Forgot your password?
Overview of all scripts
The complete list of all scripts bundled with Biskit
Content
The biskit scripts are sorted into four sub-folders of biskit/scripts. Each script has a help screen which is shown if the script is called without parameters or (especially in scripts/Mod) with the -help option. This page simply collects the help screen of each script.
- scripts/Biskit - general purpose scripts
- scripts/Dock - protein-protein docking
- scripts/Mod - homology modeling
- scripts/analysis - analysis and visualisation of results
scripts/Biskit
General purpose scripts
Convenience and management
back to Content
redump.py:
Update old python pickles. Unpickle some python objects and pickle them back to the same filename. Disable strict class checking if necessary -- this allows to load classes that have meanwhile changed their base class. Usage: redump.py file1 file2 ..
relocalize.py:
Re-create LocalPath in PDBModel or Complex from the original file name (and the current environment variables). Usage: relocalize.py -i |file1 file2 ..|
killpp.py:
kill all processes containing a given phrase on all hosts Syntax: killpp.py -n |part_of_name| [-a -f -h |host1 host2 host3 ..| -e |host4 host5 ..|] -n ... search phrase -a ... kill all without asking -f ... don't ask anything at all -h ... only look on these hosts (default: all) -e ... exclude one or more hosts from killingxload.py:
xload: show osxview for several machines. Syntax: xload |machine_file| machine_file ... text file with one machine name per linecvsrm.py:
remove files from disk and cvs Syntax: cvsrm file1 file2 file3... Changes still need to be committed (cvs ci).
rm_pvm.py:
delete all /tmp/pvm* files on all hosts
bispy:
script to start the python interpreter with standard Biskit imports
replace_wildcard_import.py:
replace_wildcard_import -i |src_file| -m |module_name| [-as |import_as| -e |exclude_method1 ..| ] example: replace_wildcard_import -i tools.py -m Numeric -as N -e Complex will replace all calls to any Numeric method (except Complex) by N.|method|fix_array_constructor.py:
Fix the import of Numeric.array_constructor in pickled Biskit modules Many Biskit pickles have e.g. array_constructor dumped as PDBModel.array_constructor
echoTestRoot.py:
prints the current test root
restartPVM.py:
Restart a distributed calculation. Syntax: restartPVM.py -i |rst_file| [-a] Options: i .. restart file containing result of TrackingJobMaster.getRst() a .. add hosts to PVM
Structure sampling
back to Content
am_restartMD.py:
Prepare the restart of a broken Amber MD run. Current *crd etc. are moved to oldName_TIMEps.* and the nstlim option in the input file is set to the number of steps remaining to the end of the MD. am_restartMD.py -f |folder| [ -t0 |time_offset| -tot |nstlim_total| -rst |rst_file| -inp |inp_file| -e |exclude_files_from_renaming| ] tot - needed for 2nd restart, total number of MD steps (w/o restart) t0 - starting time in ps of this MDamber2traj.py:
Convert single amber crd into Trajectory object amber2traj.py -i sim.crd -o traj_0.dat -r ref.pdb [-b -wat -hyd -rnres -code PDBC ] -r ref.pdb - must have identical atom content as sim.crd -b traj has box info (3 additional coordinates per frame) -wat delete WAT, Cl-, Na+ residues (after parsing) -hyd delete all hydrogens (after parsing) -rnres rename amber residues HIE/HID/HIP, CYX to HIS and CYS -code PDB code of molecule (otherwise first 4 letters of ref file name)amberConcat.py:
Concatenate 2 amber crd/vel files. amberConcat.py -i sim1.crd sim2.crd -o sim_merged.crd -n n_atoms [-b -p |int_precission| -rst |last_sim1.rst last_sim2.rst| ] -n number of atoms (obligatory) -b traj has box info (3 additional coordinates) -p when looking for overlapping block, round coord. to p positions -rst try repairing last frame of sim1, sim2 from given restart file(s)amber_ensembleMD.py:
Prepare ensemble MD with amber - requires template_pme_ensemble folder Syntax: amber_ensembleMD.py -parm |parm_file| -crd |crd_file| -out |result_folder| -pdb |0_pdb_file| [ -nb_nodes |n_nodes_per_host| -template |template_folder| -nodes_eq |2 hosts for minimiz.| -nodes_prod |10 hosts for prod.| -n_members |10| -rseed |int_random_seed| -dt |production_time_step| -n_steps |production_step_number| -ntwx |production_coordinate_writing_interval| -ntwv |production_velocities_writing_interval| -place_holder1 |value| -place_holder2 |value| ..] The script prepares a new folder |out| with all the input and start files to run multiple copies of an Amber MD. All input/start files/folders are copied from a template folder. Template files and folders ending in 'xx' are recreated |n_members| times. Strings ala '%(place_holder)s' in any template file in any template folder are replaced by the value of self.|place_holder| which can be given at the command line. If self.place_holder contains a list of values, each item is only used once (e.g. nodes_prod or nodes_eq ). Requirements: -$AMBERHOME must be set -LAM environment must be set up in .cshrc or .zshenv or etc. -start_eq must be run from first host in nodes_eq.dat !amber_pdb2parm.py:
