def test10NormalizeFromData(self):
data = """// Name SMIRKS
Nitro to N+(O-)=O [N,P,As,Sb;X3:1](=[O,S,Se,Te:2])=[O,S,Se,Te:3]>>[*+1:1]([*-1:2])=[*:3]
Sulfone to S(=O)(=O) [S+2:1]([O-:2])([O-:3])>>[S+0:1](=[O-0:2])(=[O-0:3])
Pyridine oxide to n+O- [n:1]=[O:2]>>[n+:1][O-:2]
// Azide to N=N+=N- [*,H:1][N:2]=[N:3]#[N:4]>>[*,H:1][N:2]=[N+:3]=[N-:4]
"""
normalizer1 = rdMolStandardize.Normalizer()
params = rdMolStandardize.CleanupParameters()
normalizer2 = rdMolStandardize.NormalizerFromData(data, params)
imol = Chem.MolFromSmiles("O=N(=O)CCN=N#N", sanitize=False)
mol1 = normalizer1.normalize(imol)
mol2 = normalizer2.normalize(imol)
self.assertEqual(Chem.MolToSmiles(imol), "N#N=NCCN(=O)=O")
self.assertEqual(Chem.MolToSmiles(mol1), "[N-]=[N+]=NCC[N+](=O)[O-]")
self.assertEqual(Chem.MolToSmiles(mol2), "N#N=NCC[N+](=O)[O-]")
def test13Tautomers(self):
enumerator = rdMolStandardize.TautomerEnumerator()
m = Chem.MolFromSmiles("C1(=CCCCC1)O")
ctaut = enumerator.Canonicalize(m)
self.assertEqual(Chem.MolToSmiles(ctaut), "O=C1CCCCC1")
params = rdMolStandardize.CleanupParameters()
enumerator = rdMolStandardize.TautomerEnumerator(params)
m = Chem.MolFromSmiles("C1(=CCCCC1)O")
ctaut = enumerator.Canonicalize(m)
self.assertEqual(Chem.MolToSmiles(ctaut), "O=C1CCCCC1")
tauts = enumerator.Enumerate(m)
self.assertEqual(len(tauts), 2)
ctauts = list(sorted(Chem.MolToSmiles(x) for x in tauts))
self.assertEqual(ctauts, ['O=C1CCCCC1', 'OC1=CCCCC1'])
def scorefunc1(mol):
' stupid tautomer scoring function '
p = Chem.MolFromSmarts('[OH]')
return len(mol.GetSubstructMatches(p))
def scorefunc2(mol):
' stupid tautomer scoring function '
p = Chem.MolFromSmarts('O=C')
return len(mol.GetSubstructMatches(p))
m = Chem.MolFromSmiles("C1(=CCCCC1)O")
ctaut = enumerator.Canonicalize(m, scorefunc1)
self.assertEqual(Chem.MolToSmiles(ctaut), "OC1=CCCCC1")
ctaut = enumerator.Canonicalize(m, scorefunc2)
self.assertEqual(Chem.MolToSmiles(ctaut), "O=C1CCCCC1")
# make sure lambdas work
ctaut = enumerator.Canonicalize(
m, lambda x: len(x.GetSubstructMatches(Chem.MolFromSmarts('C=O'))))
self.assertEqual(Chem.MolToSmiles(ctaut), "O=C1CCCCC1")
# make sure we behave if we return something bogus from the scoring function
with self.assertRaises(TypeError):
ctaut = enumerator.Canonicalize(m, lambda x: 'fail')
self.assertEqual(
enumerator.ScoreTautomer(Chem.MolFromSmiles('N=c1[nH]cccc1')), 99)
self.assertEqual(
enumerator.ScoreTautomer(Chem.MolFromSmiles('Nc1ncccc1')), 100)
def test19NormalizeFromParams(self):
params = rdMolStandardize.CleanupParameters()
params.normalizationsFile = "ThisFileDoesNotExist.txt"
with self.assertRaises(OSError):
rdMolStandardize.NormalizerFromParams(params)
def test17PickCanonicalCIPChangeOnChiralCenter(self):
def get_canonical_taut(res):
best_idx = max([
(rdMolStandardize.TautomerEnumerator.ScoreTautomer(t), i)
for i, t in enumerate(res.tautomers)
])[1]
return res.tautomers[best_idx]
smi = "CC\\C=C(/O)[C@@H](C)C(C)=O"
mol = Chem.MolFromSmiles(smi)
self.assertIsNotNone(mol)
self.assertEqual(
mol.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(mol.GetAtomWithIdx(5).GetProp("_CIPCode"), "R")
# here the chirality disappears as the chiral center is itself involved in tautomerism
te = rdMolStandardize.TautomerEnumerator()
can_taut = te.Canonicalize(mol)
self.assertIsNotNone(can_taut)
self.assertEqual(
can_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_UNSPECIFIED)
self.assertFalse(can_taut.GetAtomWithIdx(5).HasProp("_CIPCode"))
self.assertEqual(Chem.MolToSmiles(can_taut), "CCCC(=O)C(C)C(C)=O")
# here the chirality stays even if the chiral center is itself involved in tautomerism
# because of the tautomerRemoveSp3Stereo parameter being set to false
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
te = rdMolStandardize.TautomerEnumerator(params)
can_taut = te.Canonicalize(mol)
self.assertIsNotNone(can_taut)
self.assertEqual(
can_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(can_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "S")
self.assertEqual(Chem.MolToSmiles(can_taut), "CCCC(=O)[C@@H](C)C(C)=O")
