Source Code for Module ML.Descriptors.MoleculeDescriptors

  1  # 
  2  #  Copyright (C) 2002  greg Landrum and Rational Discovery LLC 
  3  # 
  4  """ Various bits and pieces for calculating Molecular descriptors 
  5   
  6  """ 
  7  import RDConfig 
  8  from ML.Descriptors import Descriptors 
  9  from Chem import AvailDescriptors 
 10  AvailDescriptors.Desensitize() 
 11  import re 
 12   
 13 -class MolecularDescriptorCalculator(Descriptors.DescriptorCalculator): 
 14    """ used for calculating descriptors for molecules 
 15   
 16    """ 
 17 -  def __init__(self,simpleList,*args,**kwargs): 
 18      """ Constructor 
 19   
 20        **Arguments** 
 21   
 22          - simpleList: list of simple descriptors to be calculated 
 23                (see below for format)  
 24   
 25        **Note** 
 26   
 27          - format of simpleList: 
 28   
 29             a list of strings which are keys into _AvailDescriptors.descDict_ 
 30   
 31      """ 
 32      self.simpleList = simpleList[:] 
 33      self.descriptorNames = self.simpleList[:] 
 34      self.compoundList = None 
 35     
 36 -  def SaveState(self,fileName): 
 37      """ Writes this calculator off to a file so that it can be easily loaded later 
 38   
 39       **Arguments** 
 40   
 41         - fileName: the name of the file to be written 
 42          
 43      """ 
 44      import cPickle 
 45      try: 
 46        f = open(fileName,'wb+') 
 47      except: 
 48        print 'cannot open output file %s for writing'%(fileName) 
 49        return 
 50      cPickle.dump(self,f) 
 51      f.close() 
 52   
 53 -  def CalcDescriptors(self,mol,*args,**kwargs): 
 54      """ calculates all descriptors for a given molecule 
 55   
 56        **Arguments** 
 57   
 58          - mol: the molecule to be used 
 59   
 60        **Returns**   
 61          a tuple of all descriptor values 
 62   
 63      """ 
 64      res = [None]*len(self.simpleList) 
 65      for i in range(len(self.simpleList)): 
 66        nm = self.simpleList[i] 
 67        fn = AvailDescriptors.descDict.get(nm,lambda x:777) 
 68        #print '>',nm 
 69        try: 
 70          res[i] = fn(mol) 
 71        except: 
 72          import traceback 
 73          traceback.print_exc() 
 74          res[i]=-666 
 75      return tuple(res) 
 76   
 77 -  def GetDescriptorNames(self): 
 78      """ returns a tuple of the names of the descriptors this calculator generates 
 79   
 80      """ 
 81      return tuple(self.descriptorNames) 
 82   
 83 -  def GetDescriptorSummaries(self): 
 84      """ returns a tuple of summaries for the descriptors this calculator generates 
 85   
 86      """ 
 87      res = [] 
 88      for nm in self.simpleList: 
 89        fn = AvailDescriptors.descDict.get(nm,lambda x:777) 
 90        if hasattr(fn,'__doc__') and fn.__doc__: 
 91          doc = fn.__doc__.split('\n\n')[0].strip() 
 92          doc = re.sub('\ *\n\ *',' ',doc) 
 93        else: 
 94          doc = 'N/A' 
 95        res.append(doc) 
 96      return res 
 97   
 98 -  def GetDescriptorFuncs(self): 
 99      """ returns a tuple of the functions used to generate this calculator's descriptors 
100   
101      """ 
102      res = [] 
103      for nm in self.simpleList: 
104        fn = AvailDescriptors.descDict.get(nm,lambda x:777) 
105        res.append(fn) 
106      return tuple(res)   
107       
108  if __name__ == '__main__': 
109    from Chem import * 
110    from ML.Descriptors import MoleculeDescriptors 
111    descs = ['MolLogP','IWd','Chi1v'] 
112    smis = ['CCOC','CC=O','CCC(=O)O'] 
113    calc = MoleculeDescriptors.MolecularDescriptorCalculator(descs) 
114    calc.SaveState('test_data/molcalc.dsc') 
115    print calc.GetDescriptorNames() 
116   
117    for smi in smis: 
118      mol = MolFromSmiles(smi) 
119      print smi,calc.CalcDescriptors(mol) 
120