evaluating rotational diffusion from protein md\nsimulations. J. Phys. Chem. B, 112:6013–6024, 2008.\n[40] B. Schneider, S. Neidle, and H. M. Berman. Conformations of the\nsugar-phosphate backbone in helical dna crystal structures. Biopolymers,\n42(1):113–124, 1997.\n[41] B.Schneider, Z.Moravek, andH.M.Berman. Rnaconformationalclasses.\nNucleic Acids Res., 32(5):1666–1677, 2004.\n[42] Frank Noé, Ralf Banisch, and Cecilia Clementi. Commute maps: Separat-\ning slowly mixing molecular configurations for kinetic modeling. Journal\nof Chemical Theory and Computation, 12(11):5620–5630, 2016.\n[43] Guillermo Pérez-Hernández, Fabian Paul, Toni Giorgino, Gianni De Fab-\nritiis, and Frank Noé. Identification of slow molecular order parame-\nters for Markov model construction. The Journal of Chemical Physics,\n139(1):015102, 07 2013.\n[44] C. Torrence and G. P. Compo. A practical guide to wavelet analysis. Bull.\nAm. Meteorol. Soc., 79(1):61–78, 1998.\n[45] N. C. Benson and V. Dagget. Wavelet analysis of protein motion. Int. J.\nWavelets Multi., 10(4), 2012.\n[46] Z. Heidari, D. R. Roe, R. Galindo-Murillo, J. B. Ghasemi, and T. E.\nCheatham, III. Using Wavelet Analysis To Assist in Identification of Sig-\nnificant Events in Molecular Dynamics Simulations. J. Chem. Inf. Model.,\n56:1282–1291, 2016.\n290"}]