# here the chirality disappears as the chiral center is itself involved in tautomerism
# the reassignStereo setting has no influence
te = rdMolStandardize.TautomerEnumerator()
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 8)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_UNSPECIFIED)
self.assertFalse(best_taut.GetAtomWithIdx(5).HasProp("_CIPCode"))
self.assertEqual(Chem.MolToSmiles(best_taut), "CCCC(=O)C(C)C(C)=O")
# here the chirality disappears as the chiral center is itself involved in tautomerism
# the reassignStereo setting has no influence
params = rdMolStandardize.CleanupParameters()
params.tautomerReassignStereo = False
te = rdMolStandardize.TautomerEnumerator(params)
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 8)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_UNSPECIFIED)
self.assertFalse(best_taut.GetAtomWithIdx(5).HasProp("_CIPCode"))
self.assertEqual(Chem.MolToSmiles(best_taut), "CCCC(=O)C(C)C(C)=O")
# here the chirality stays even if the chiral center is itself involved in tautomerism
# because of the tautomerRemoveSp3Stereo parameter being set to false
# as reassignStereo by default is true, the CIP code has been recomputed
# and therefore it is now S (correct)
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
te = rdMolStandardize.TautomerEnumerator(params)
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 8)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(best_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "S")
self.assertEqual(Chem.MolToSmiles(best_taut),
"CCCC(=O)[C@@H](C)C(C)=O")
# here the chirality stays even if the chiral center is itself involved in tautomerism
# because of the tautomerRemoveSp3Stereo parameter being set to false
# as reassignStereo is false, the CIP code has not been recomputed
# and therefore it is still R (incorrect)
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
params.tautomerReassignStereo = False
te = rdMolStandardize.TautomerEnumerator(params)
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 8)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(best_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "R")
self.assertEqual(Chem.MolToSmiles(best_taut),
"CCCC(=O)[C@@H](C)C(C)=O")
smi = "CC\\C=C(/O)[C@@](CC)(C)C(C)=O"
mol = Chem.MolFromSmiles(smi)
self.assertIsNotNone(mol)
self.assertEqual(mol.GetAtomWithIdx(5).GetProp("_CIPCode"), "S")
self.assertEqual(
mol.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
# here the chirality stays no matter how tautomerRemoveSp3Stereo
# is set as the chiral center is not involved in tautomerism
te = rdMolStandardize.TautomerEnumerator()
can_taut = te.Canonicalize(mol)
self.assertIsNotNone(can_taut)
self.assertEqual(
can_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(can_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "R")
self.assertEqual(Chem.MolToSmiles(can_taut),
"CCCC(=O)[C@](C)(CC)C(C)=O")
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
te = rdMolStandardize.TautomerEnumerator(params)
can_taut = te.Canonicalize(mol)
self.assertIsNotNone(can_taut)
self.assertEqual(
can_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(can_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "R")
self.assertEqual(Chem.MolToSmiles(can_taut),
"CCCC(=O)[C@](C)(CC)C(C)=O")
# as reassignStereo by default is true, the CIP code has been recomputed
# and therefore it is now R (correct)
te = rdMolStandardize.TautomerEnumerator()
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 4)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(best_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "R")
self.assertEqual(Chem.MolToSmiles(best_taut),
"CCCC(=O)[C@](C)(CC)C(C)=O")
# as reassignStereo is false, the CIP code has not been recomputed
# and therefore it is still S (incorrect)
params = rdMolStandardize.CleanupParameters()
params.tautomerReassignStereo = False
te = rdMolStandardize.TautomerEnumerator(params)
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 4)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(best_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "S")