Create amber topology and coordinate files from a PDB. Syntax: am_pdb2parm.py -i |PDBfile| [-o |out.parm| ...any of options below ] OR: am_pdb2parm.py -i |PDBfile| -mirror [-o |out.parm| ] Result: out.parm, out.crd, out.pdb, (and leap.log in current folder) Special option -mirror: create a parm for exact atom content of input PDB (no S-S bonds formed, atoms missing from residues..) This parm can be used for ptraj but not for simulations! Options: ocrd - str, target file for crd (coordinates) [|f_out_base|.crd] opdb - str, target file for pdb [|f_out_base|.pdb] hetatm - keep hetero atoms [don't] cap - put ACE and NME capping residue on chain breaks [don't] capN - int int, indices of chains that should get ACE cap [] capC - int int, indices of chains that should get NME cap [] box - float, minimal distance of solute from box edge [10.0] fmod - str str, list of files with amber parameter modifications (to be loaded into leap with loadAmberParams) [] fprep - str str, list of files with amber residue definitions (to be loaded into leap with loadAmberPrep) [] leap_template - str, path to template file for leap input [use default] leaprc - str, path to parameter file for leap [use default] leap_out - str, target file for leap.log [default: discard] leap_in - str, target file for leap.in script [default: discard] leap_pdb - str, target file for cleaned input pdb [discard] leap_bin - str, path to tleap binary [use default] norun - do not run leap, only prepare files debug - keep all temporary files more -key value pairs for place holders in leap input template Comments: - The protonation state of histidines is decided from the H-atoms that are found (HD, HE, or both). After that all H are removed to be added again by leap. - Cleaning tries to convert non-standard residues to the closest standard one. - Cleaning removes non-standard atoms (and atoms following them) from standard residues. - Cleaning keeps the largest / first of multiple occupancies - Ends of chains are assumed if the residue numbering jumps backward, if there is a TER record or chain ID or segid change, or if there is a chain break. - A chain break is assumed if there is an untypical gap in the chain of back- bone atoms (see PDBModel.chainBreaks() ). - The index of the first chain is 0. - Original waters are deleted. - As usual, options can also be put into a file and loaded with the -x optionpcr_crd2pdb.py:
Extract PDB files from Xplor PCR trajectories. Syntax: pcr_crd2pdb -i |pcrFolder| -t |psfFolder| -o |outFolder| [ -n_iter |n_iterations| -skip |stepping| -z ] -z gzip crd files -skip MD-step intervall for PDBs (500 = 1/ps) -n_iter number of iterations per ensemble member (50 = 50ps)reduceTraj.py:
Reduce all-atom trajectory to trajectory with only one backbone and up to 2 side chain atoms per residue. The new atoms are the centers of mass of several atoms and carry the weight of the pooled atoms in an atom profile called 'mass'. reduceTraj.py -i traj.dat [-o traj_reduced.dat -amber -red |red_traj.dat|] i - pickled Trajectory object (all atoms) o - alternative output file name (default: 'reduced_' + input.dat) red - pickled reduced traj, just update ref model in given traj. amber - rename amber HIE/HIP/HID, CYX -> HIS, CYS; unwrap atom namesrunPcr.py:
Start single Xplor PCR job on remote host. Syntax: runPCR -t |psfFolder| -h |host| [-f |Force| -r |resultFolder| -n |nice| -i |inpFolder| ] Options: -f force constant for PCR restraint -r base folder for result (sub-folder will be created) -t folder with topology (psf, pdb) -n nice value -h host computer (accessible via ssh) -i folder with all input files, must contain restart_h2o.inp' -parm folder with param19.* files A MD folder called pcr_<PDBCode> is created. The force constant is written to a file 'oldenergy' in this folder. The topology folder is copied to the new md folder and renamed to the PDB code (which is taken from the first 4 letters of the psf file name). The job is started via ssh on the remote host. A summary of all used parameters is written to runReport.out. NOTE: The pdb/psf file has to be 4 characters long and start with a number (the segId has to conform to the same format). By using python -i runPCR.py .. a python shell remains open and the job can be killed with the command r.kill()thinTraj.py:
This script is used only for the test_multidock example and is used to remove frames from the test trajectory to speed up subsequent test steps. With he default setting of step=5 this will result in a 100 frame trajectory. thinTraj.py -i traj.dat [-step |int|] i - pickled Trajectory object step - int, 1..keep all frames, 2..skip first and every second, ..traj2ensemble.py:
Pool several trajectory objects to one ensemble trajectory. Each sub-trajectory is considered as traj of one ensemble member. This script is ignoring any profiles of the given trajectories and re-assigns new frame names ( the Trajectory.concat() method is not used to allow handling of larger trajectories ) traj2ensemble.py -i |in_traj1 in_traj2 ..| -o |out_traj| [-s |start_frame(incl.)| -e |end_frame(excl.)| -step |step| -ref |ref_pdb_or_model| -pdb |PDBCode| -prot ] s,e,step - start, end position and stepping for each of the input traject. ref - PDB or pickled PDBModel with reference coordinates, by default, the reference frame of the first trajectory is taken pdb - PDB code to be stored in trajectory prot - delete all non-protein atoms (not by default)trajAddNames.py:
Add file names of frames to trajectory trajAddNames.py -i |in_traj.| -o |out_traj.| -f |file1 file2 file..| Add/replace file names of frames to existing (pickled) Trajectory.
