Source Code for Module rdkit.Chem.EState.Fingerprinter

 1  ## Automatically adapted for numpy.oldnumeric Jun 27, 2008 by -c 
 2   
 3  # $Id: Fingerprinter.py 1528 2010-09-26 17:04:37Z glandrum $ 
 4  # 
 5  #  Copyright (C) 2002-2006  greg Landrum and Rational Discovery LLC 
 6  # 
 7  #   @@ All Rights Reserved @@ 
 8  #  This file is part of the RDKit. 
 9  #  The contents are covered by the terms of the BSD license 
10  #  which is included in the file license.txt, found at the root 
11  #  of the RDKit source tree. 
12  # 
13  """  EState fingerprinting 
14   
15  """ 
16  import numpy 
17  from rdkit.Chem.EState import EStateIndices 
18  from rdkit.Chem.EState import AtomTypes 
19   
20 -def FingerprintMol(mol): 
21    """ generates the EState fingerprints for the molecule 
22   
23    Concept from the paper: Hall and Kier JCICS _35_ 1039-1045 (1995) 
24   
25    two numeric arrays are returned: 
26      The first (of ints) contains the number of times each possible atom type is hit 
27      The second (of floats) contains the sum of the EState indices for atoms of 
28        each type. 
29   
30    """ 
31    if AtomTypes.esPatterns is None: 
32      AtomTypes.BuildPatts() 
33    esIndices = EStateIndices(mol) 
34   
35    nPatts = len(AtomTypes.esPatterns) 
36    counts = numpy.zeros(nPatts,numpy.int) 
37    sums = numpy.zeros(nPatts,numpy.float) 
38   
39    for i,(name,pattern) in enumerate(AtomTypes.esPatterns): 
40      matches = mol.GetSubstructMatches(pattern,uniquify=1) 
41      counts[i] = len(matches) 
42      for match in matches: 
43        sums[i] += esIndices[match[0]] 
44    return counts,sums 
45   
46   
47  if __name__ == '__main__': 
48    from rdkit import Chem 
49    smis = ['CC','CCC','c1[nH]cnc1CC(N)C(O)=O','NCCc1ccc(O)c(O)c1'] 
50    for smi in smis: 
51      m = Chem.MolFromSmiles(smi) 
52      print smi,Chem.MolToSmiles(m) 
53      types = AtomTypes.TypeAtoms(m) 
54      for i in range(m.GetNumAtoms()): 
55        print '%d %4s: %s'%(i+1,m.GetAtomWithIdx(i).GetSymbol(),str(types[i])) 
56      es = EStateIndices(m) 
57      counts,sums = FingerprintMol(m) 
58      for i in range(len(AtomTypes.esPatterns)): 
59        if counts[i]: 
60          name,patt = AtomTypes.esPatterns[i] 
61          print '%6s, % 2d, % 5.4f'%(name,counts[i],sums[i]) 
62      for i in range(len(es)): 
63        print '% 2d, % 5.4f'%(i+1,es[i]) 
64      print '--------' 
65