self.assertEqual(Chem.MolToSmiles(best_taut),
"CCCC(=O)[C@](C)(CC)C(C)=O")
# as reassignStereo by default is true, the CIP code has been recomputed
# and therefore it is now R (correct)
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
te = rdMolStandardize.TautomerEnumerator(params)
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 4)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(best_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "R")
self.assertEqual(Chem.MolToSmiles(best_taut),
"CCCC(=O)[C@](C)(CC)C(C)=O")
# here the chirality stays even if the tautomerRemoveSp3Stereo parameter
# is set to false as the chiral center is not involved in tautomerism
# as reassignStereo is false, the CIP code has not been recomputed
# and therefore it is still S (incorrect)
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
params.tautomerReassignStereo = False
te = rdMolStandardize.TautomerEnumerator(params)
res = te.Enumerate(mol)
self.assertEqual(res.status,
rdMolStandardize.TautomerEnumeratorStatus.Completed)
self.assertEqual(len(res.tautomers), 4)
best_taut = get_canonical_taut(res)
self.assertIsNotNone(best_taut)
self.assertEqual(
best_taut.GetAtomWithIdx(5).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CW)
self.assertEqual(best_taut.GetAtomWithIdx(5).GetProp("_CIPCode"), "S")
self.assertEqual(Chem.MolToSmiles(best_taut),
"CCCC(=O)[C@](C)(CC)C(C)=O")
def test16EnumeratorCallback(self):
class MyTautomerEnumeratorCallback(
rdMolStandardize.TautomerEnumeratorCallback):
def __init__(self, parent, timeout_ms):
super().__init__()
self._parent = parent
self._timeout = timedelta(milliseconds=timeout_ms)
self._start_time = datetime.now()
def __call__(self, mol, res):
self._parent.assertTrue(isinstance(mol, Chem.Mol))
self._parent.assertTrue(
isinstance(res, rdMolStandardize.TautomerEnumeratorResult))
return (datetime.now() - self._start_time < self._timeout)
class MyBrokenCallback(rdMolStandardize.TautomerEnumeratorCallback):
pass
class MyBrokenCallback2(rdMolStandardize.TautomerEnumeratorCallback):
__call__ = 1
# Test a structure with hundreds of tautomers.
smi68 = "[H][C](CO)(NC(=O)C1=C(O)C(O)=CC=C1)C(O)=O"
m68 = Chem.MolFromSmiles(smi68)
params = rdMolStandardize.CleanupParameters()
params.maxTransforms = 10000
params.maxTautomers = 10000
enumerator = rdMolStandardize.TautomerEnumerator(params)
enumerator.SetCallback(MyTautomerEnumeratorCallback(self, 50.0))
res68 = enumerator.Enumerate(m68)
# either the enumeration was canceled due to timeout
# or it has completed very quickly
hasReachedTimeout = (
len(res68.tautomers) < 375 and res68.status
== rdMolStandardize.TautomerEnumeratorStatus.Canceled)
hasCompleted = (len(res68.tautomers) == 375 and res68.status
== rdMolStandardize.TautomerEnumeratorStatus.Completed)
if hasReachedTimeout:
print("Enumeration was canceled due to timeout (50 ms)",
file=sys.stderr)
if hasCompleted:
print("Enumeration has completed", file=sys.stderr)
self.assertTrue(hasReachedTimeout or hasCompleted)
self.assertTrue(hasReachedTimeout ^ hasCompleted)
enumerator = rdMolStandardize.TautomerEnumerator(params)
enumerator.SetCallback(MyTautomerEnumeratorCallback(self, 10000.0))
res68 = enumerator.Enumerate(m68)
# either the enumeration completed
# or it ran very slowly and was canceled due to timeout
hasReachedTimeout = (
len(res68.tautomers) < 375 and res68.status
== rdMolStandardize.TautomerEnumeratorStatus.Canceled)
hasCompleted = (len(res68.tautomers) == 375 and res68.status
== rdMolStandardize.TautomerEnumeratorStatus.Completed)
if hasReachedTimeout:
print("Enumeration was canceled due to timeout (10 s)",
file=sys.stderr)
if hasCompleted:
print("Enumeration has completed", file=sys.stderr)
self.assertTrue(hasReachedTimeout or hasCompleted)
self.assertTrue(hasReachedTimeout ^ hasCompleted)
enumerator = rdMolStandardize.TautomerEnumerator(params)
with self.assertRaises(AttributeError):
enumerator.SetCallback(MyBrokenCallback())
with self.assertRaises(AttributeError):
enumerator.SetCallback(MyBrokenCallback2())
def test15EnumeratorParams(self):