trajFluct.py:
trajFluct.py: Calculate global and side chain fluctuation per atom for a trajectory. Syntax: trajFluct -i trajectory_file [-o result_trajectory] Options: -i pickled trajectory -o file name for pickled result Trajectorytrajpool2ensemble.py:
Convert one normal Trajectory into EnsembleTraj. The input trajectory must have frame names that allow sorting by time and ensemble member (see EnsembleTraj.py for details). traj2ensemble.py -i |in_traj| -n |n_members| -o |out_traj| -pdb |PDBCode| ] o - out file name (default: replace input file) n - number of ensemble members to expect (default: 10) pdb - PDB code to be stored in trajectory
Biskit Setup
back to Content
setup_biskit.py:
setup the biskit environment
setup_hosts.py:
setup_hosts.py: Setup the host list neded for Biskit distributed calculations. Usage: Run this script once and let it create the empty host list in ~/.biskit/hosts.dat. Add your avaliable hosts to the list. There are three different sections to which you can choose to add a host: - own_hosts: omputers reserved for own use, highest priority - shared_hosts: computers shared with others, medium priority - others_hosts: computers mainly used by others, lowest priority Add your hosts to the corresponding 'dual' or 'single' cpu option. Separat the different hosts with a blank space. If you whish to temporarily exclude a host from being used, add it to the 'exclude' option Optional settings (noce and ram): The nice settings can be changed for a specific computer (default values are 0 for 'own' and 'shared' and 17 for 'others'). To add nice value add the host(s) and the nice value separated by a colon (:) to the 'option' 'nice'. Separate multiple hosts with a blank space. Example: computer1.local.net:12 computer2:8 computer3.local.net:5 In the same way avaliablbe RAM in GB can be added. The default values here are 0.5 for a single cpu machine and 1.0 GB for a dual cpu machine. Syntax: setup_hosts -i |list, input file| -w |str, out file| Options: -i |filename| read variable names from this file -w |filename| write variables to this file, if the file already exists it will be updated, -d |Yes| accept default values for -i and -w
Structure manipulation
back to Content
1pdb2model.py:
Syntax: 1pdb2model.py -i |file1| [-o |outfile| -psf |psf_file| -wat -amber -pdb |PDBCode| ] Result: self-sufficient pickled PDBModel or PCRModel, with itself as source Options: -i input PDB or pickled PDBModel -psf psf file name -> will generate PCRModel instead -o output file name (default: pdbfile.model) -wat skip water residues (WAT TIP3 WWW H2O) and Cl-, Na+ -amber rename CYX -> CYS, HID/HIE/HIP -> HIS -pdb pdb code to be stored in model.pdbCodeaverageASA.py:
averageASA.py - a script that collects the average (of 500 structures) molecular surface (MS) and solvent accessible surface (AS) for all 20 amino acids in a GLY-XXX-GLY tripeptide. Syntax: AverageASA.py -i |template.pdb| -r |path| [ -l |str| -mask |resmask|] Options: -i file, pdb tripeptide template file -r path, calculation root folder (many directories with be created in this folder) -l str, label for the three result dictionaries -mask residue mask to delete padding residues (i.e GLY) Result: 4 dictionaries AS, AS_sd, MS and MS_sd, written to root foldercastPdbs.py:
castPdbs: Convert two similar PDBs in two PDBs with equal atom content. PDBs must not have any HETATOMs. TIP3 residues are removed. Syntax: castPdbs.py -i1 |pdb1| -i2 |pdb2| -o1 |outFile1| -o2 |outFile2| [ -c1 |0 1 ..| -c2 |2 3 ..| ] i1, i2 file names of PDBs to be compared o1, o2 file names for result pdbs c1, c2 chain numbers (starting 0) to take from i1 and i2 (default: all)dope.py:
Syntax: dope.py -s sourceModel -i otherModels [-so sourceOut -o othersPrefix -dic old_model_dic ] Add conservation, accessibility profiles and foldX energies to a reference model and models linking to this reference. 1) if sourceOut is given: Remove waters from source, put conservation score into profiles, saveAs sourceOut 2) update each of |otherModels| from their respective source, make |sourceOut| their new source, remove atoms (shouldn't be changed from |sourceOut|) and pickle them down to same file name plus |othersPrefix| if given. 3) update old model dic if given Example 1: dope.py -s ../../rec_wet/1B39.pdb -so ../../rec_wet/dry.model \ -i *.model -dic 1B39_model.dic -> create source and update model.dic Example 2: dope.py -s ../../rec_wet/dry.model \ -i *.model -dic 1B39_model.dic -> source already there, update model.dicgetSS.py:
Count number of SS bonds in protein. Syntax: getSS.py |input1.pdb| |input_2.pdb| ..
