def test_sanitize(self):
for status, block in test_data:
print("*" * 44)
rxna = AllChem.ReactionFromRxnBlock(block)
rxnb = AllChem.ReactionFromRxnBlock(block)
rxna.Initialize()
res = rdChemReactions.PreprocessReaction(rxna)
print(AllChem.ReactionToRxnBlock(rxna))
if status == "good":
self.assertEquals(res, good_res)
elif status == "bad":
self.assertEquals(res, bad_res)
print(">" * 44)
rxnb.Initialize()
try:
rdChemReactions.SanitizeRxn(rxnb)
res = rdChemReactions.PreprocessReaction(rxnb)
print(AllChem.ReactionToRxnBlock(rxnb))
self.assertEquals(res, good_res)
assert not status == "fail"
except:
print("$RXN Failed")
if status == "fail":
continue
raise
def testPreprocess(self):
testFile = os.path.join(RDConfig.RDCodeDir, 'Chem', 'SimpleEnum', 'test_data', 'boronic1.rxn')
rxn = rdChemReactions.ReactionFromRxnFile(testFile)
rxn.Initialize()
res = rdChemReactions.PreprocessReaction(rxn)
self.assertEquals(res,
(0, 0, 2, 1, (((0, 'halogen.bromine.aromatic'), ), ((1, 'boronicacid'), ))))
def PreprocessReaction(reaction,
funcGroupFilename=None,
propName='molFileValue'):
"""
>>> from rdkit.Chem import AllChem
>>> testFile = os.path.join(RDConfig.RDCodeDir,'Chem','SimpleEnum','test_data','boronic1.rxn')
>>> rxn = AllChem.ReactionFromRxnFile(testFile)
>>> rxn.Initialize()
>>> nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
>>> nWarn
0
>>> nError
0
>>> nReacts
2
>>> nProds
1
>>> reactantLabels
(((0, 'halogen.bromine.aromatic'),), ((1, 'boronicacid'),))
If there are functional group labels in the input reaction (via atoms with molFileValue properties),
the corresponding atoms will have queries added to them so that they only match such things. We can
see this here:
>>> rxn = AllChem.ReactionFromRxnFile(testFile)
>>> rxn.Initialize()
>>> r1 = rxn.GetReactantTemplate(0)
>>> m1 = Chem.MolFromSmiles('CCBr')
>>> m2 = Chem.MolFromSmiles('c1ccccc1Br')
These both match because the reaction file itself just has R1-Br:
>>> m1.HasSubstructMatch(r1)
True
>>> m2.HasSubstructMatch(r1)
True
After preprocessing, we only match the aromatic Br:
>>> d = PreprocessReaction(rxn)
>>> m1.HasSubstructMatch(r1)
False
>>> m2.HasSubstructMatch(r1)
True
We also support or queries in the values field (separated by commas):
>>> testFile = os.path.join(RDConfig.RDCodeDir,'Chem','SimpleEnum','test_data','azide_reaction.rxn')
>>> rxn = AllChem.ReactionFromRxnFile(testFile)
>>> rxn.Initialize()
>>> reactantLabels = PreprocessReaction(rxn)[-1]
>>> reactantLabels
(((1, 'azide'),), ((1, 'carboxylicacid,acidchloride'),))
>>> m1 = Chem.MolFromSmiles('CC(=O)O')
>>> m2 = Chem.MolFromSmiles('CC(=O)Cl')
>>> m3 = Chem.MolFromSmiles('CC(=O)N')
>>> r2 = rxn.GetReactantTemplate(1)
>>> m1.HasSubstructMatch(r2)
True
>>> m2.HasSubstructMatch(r2)
True
>>> m3.HasSubstructMatch(r2)
False
unrecognized final group types are returned as None:
>>> testFile = os.path.join(RDConfig.RDCodeDir,'Chem','SimpleEnum','test_data','bad_value1.rxn')
>>> rxn = AllChem.ReactionFromRxnFile(testFile)
>>> rxn.Initialize()
>>> nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
Traceback (most recent call last):
File "/usr/prog/python/2.6.6_gnu/lib/python2.6/doctest.py", line 1253, in __run
compileflags, 1) in test.globs
File "<doctest __main__.PreprocessReaction[36]>", line 1, in <module>
nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
File "Enumerator.py", line 105, in PreprocessReaction
reactantLabels = reaction.AddRecursiveQueriesToReaction(queryDict, propName='molFileValue', getLabels=True)
RuntimeError: KeyErrorException
One unrecognized group type in a comma-separated list makes the whole thing fail:
>>> testFile = os.path.join(RDConfig.RDCodeDir,'Chem','SimpleEnum','test_data','bad_value2.rxn')
>>> rxn = AllChem.ReactionFromRxnFile(testFile)
>>> rxn.Initialize()
>>> nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
Traceback (most recent call last):
File "/usr/prog/python/2.6.6_gnu/lib/python2.6/doctest.py", line 1253, in __run
compileflags, 1) in test.globs
File "<doctest __main__.PreprocessReaction[36]>", line 1, in <module>
nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
File "Enumerator.py", line 105, in PreprocessReaction
reactantLabels = reaction.AddRecursiveQueriesToReaction(queryDict, propName='molFileValue', getLabels=True)
RuntimeError: KeyErrorException
>>> testFile = os.path.join(RDConfig.RDCodeDir,'Chem','SimpleEnum','test_data','bad_value3.rxn')
>>> rxn = AllChem.ReactionFromRxnFile(testFile)
>>> rxn.Initialize()
>>> nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
Traceback (most recent call last):
File "/usr/prog/python/2.6.6_gnu/lib/python2.6/doctest.py", line 1253, in __run
compileflags, 1) in test.globs
File "<doctest __main__.PreprocessReaction[36]>", line 1, in <module>
nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
File "Enumerator.py", line 105, in PreprocessReaction
reactantLabels = reaction.AddRecursiveQueriesToReaction(queryDict, propName='molFileValue', getLabels=True)
RuntimeError: KeyErrorException
>>> rxn = rdChemReactions.ChemicalReaction()
>>> rxn.Initialize()
>>> nWarn,nError,nReacts,nProds,reactantLabels = PreprocessReaction(rxn)
>>> reactantLabels
()
>>> reactantLabels == ()
True
"""
if funcGroupFilename:
try:
queryDict = Chem.ParseMolQueryDefFile(funcGroupFilename)
except Exception:
raise IOError('cannot open', funcGroupFilename)
return rdChemReactions.PreprocessReaction(reaction, queryDict,
propName)
return rdChemReactions.PreprocessReaction(reaction, propName=propName)