# Test a structure with hundreds of tautomers.
smi68 = "[H][C](CO)(NC(=O)C1=C(O)C(O)=CC=C1)C(O)=O"
m68 = Chem.MolFromSmiles(smi68)
enumerator = rdMolStandardize.TautomerEnumerator()
res68 = enumerator.Enumerate(m68)
self.assertEqual(len(res68), 292)
self.assertEqual(len(res68.tautomers), len(res68))
self.assertEqual(
res68.status,
rdMolStandardize.TautomerEnumeratorStatus.MaxTransformsReached)
params = rdMolStandardize.CleanupParameters()
params.maxTautomers = 50
enumerator = rdMolStandardize.TautomerEnumerator(params)
res68 = enumerator.Enumerate(m68)
self.assertEqual(len(res68), 50)
self.assertEqual(
res68.status,
rdMolStandardize.TautomerEnumeratorStatus.MaxTautomersReached)
sAlaSmi = "C[C@H](N)C(=O)O"
sAla = Chem.MolFromSmiles(sAlaSmi)
# test remove (S)-Ala stereochemistry
self.assertEqual(
sAla.GetAtomWithIdx(1).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CCW)
self.assertEqual(sAla.GetAtomWithIdx(1).GetProp("_CIPCode"), "S")
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = True
enumerator = rdMolStandardize.TautomerEnumerator(params)
res = enumerator.Enumerate(sAla)
for taut in res:
self.assertEqual(
taut.GetAtomWithIdx(1).GetChiralTag(),
Chem.ChiralType.CHI_UNSPECIFIED)
self.assertFalse(taut.GetAtomWithIdx(1).HasProp("_CIPCode"))
for taut in res.tautomers:
self.assertEqual(
taut.GetAtomWithIdx(1).GetChiralTag(),
Chem.ChiralType.CHI_UNSPECIFIED)
self.assertFalse(taut.GetAtomWithIdx(1).HasProp("_CIPCode"))
for i, taut in enumerate(res):
self.assertEqual(Chem.MolToSmiles(taut),
Chem.MolToSmiles(res.tautomers[i]))
self.assertEqual(len(res), len(res.smiles))
self.assertEqual(len(res), len(res.tautomers))
self.assertEqual(len(res), len(res()))
self.assertEqual(len(res), len(res.smilesTautomerMap))
for i, taut in enumerate(res.tautomers):
self.assertEqual(Chem.MolToSmiles(taut), Chem.MolToSmiles(res[i]))
self.assertEqual(Chem.MolToSmiles(taut), res.smiles[i])
self.assertEqual(
Chem.MolToSmiles(taut),
Chem.MolToSmiles(res.smilesTautomerMap.values()[i].tautomer))
for i, k in enumerate(res.smilesTautomerMap.keys()):
self.assertEqual(k, res.smiles[i])
for i, v in enumerate(res.smilesTautomerMap.values()):
self.assertEqual(Chem.MolToSmiles(v.tautomer),
Chem.MolToSmiles(res[i]))
for i, (k, v) in enumerate(res.smilesTautomerMap.items()):
self.assertEqual(k, res.smiles[i])
self.assertEqual(Chem.MolToSmiles(v.tautomer),
Chem.MolToSmiles(res[i]))
for i, smiles in enumerate(res.smiles):
self.assertEqual(smiles, Chem.MolToSmiles(res[i]))
self.assertEqual(smiles, res.smilesTautomerMap.keys()[i])
self.assertEqual(Chem.MolToSmiles(res.tautomers[-1]),
Chem.MolToSmiles(res[-1]))
self.assertEqual(Chem.MolToSmiles(res[-1]),
Chem.MolToSmiles(res[len(res) - 1]))
self.assertEqual(Chem.MolToSmiles(res.tautomers[-1]),
Chem.MolToSmiles(res.tautomers[len(res) - 1]))
with self.assertRaises(IndexError):
res[len(res)]
with self.assertRaises(IndexError):
res[-len(res) - 1]
with self.assertRaises(IndexError):
res.tautomers[len(res)]
with self.assertRaises(IndexError):
res.tautomers[-len(res.tautomers) - 1]
# test retain (S)-Ala stereochemistry
self.assertEqual(
sAla.GetAtomWithIdx(1).GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CCW)
self.assertEqual(sAla.GetAtomWithIdx(1).GetProp("_CIPCode"), "S")
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveSp3Stereo = False
enumerator = rdMolStandardize.TautomerEnumerator(params)
res = enumerator.Enumerate(sAla)