model2pdb.py:
Convert a pickled PDBModel into a PDB file. Syntax: model2pdb.py -i |infile| -o |outfile| [ -wat -ter 0|1|2 -codeprefix ] Options: -i one or more pickled PDBModel(s) -o output file name (default: infile.pdb ) (ignored if >1 input file) -wat skip water residues (WAT TIP3 WWW H2O) and Cl-, Na+ -ter 0 don't write any TER statements -ter 1 try restoring original TER statements -ter 2 put TER between all detected chains -codeprefix add model's pdbCode entry as prefix to out file namepdb2model.py:
Syntax: pdbs2struct.py -i |file1 file2 ..| [-h |host| -c |chunk| -a -w -o |other_outFolder| -wat -s] pvm must be running on the local machine! Result: pickled PDBModel object for each pdb file with same file name but ending in '.model' Options: -h number of hosts to be used -a first add hosts to pvm -c chunk size, number of pdb's passed to each node at once -w display a xterm window for each node -o destination folder (default: same where pdb file comes from) -wat skip water residues (WAT TIP3 WWW H2O) -amber rename CYX -> CYS, HID/HIE/HIP -> HIS, unwrap atom names (this creates models with the same atom/res names as pdbs created with ambpdb -p top.parm -aatm -bres < some.crd > some.pdb ) -s sort atoms alphabetically within residuespdb2seq.py:
Extract AA sequence from PDB. Syntax pdb2seq.py |pdb_file|
pdb2traj.py:
pdb2traj.py: Collect many coordinate frames ( pdb or pickled PDBModel ) of one molecule. Write Trajectory object. Waters are removed. Syntax: pdb2traj -i pdb1 pdb2 .. [ -e -r |ref_structure| -o |out_file| -f -wat -c ] OR: pdb2traj -d folder/with/pdbs/or/models [ -r ... ] Options: -i input pdb files or pickled PDBModel objects -d folder containing input pdb or pickled PDBModel files -e create EnsembleTraj, input files must be ordered first by time then by member; x_30_10.pdb sorts before x_100_09.pdb -r reference structure with PDB records (incl. waters), if not given, the first file from -i is used -wat delete TIP3, HOH, Cl-, Na+ from ref and all frames -o file name for resulting pickled Trajectory object -f fit to reference (dry reference if given) -c analyze atom content of all files seperately before casting them to reference. Default: only analyze first file in -i. Note about reference structure: The atom order and content of the files given with -i is adapted to the order/content of the reference PDB but NOT vice-versa. Snapshots can hence have additional atoms (which are removed) but they must have, at least, all the atoms that are in the reference.pdb2xplor.py:
The xplor input file will be assembled from 3 template files. The template files should be independent of the particular PDB and should instead contain place holders which pdb2xplor will replace by actual file names, numbers, etc. Place holders look like that: %(segment_pdb)s .. means, insert value of variable segment_id as string All the variables of the Xplor class (see Xplor.__init__()) can be adressed this way. The available variables are listed in the log file. Some variables, like segment_pdb, amber_patch, segment_id will only have meaningfull values in a segment template. pdb2xplor combines the templates as follows: one header_template + (one segment template per segment) + disulfide patches (generated without template) + one tail template the most relevant variables are: for header: project_root .. root folder of cvs project for segment: segment_id .. segid of currently processed segment segment_pdb .. file name of segment pdb (generated) amber_patch .. terminal patches for amber ff (generated) for tail: pdbcode .. first 4 letters of input pdb file name outname .. suggested file name for output pdb and psf (with absolute path but w/o '.pdb' or '.psf') path .. output path (specified with option -o) For hackers: All command line options are also available as variables (e.g. i, o, t). Even more, you can invent any command line option (e.g. -temperature 298) which will then be available as variable. Taken the example you could add a place holder %(temperature)i to your template file which would be translated to 298 (or whatever you specify). For hackers++: With option -x you can specify a file containing variable - value pairs ala: temperature 298 # the temperature in K steps 100 !! minimization steps Give one pair per line, only the first 2 words are parsed.
scripts/Dock
protein-protein docking scripts -- back to Content
PCR2hex.py:
pcr2hex pool many pdb's into one seperated by MODEL/ENDMDL to be used by hex. Syntax: pcr2hex -psf |in.psf| -pdb |in1.pdb| |in2.pdb| ... [-s |modelFolder| ] -psf psf file name -pdb list of pdb file names -nos don't pickle each PDB as pickled w/o waters to this folder Result: -pdb file, pdbCode_hex.pdb (first 4 characters of the first pdb file are taken as pdbCode) -model dictionary, pdbCode_models.dic -modelFolder/in1.model, in2.model, unless -nos has been givenconcat_complexLists.py:
concat_complexLists.py -i complexes1.cl complexes2.cl -o complexes_out.cl -mo out_folder_for_changed_models -rdic correct_rec.dic -ldic correct_lig_models.diccontacter.py:
contacter: Take ComplexList, calculate contactMatrix and other stuff for all complexes on several nodes. Pickle ComplexList to a file. The result values are put into the info dict of each complex. Syntax: contacter [-i |complex_lst| -o |file_complex_lst| -c |chunk_value| -ref |ref_complex| -v |complex_version| -a -h |n_hosts| -u -f |name| -s | -n |min_nice| -all -e |host1 host2..|] Options: -i pickeled list of Complex objects (file name) -o file name for pickled complex dictionary -c chunk size (number of complexes passed to each node) -a add hosts to pvm -h number of nodes to use (default: all available) -e exclude hosts -ref pickled reference Complex for fraction of native Contacts -w show xterm for each node (default: off) -u only fill empty info fields, or missing keys from -f -f force calculation on sub-set of measures, current measures: 'fnrc_4.5', 'fnac_10', 'fnac_4.5', 'fnarc_9', 'fnarc_10', 'c_ratom_9', 'c_ratom_10', 'eProsa', 'ePairScore', 'foldX', 'cons_ent', 'cons_max', 'cons_abs' 'rms_if', 'rms_if_bb', 'xplorEnergy' -v work on a previous version of each complex, only valid if input is ComplexEvolvingList (e.g. status before and after refinement). 0..use oldest version, -1..use latest version -n renice calc to, at least, this nice value -s splits complex list into sublists of this size, dumps temporary contacted lists, collects result (can be resumed) -all allow more than 512 solutions per model pair (keep all)hex2complex.py:
hex2complex: Parse output file from hex docking run, create dictionary of Complex(es), and pickle it to a file. Creates a plot of the cluster distribution (using hex rmsd). rec, lig - receptor and ligang model dictionary hex - output file from hex o - name of resulting complex list p - create biggles plot of rmsd vs. solution mac - force rec and lig 'model 1' to be used for all Syntax: hex2complex -rec |models_rec| -lig |models_lig| -hex |hex.out| -o |output name| -p |create plot| Example: hex2complex -rec 1BZY_models.dic -lig 2AKZ_models.dic -hex 1BZY_2AKZ_hex.out -o complexes.cl -phexInput.py:
hexInput Create a macro file for hex. Syntax hexInput -r |rec pdb| -l |lig pdb| [-c |com pdb| -rm |rec model| -lm |lig model|] r, l - pdb file in hex format (single or multi model) rm, lm - model number to use, if not given perform multi model docking c - a reference complex pdb file (for rmsd output) sol - number of solutions to save Result Hex macro filehexResults.py:
hexResults Get info about docking results from one or more complexGroup files. Syntax hexResult -cg |complexGroup.cg| [ -p |plot name| -o |file name| ] Result Plot and report file
inspectComplexList.py:
Check info dict of ComplexList for missing values. Syntax: checkComplexes.py |complex_cont.cl|
multidock.py:
Seperately dock several receptor models against several ligand models. multidock.py -rdic |rec_model.dic| -ldic |lig_model.dic| [-rpdb |rec_hex.pdb| -lpdb |lig_hex.pdb| -com |refcomplex_hex.pdb| -out |outfolder| -e |excludeHost1 excludeHost2..| mac |1 or 0| -rid |A A ..| -lid |B| -soln |int|] rdic, ldic .. dict with PCRModels indexed by 1 .. n (rec, lig) rpdb, lpdb .. HEX-formatted PDB with same models (rec, lig) com .. HEX-formatted PDB with reference complex out .. folder for results (created), may contain %s for date e .. dont use these hosts mac .. 1|0 force the use of macro docking, if not given, the size of the receptor will decide if macro docking is used. rid,lid .. force these chain IDs into HEX PDB file of rec / lig soln .. number of solutions to keep from each dockingpdb2complex.py:
pdb2complex.py - create a reference Complex (without waters) Syntax: pdb2complex.py -c |complex pdb| -r |chain index| -l |chain index| -o |output name| Options: -c complex pdb file or pickled PDBModel object -r receptor chain list (e.g. 0 1 ) -l ligand ~ (e.g. 2 ) -o output file -lo,lr ligand, receptor model output filereduceComplexList.py:
Reduce macro docked complex list (rec * lib * 5120) to a normal complex list (rec * lig * 512) i - complexList to be reduced o - name of reduced list
selectModels.py:
selectModels: Select non-redundant frames from a trajectory dump them and put them into a PDB file for HEX docking. Syntax: selectModels -i |traj.dat| -o |out_folder| [ -psf |psf_file| -dic |out_model_dic| -n |number| -ref -co |TrajCluster.out| -a |atom1 atom2 atom..| -s |startFrame| -e |endFrame| -step |frameSkip| -id |chaiID] -conv |convergence_diff| ] i - pickled Trajectory object dic - alternative name for model.dic psf - create PCRModels with psf file info ref - add trajectory's reference model to dictionary and pdb if a reference pdb file is given this will be used insted id - set ligand and receptor chainID a - atoms to use for clustering, default: C and roughly every second side chain heavy atom conv - float, convergence criterium [1e-11] Result: - n pickled PDBModels '|PDBCode|_|frame|.model' in out_folder - pickled TrajCluster if requested - |PDBCode|_model.dic with these n PDBModels indexed from 1 to n
scripts/Mod
homology modeling scripts -- back to Content
align.py:
Syntax: align.py [ -o |outFolder| -log |logFile| -h |host_computer| ] Options: -o output folder for results (default: .) -log log file (default: STDOUT) -h host computer for calculation (default: local computer) -> must be accessible w/o password via ssh, check! -? or help .. this help screenalign_parallel.py:
Built multiple alignment for each project given directory (parallelised). If run from within a standardized modeling/validation folder structure, i.e from the project root where the folders templates, sequences, and validation reside all options will be set by the script. Syntax: align_parallel.py -d |list of folders| -h |hosts| [-pdb |pdbFolder| -ft |fastaTemplates| -fs |fastaSequences| -fta |fastaTarget| -fe |ferror|] Note: pvm must be running on the local machine! Options: -d [str], list of project directory (full path) -h int, number of hosts to be used -a first add hosts to pvm -pdb str, pdbFolder for the pdb *.alpha -ft str, path to 'templates.fasta' -fs str, path to 'nr.fasta' -fta str, path to 'target.fasta' -fe str, path to the error file for the AlignerMasteranalyse.py:
Syntax: analyse.py -d |main project folder| [-s |1||0] ] Result: Performing model analysis for each main project folder given. Outputs a folder 'analyse' containing: * analyse/global_results.out various data about the model, see file header. * analyse/local_results.out: residue rmsd profile to taget and mean rmsd to tagret * modeller/final.pdb: the 'best' model with the mean residue rmsd in the B-factor column Options: -d [str], list of project directory -s show the structure final.pdb im PyMolbenchmark.py:
Syntax: benchmark.py -d |list of folders| [ -modlist |model_list| -ref |reference|] Result: Performing various benchmark tasks for each folder given. A folder validation/benchmark containing: * validation/????/benchmark/Fitted_??.pdb: Benchmark model iteratively superimposed on its known structure. * validation/????/benchmark/rmsd_aa.out: All-atom rmsd of the benchamark modela. (1) without iterative fitting, (2) with iterative fitting and (3) the percentage of atoms that has been removed during the iterative fitting. * validation/????/benchmark/rmsd_ca.out: same as above, but only for C-alpha atoms * validation/????/benchmark/rmsd_res_??: gives the C-alpha rmsd for each residue. * validation/????/benchmark/PDBModels.list: pickled PYTHON list of PDBModels. Each model contains the benchmark information in the atom and residue profiles: 'rmsd_aa', 'rmsd_ca', 'rmsd_res'. See PDBModel.profile()! Options: -d [str], list of project validation directories -modlist str, the path to the 'PDBModels.list' from the project directory -ref str, the path to the 'reference.pdb' from the project directory (known structure)clean_templates.py:
Syntax: clean_templates.py [-o |output_folder| -i |chainIndex| -log |logFile| input: templates/nr/*.pdb templates/nr/chain_index.txt output: templates/t_coffee/*.alpha (input for Alignar) templates/modeller/*.pdb (input for Modeller) Options: -o output folder for results (default: .) -i chain index file for templates (default: '/templates/nr/chain_index.txt') -log log file (default: STDOUT)model.py:
Build model using Modeller. Syntax: model.py [ -o |outFolder| -log |logFile| -h |host_computer| ] Options: -o output folder for results (default: .) -log log file (default: STDOUT) -h host computer for calculation (default: local computer) -> must be accessible w/o password via ssh, check! -s show structures on Pymol superimposed on average -? or help .. this help screen input: templates/modeller/*.pdb t_coffee/final.pir_aln output: modeller/modeller.log /*.B9999000?? <- modelsmodel_parallel.py:
Syntax: model_parallel.py -d |list of folders| -h |host| [-fta |fastaTarget| -pir |f_pir| -tf |template_folder| -sm |starting_model| -em |ending_model| -fe |ferror|] pvm must be running on the local machine! Result: Parallel modelling for each project directory given Options: -d [str], project directories (default: ./validation/*) -h int, number of hosts to be used (default: 10) -fta str, path to find 'target.fasta' -pir str, alignment filename -tf str, directories for input atom files -sm int, index of the first model -em int, index of the last model -fe str, filename to output errors from the Slavemodelling_example.py:
Biskit.Mod example script that models a structure from a fasta formated sequence file in 4 steps: 1) Searches for homologe sequences and clusters the found sequences to a representative set using NCBI-Tools. 2) Searches for temptale structures for the homology modeling. Similar structures are removed by clustering. 3) Build a combined sequence/structure alignment using T-Coffee. 4) Build models using Modeller. Syntax: modelling_example.py -q |query file| -o |outputFolder| [-h |host| -log -view ] Options: -q file; fasta formated sequence file to model -o folder; directory in which all project files will be written -h host name; the quite cpu consuming stasks of aligning and modeling can be sent to a remote host that also has access to the output directory -log write stdOut messages to log file (~project/modelling.log) -view show the superimposed models in PyMol HINT: If you want to inspect the alignment used for modeling: ~project/t_coffee/final.score_htmlmodel_for_docking.py:
another modelling example
search_sequences.py:
Syntax: search_sequences.py [-q |target.fasta| -o |outFolder| -log |logFile| -db |database| -limit |max_clusters| -e |e-value-cutoff| -aln |n_alignments| -psi |psi-blast rounds| -... additional options for blastall (see SequenceSearcher.py) ] Result: folder 'sequences' with files: - blast.out - result from blast search (all alignments) - cluster_blast.out - blast alignments of cluster sequences - cluster_result.out - clustering output - all.fasta - all found sequences in fasta format - nr.fasta - clustered sequences in fasta format Options: -q fasta file with query sequence (default: ./target.fasta) -o output folder for results (default: .) -log log file (default: STDOUT) -db sequence data base -limit Largest number of clusters allowed -e E-value cutoff for sequence search -aln number of alignments to be returned -simcut similarity threshold for blastclust (score < 3 or % identity) -simlen length threshold for clustering -ncpu number of CPUs for clustering -psi int, use PSI Blast with specified number of iterationssearch_templates.py:
Syntax: search_templates.py [-q |target.fasta| -o |outFolder| -log |logFile| -db |database| -e |e-value-cutoff| -limit |max_clusters| -aln |n_alignments| -psi -... additional options for blastall (see SequenceSearcher.py) ] Options: -q fasta file with query sequence (default: ./target.fasta) -o output folder for results (default: .) -log log file (default: STDOUT) -db sequence data base -limit Largest number of clusters allowed -e E-value cutoff for sequence search -aln number of alignments to be returned -simcut similarity threshold for blastclust (score < 3 or % identity) -simlen length threshold for clustering -ncpu number of CPUs for clustering -psi use PSI Blast instead, experimental!!setup_validation.py:
Setup the cross-validation folder for one or several projects Syntax: setup_validation.py [ -o |project folder(s)| ] Options: -o .. one or several project folders (default: current) -? or -help .. this help screen
scripts/analysis
