Source Code for Module rdkit.Chem.Fragments

 1  # $Id$ 
 2  # 
 3  # Copyright (C) 2002-2010 greg Landrum and Rational Discovery LLC 
 4  # 
 5  #   @@ All Rights Reserved @@ 
 6  #  This file is part of the RDKit. 
 7  #  The contents are covered by the terms of the BSD license 
 8  #  which is included in the file license.txt, found at the root 
 9  #  of the RDKit source tree. 
10  # 
11  """ functions to match a bunch of fragment descriptors from a file 
12   
13  No user-servicable parts inside.  ;-) 
14   
15  """ 
16  import os 
17  from rdkit import RDConfig 
18  from rdkit import Chem 
19   
20  defaultPatternFileName = os.path.join(RDConfig.RDDataDir, 'FragmentDescriptors.csv') 
21   
22   
23 -def _CountMatches(mol, patt, unique=True): 
24    return len(mol.GetSubstructMatches(patt, uniquify=unique)) 
25   
26   
27  fns = [] 
28   
29   
30 -def _LoadPatterns(fileName=None): 
31    if fileName is None: 
32      fileName = defaultPatternFileName 
33    try: 
34      with open(fileName, 'r') as inF: 
35        for line in inF.readlines(): 
36          if len(line) and line[0] != '#': 
37            splitL = line.split('\t') 
38            if len(splitL) >= 3: 
39              name = splitL[0] 
40              descr = splitL[1] 
41              sma = splitL[2] 
42              descr = descr.replace('"', '') 
43              patt = Chem.MolFromSmarts(sma) 
44              if not patt or patt.GetNumAtoms() == 0: 
45                raise ImportError('Smarts %s could not be parsed' % (repr(sma))) 
46              fn = lambda mol, countUnique=True, pattern=patt: _CountMatches(mol, pattern, unique=countUnique) 
47              fn.__doc__ = descr 
48              name = name.replace('=', '_') 
49              name = name.replace('-', '_') 
50              fns.append((name, fn)) 
51    except IOError: 
52      pass 
53   
54   
55  _LoadPatterns() 
56  for name, fn in fns: 
57    exec('%s=fn' % (name)) 
58  fn = None 
59