# HG changeset patch # User bgruening # Date 1376552110 14400 # Node ID 6bb56f3a8de0a1224c6534619649bba9b12080d5 Uploaded diff -r 000000000000 -r 6bb56f3a8de0 rdkit_descriptors.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/rdkit_descriptors.py Thu Aug 15 03:35:10 2013 -0400 @@ -0,0 +1,77 @@ +#!/usr/bin/env python + +from rdkit.Chem import Descriptors +from rdkit import Chem +import sys, os, re +import argparse +import inspect + +def get_supplier( infile, format = 'smiles' ): + """ + Returns a generator over a SMILES or InChI file. Every element is of RDKit + molecule and has its original string as _Name property. + """ + with open(infile) as handle: + for line in handle: + line = line.strip() + if format == 'smiles': + mol = Chem.MolFromSmiles( line, sanitize=True ) + elif format == 'inchi': + mol = Chem.inchi.MolFromInchi( line, sanitize=True, removeHs=True, logLevel=None, treatWarningAsError=False ) + if mol is None: + yield False + else: + mol.SetProp( '_Name', line.split('\t')[0] ) + yield mol + + +def get_rdkit_descriptor_functions(): + """ + Returns all descriptor functions under the Chem.Descriptors Module as tuple of (name, function) + """ + ret = [ (name, f) for name, f in inspect.getmembers( Descriptors ) if inspect.isfunction( f ) and not name.startswith( '_' ) ] + ret.sort() + return ret + + +def descriptors( mol, functions ): + """ + Calculates the descriptors of a given molecule. + """ + for name, function in functions: + yield (name, function( mol )) + + +if __name__ == "__main__": + parser = argparse.ArgumentParser() + parser.add_argument('-i', '--infile', required=True, help='Path to the input file.') + parser.add_argument("--iformat", help="Specify the input file format.") + + parser.add_argument('-o', '--outfile', type=argparse.FileType('w+'), + default=sys.stdout, help="path to the result file, default it sdtout") + + parser.add_argument("--header", dest="header", action="store_true", + default=False, + help="Write header line.") + + args = parser.parse_args() + + if args.iformat == 'sdf': + supplier = Chem.SDMolSupplier( args.infile ) + elif args.iformat =='smi': + supplier = get_supplier( args.infile, format = 'smiles' ) + elif args.iformat == 'inchi': + supplier = get_supplier( args.infile, format = 'inchi' ) + + functions = get_rdkit_descriptor_functions() + + if args.header: + args.outfile.write( '%s\n' % '\t'.join( [name for name, f in functions] ) ) + + for mol in supplier: + if not mol: + continue + descs = descriptors( mol, functions ) + molecule_id = mol.GetProp("_Name") + args.outfile.write( "%s\n" % '\t'.join( [molecule_id]+ [str(res) for name, res in descs] ) ) + diff -r 000000000000 -r 6bb56f3a8de0 rdkit_descriptors.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/rdkit_descriptors.xml Thu Aug 15 03:35:10 2013 -0400 @@ -0,0 +1,143 @@ + + calculated with RDKit + + + rdkit + + rdkit_descriptors.py -i "${infile}" --iformat "${infile.ext}" -o "${outfile}" $header 2>&1 + + + + + + + + + + + +.. class:: infomark + +**What this tool does** + +| RDKit is an open source toolkit for cheminformatics and machine learning. +| This implementation focuses on descriptor calculation, though, RDKit offers a vast number of other functions. +| +| The table below shows a brief overview of the descriptors. +| + ++-----------------------------------+------------+ +| Descriptor/Descriptor Family | Language | ++===================================+============+ +| Gasteiger/Marsili Partial Charges | C++ | ++-----------------------------------+------------+ +| BalabanJ | Python | ++-----------------------------------+------------+ +| BertzCT | Python | ++-----------------------------------+------------+ +| Ipc | Python | ++-----------------------------------+------------+ +| HallKierAlpha | Python | ++-----------------------------------+------------+ +| Kappa1 - Kappa3 | Python | ++-----------------------------------+------------+ +| Chi0, Chi1 | Python | ++-----------------------------------+------------+ +| Chi0n - Chi4n | Python | ++-----------------------------------+------------+ +| Chi0v - Chi4v | Python | ++-----------------------------------+------------+ +| MolLogP | C++ | ++-----------------------------------+------------+ +| MolMR | C++ | ++-----------------------------------+------------+ +| MolWt | C++ | ++-----------------------------------+------------+ +| HeavyAtomCount | Python | ++-----------------------------------+------------+ +| HeavyAtomMolWt | Python | ++-----------------------------------+------------+ +| NHOHCount | C++ | ++-----------------------------------+------------+ +| NOCount | C++ | ++-----------------------------------+------------+ +| NumHAcceptors | C++ | ++-----------------------------------+------------+ +| NumHDonors | C++ | ++-----------------------------------+------------+ +| NumHeteroatoms | C++ | ++-----------------------------------+------------+ +| NumRotatableBonds | C++ | ++-----------------------------------+------------+ +| NumValenceElectrons | Python | ++-----------------------------------+------------+ +| RingCount | C++ | ++-----------------------------------+------------+ +| TPSA | C++ | ++-----------------------------------+------------+ +| LabuteASA | C++ | ++-----------------------------------+------------+ +| PEOE_VSA1 - PEOE_VSA14 | Python/C++ | ++-----------------------------------+------------+ +| SMR_VSA1 - SMR_VSA10 | Python/C++ | ++-----------------------------------+------------+ +| SlogP_VSA1 - SlogP_VSA12 | Python/C++ | ++-----------------------------------+------------+ +| EState_VSA1 - EState_VSA11 | Python | ++-----------------------------------+------------+ +| VSA_EState1 - VSA_EState10 | Python | ++-----------------------------------+------------+ +| Topliss fragments | Python | ++-----------------------------------+------------+ + +| +| A full list of the descriptors can be obtained here_. + +.. _here: https://code.google.com/p/rdkit/wiki/DescriptorsInTheRDKit + +----- + +.. class:: warningmark + +**HINT** + +Use the **cut columns from a table** tool to select just the desired descriptors. + +----- + +.. class:: infomark + +**Input** + +| - `SD-Format`_ +| - `SMILES Format`_ +| - TDT_ +| - SLN +| - `Corina MOL2`_ + +.. _SD-Format: http://en.wikipedia.org/wiki/Chemical_table_file +.. _SMILES Format: http://en.wikipedia.org/wiki/Simplified_molecular_input_line_entry_specification +.. _TDT: https://earray.chem.agilent.com/earray/helppages/index.htm#tdt_format_files.htm +.. _Corina MOL2: http://www.molecular-networks.com/products/corina + +----- + +.. class:: infomark + + **Output** + +Tabularfile, where each descriptor (value) is shown in a seperate column. + +----- + +.. class:: informark + +**Cite** + +Greg Landrum - RDKit_: Open-source cheminformatics + +.. _RDKit: http://www.rdkit.org + + + + diff -r 000000000000 -r 6bb56f3a8de0 repository_dependencies.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/repository_dependencies.xml Thu Aug 15 03:35:10 2013 -0400 @@ -0,0 +1,4 @@ + + + + diff -r 000000000000 -r 6bb56f3a8de0 tool_dependencies.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tool_dependencies.xml Thu Aug 15 03:35:10 2013 -0400 @@ -0,0 +1,6 @@ + + + + + +