scripts for analysis and visualisation of results -- back to Content
a_baharEntropy.py:
a_baharEntropy.py -i |com_folder1 com_folder2 com_folder3..| com_folder must contain com_wet/dry_com.model, rec_wet/dry_rec.model, lig_wet/dry_lig.model and analysis_500-5.opt for -cl and -cr option E.g: a_baharEntropy.py -i c11 c12 c13 > result.txt 2> log.txta_baharFluct.py:
no documentation
a_cad.py:
CAD (contact area difference) calculation by icmbrowser. The calculation is performed only for residues that are in contact in the reference complex. cl - complexList, has to contain info dictionary data for key ref - reference complexa_comEntropy.py:
Run many AmberEntropist calculations on many nodes. The Master has a standard set of 13 protocols to run on rec, lig, and com trajectories, as well as on every single member trajectory - in total 113. It accepts one variable parameter, e.g. s(tart). Each protocol is then run for all values of the variable parameter. The script puts many temporary trajectories into the folder with the input trajectories -- consider creating a new folder for each trajectory! Syntax: a_comEntropy.py -rec |rec.traj| -lig |lig.traj| -com |com.traj| -out |out.dat| [ -cr |rec_chains| -zfilter |cutoff| -s |from| -e |to| -ss |from| -se |to| -thin |fraction| -step |offset| -var |option| -vrange |v1 v2..| -jack -exrec |members| -exlig |members| -excom |members| -hosts |name1 name2..| -clean -single ] Options: rec - str, free rec trajectory lig - str, free lig trajectory com - str, complex trajectory out - str, file name for pickled result cr - [int], chains of receptor in complex trajectory [n_chains rec] var - str, name of variable option [ s ] vrange - [any], set of values used for variable option OR 'start:stop:step' i.e string convertable to arange() input jack - set up leave-one-trajectory-out jackknife test [don't] (replaces var with 'ex1' and vrange with range(1,n_members+1)) zfilter- float, kick out outlyer trajectories using z-score threshold [None->don't] exrec - [int], exclude certain members of receptor ensemble [[]] exlig - [int], exclude certain members of ligand ensemble [[]] excom - [int], exclude certain members of complex ensemble [[]] clean - remove pickled ref models and member trajectories [0] hosts - [str], nodes to be used [all known] h - int, number of nodes to be used from all known [all] single - run only one job on multi-processor nodes [0] mem - float, run only on machines with more than |mem| GB RAM [0] debug - don't delete output files [0] ... parameters for AmberEntropist -- can also be given as -var cast - equalize free and bound atom content [1] s,e - int, start and stop frame [0, to end] ss, se - int, start and stop frame of single member trajectories (only works with EnsembleTraj; overrides s,e) atoms - [ str ], names of atoms to consider [all] heavy - delete all hydrogens [don't] step - int, frame offset [no offset] thin - float, use randomly distributed fraction of frames [all] (similar to step but sometimes better) all - only calculate with all members, no single member values ex - [int] exclude same members from rec, lig and com ex_n - int, exclude last n members OR... [0] ex3 - int, exclude |ex3|rd tripple of trajectories [0] (0 excludes nothing, 1 excludes [0,1,2] ) ex1 - int, exclude ex1-th member remaining after applying ex [None] (0 excludes nothing, 1 excludes [0] ) ... parameters for AmberCrdEntropist, Executor, Master f_template - str, alternative ptraj input template [default] verbose - print progress messages to log [log != STDOUT] w - show x-windows [no] a - 0|1, add hosts to PVM [1]a_compare_rms_vs_fnc.py:
a_compare_rms_vs_fnc.py: Plot interface rmsd (heavy and/or backbone) vs. fraction native atom/residue contacts at different cutoffs. creates up to 4 plots: rms_if_vs_cont.eps rms_if_bb_vs_cont.eps rms_if_bb_vs_rms_if.eps rms_hex_vs_rms_if.eps Syntax: -i complexes_cont.cl -o str, output directory -v [str], list of keys to plot -if 1||0 create plot of key vs. interface rmsd -if_bb 1||0 create plot of key vs. interface backbone rmsd Abbreviations: fnac - Fraction of Native Atom Contacts fnrc - Fraction of Native Residue Contacts fnarc - fnac with Reduced atom modelsa_dasa.py:
Calculate change in access. and molecular surface upon binding. Syntax: a_dasa.py -r |rec_model| -l |lig_model| -c |com_model|
a_ensemble.py:
Analyze ensemble MD. Syntax: a_ensemble.py -i traj.dat [ -z |outlier-z-score| -step |offset| -o |out.eps| -title |plot_title| ]a_foldX.py:
Syntax: a_foldX -c |complexes.cl| -o |out_folder| -ref |ref.complex|
a_model_rms.py:
load model dictionary and report * average pairwise rmsd * average rmsd to free structure (assumed to be model 1) * average rmsd to bound structure model 1 is not included in the calculations
a_multiDock.py:
a_multiDock Visualize multidock results Note: interface rms values are for contact atoms not contact residues. Syntax a_multiDock -cl |complexList.cl| cl - complexList, has to contain info dictionary data for key r - hex receptor pdbs (i.e rec/*_hex.pdb) l - hex ligand pdbs (i.e lig/*_hex.pdb) ref - reference complex key - info dictionary key to plot (high values are considered good) inv - 1||0 inverse data associated with key (i.e. for rmds plots) maxContour - scale contour circles to fit at most x solutions additional_profile - add to profile plot (rec_model lig_model) Result 5 plots, info txt file, dumped dataa_multidock_contour.py:
a_multiDock Visualize multidock results Syntax a_multiDock -cl |complexList.cl| cl - complexList, has to contain info dictionary data for key inv - 1||0 inverse data associated with key (i.e. for rmds plots) maxContour - scale contour circles to fit at most x solutionsa_random_contacting.py:
a_random_contacting.py -i 1.cl 2.cl 3.cl .. -ref ref.complex -nat natively_contacted.cl [ -t ] [ -nout summary_output_file -rout random_output_file ] Get confidence of native scores from several scores to random reference. Prints table 3 of multidock paper, i.e. fnac, score, rms, .. of free vs. free docking, the docking with highest fnac, and the docking with highest score - FOR EACH complex list. The ref.complex is used to calculate the interface rmsd to the bound. Calculates averages and confidence of highest score, rms.. The 'real' table 3 line for the native is appended to a separate file. t .. print table header nout .. append line with native fnac, scores,.. and confidence to this file rout .. append lines with random fnac, scores,.. to this file [STDOUT]a_report_comEntropy.py:
Analyze the result of a_comEntropy.py Syntax: a_report_comEntropy.py -i |result.dic| [-eps |file| -tall -tsd -t -prefix |str|] Options: -i dictionary from a_comEntropy.py -eps output plot with entropies for all values of variable parameter -tall print table with all entropy values -tsd print single line with entropy values and standard dev (var=='ex3') -t print header row of table -prefix prefix for tables (e.g. a01)a_rmsd_vs_dock_performance.py:
Collect, calculate and save to disc (both as text files and pickled dictionaries) various data about a complex list. This script is written to collect data from multidocking runs and assums that the first ligand and receptor model is the free xray structure. Syntax: a_rmsd_vs_dock_performance.py -cl |complexList.cl| -ref |ref.complex| [-key |str| -inv [int|] cl - complexList, has to contain info dictionary data for key ref - reference complex key - info dictionary key to plot (high values are considered good) inv - 1||0 inverse data associated with key (i.e. for rmds plots) Output: An output directory 'model_vs_rmsd' is created and various text files and corresponding dictionaries are written to it.a_rmsd_vs_performance.py:
Create figure for multidock paper with the change in docking performance plotted against the change in rms_to_bound (both relative to the free-free docking). k .. PDBModel.info key to be plotted (rms_to_bound, rmsCA_to_bound, ..) o .. file name of output eps i .. one or many complexists with
a_table_fnac_rms_score.py:
a_table_fnac_rms_score.py [ -i complexes_cont.cl -ref ref.complex -t ] Creates one line of table 3 of multidock paper, i.e. fnac, score, rms, .. of free vs. free docking, the docking with highest fnac, and the docking with highest score. t .. print table header
a_trajEntropy.py:
Analyze entropy of a single or one rec and one lig trajectory with ptraj. Syntax: a_trajEntropy.py -i |traj1.dat+traj2.dat| [ -o |result.dic| -ref |ref_structure| -cast -chains |chain_indices| -border |chain| -split -shift -shuffle -s |startFrame| -e |endFrame| -step |frameOffset| -ss |member_startFrame| -se |member_endFrame| -ex_n |exclude_n_members| -ex1 |ex_from_traj1| -ex2 |ex_from_traj2| -ex3 |exclude_member_tripple| -atoms |CA CB ..| -heavy -nice |level| -parm |parm_file| -crd |crd_file| -f_out |ptraj_out| -f_template |ptraj_template| -log |log_file| -debug -verbose ] Options: i 1 trajectory or 2 trajectories connected by '+' o file name for pickled result dictionary s skip first |s| frames (of complete trajectory) e skip frames after |e| (of complete trajectory) ss skip first |ss| frames of each member trajectory se skip frames after |se| of each member trajectory atoms considered atoms (default: all) heavy remove hydrogens (default: don't) ref pickled PDBModel, Complex, or Trajectory cast equalize atom content of traj and ref [no] chains list of integer chain indices e.g -chains 0 1 [all] border 1st chain of 2nd molecule for -split, -shift, -shuffle split split complex trajectory and fit rec and lig separately (requires -border with first lig chain) [no] shuffle shuffle the order of rec vs. lig frames thin use randomly distributed fraction of frames, e.g. 0.2 [all] step frame offset, use every step frame, e.g. 5 [all] ex1 exclude these members from 1st trajectory, e.g. 3 6 ex2 exclude these members from 2nd trajectory (if given) ex_n exclude first n members [0] ex3 exclude |ex3|rd tripple of members, e.g. 2 excludes 3,4,5 (0 excludes nothing) [0] f_template alternative ptraj input template [default template] f_out target name for ptraj output file [discard] nice nice level [0] log file for program log [STOUT] debug keep all temporary files verbose print extended progress messages to log [log != STDOUT]a_trajQuality.py:
Syntax: a_trajQuality -i |traj_1 traj_2 .. traj_n| [-a -h |n_hosts| -w] pvm must be running on the local machine! Result: eps with quality plots in folder of traj files Options: -h number of hosts to be used -a first add hosts to pvm -w display a xterm window for each noderandom_complexes.py:
random_complexes.py -r |rec_model| -l |lig_model| [ -o |out_file| -n |number| -traj |traj_out_name| -ro |rec_out| -lo |lig_out| -debug copy_inp_file ] Remark: to create valid PCRModels for rec and lig use (in rec_wet/, lig_wet/) 1pdb2model.py -i ????.pdb -psf ????.psf -o xplor.model The waters in the PSF are deleted. They can be in the model but don't have to. Options: r pickled PCRModel, receptor (psf file has to be valid) l pickled PCRModel, ligand (psf file has to be valid) ro file name for rec copy (centered and no waters) lo file name for lig copy (centered and no waters) o file name for result ComplexList n number of random complexes to generate traj file name for optional amber crd and pdb (for visualisation) debug keep Xplor input file and don't delete temporary filesrandom_grouping.py:
select set of non-redundant, non-native random complexes random_grouping.py -cl |complex_list| Options: ref .. pickled native complex h .. number of hosts co .. folder name for result complexes o .. base name for other result files a .. add hosts to PVM before starting