for taut in res:
tautAtom = taut.GetAtomWithIdx(1)
if (tautAtom.GetHybridization() == Chem.HybridizationType.SP3):
self.assertEqual(tautAtom.GetChiralTag(),
Chem.ChiralType.CHI_TETRAHEDRAL_CCW)
self.assertTrue(tautAtom.HasProp("_CIPCode"))
self.assertEqual(tautAtom.GetProp("_CIPCode"), "S")
else:
self.assertFalse(tautAtom.HasProp("_CIPCode"))
self.assertEqual(tautAtom.GetChiralTag(),
Chem.ChiralType.CHI_UNSPECIFIED)
eEnolSmi = "C/C=C/O"
eEnol = Chem.MolFromSmiles(eEnolSmi)
self.assertEqual(
eEnol.GetBondWithIdx(1).GetStereo(), Chem.BondStereo.STEREOE)
# test remove enol E stereochemistry
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveBondStereo = True
enumerator = rdMolStandardize.TautomerEnumerator(params)
res = enumerator.Enumerate(eEnol)
for taut in res.tautomers:
self.assertEqual(
taut.GetBondWithIdx(1).GetStereo(), Chem.BondStereo.STEREONONE)
# test retain enol E stereochemistry
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveBondStereo = False
enumerator = rdMolStandardize.TautomerEnumerator(params)
res = enumerator.Enumerate(eEnol)
for taut in res.tautomers:
if (taut.GetBondWithIdx(1).GetBondType() == Chem.BondType.DOUBLE):
self.assertEqual(
taut.GetBondWithIdx(1).GetStereo(),
Chem.BondStereo.STEREOE)
zEnolSmi = "C/C=C\\O"
zEnol = Chem.MolFromSmiles(zEnolSmi)
self.assertEqual(
zEnol.GetBondWithIdx(1).GetStereo(), Chem.BondStereo.STEREOZ)
# test remove enol Z stereochemistry
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveBondStereo = True
enumerator = rdMolStandardize.TautomerEnumerator(params)
res = enumerator.Enumerate(zEnol)
for taut in res:
self.assertEqual(
taut.GetBondWithIdx(1).GetStereo(), Chem.BondStereo.STEREONONE)
# test retain enol Z stereochemistry
params = rdMolStandardize.CleanupParameters()
params.tautomerRemoveBondStereo = False
enumerator = rdMolStandardize.TautomerEnumerator(params)
res = enumerator.Enumerate(zEnol)
for taut in res:
if (taut.GetBondWithIdx(1).GetBondType() == Chem.BondType.DOUBLE):
self.assertEqual(
taut.GetBondWithIdx(1).GetStereo(),
Chem.BondStereo.STEREOZ)
def test21UpdateFromJSON(self):
params = rdMolStandardize.CleanupParameters()
# note: these actual parameters aren't useful... they are for testing
rdMolStandardize.UpdateParamsFromJSON(
params, """{
"normalizationData":[
{"name":"silly 1","smarts":"[Cl:1]>>[F:1]"},
{"name":"silly 2","smarts":"[Br:1]>>[F:1]"}
],
"acidbaseData":[
{"name":"-CO2H","acid":"C(=O)[OH]","base":"C(=O)[O-]"},
{"name":"phenol","acid":"c[OH]","base":"c[O-]"}
],
"fragmentData":[
{"name":"hydrogen", "smarts":"[H]"},
{"name":"fluorine", "smarts":"[F]"},
{"name":"chlorine", "smarts":"[Cl]"}
],
"tautomerTransformData":[
{"name":"1,3 (thio)keto/enol f","smarts":"[CX4!H0]-[C]=[O,S,Se,Te;X1]","bonds":"","charges":""},
{"name":"1,3 (thio)keto/enol r","smarts":"[O,S,Se,Te;X2!H0]-[C]=[C]"}
]}""")
m = Chem.MolFromSmiles("CCC=O")
te = rdMolStandardize.TautomerEnumerator(params)
tauts = [Chem.MolToSmiles(x) for x in te.Enumerate(m)]
self.assertEqual(tauts, ["CC=CO", "CCC=O"])
self.assertEqual(
Chem.MolToSmiles(rdMolStandardize.CanonicalTautomer(m, params)),
"CCC=O")
# now with defaults
te = rdMolStandardize.TautomerEnumerator()
tauts = [Chem.MolToSmiles(x) for x in te.Enumerate(m)]
self.assertEqual(tauts, ["CC=CO", "CCC=O"])
self.assertEqual(
Chem.MolToSmiles(rdMolStandardize.CanonicalTautomer(m)), "CCC=O")
m = Chem.MolFromSmiles('ClCCCBr')
nm = rdMolStandardize.Normalize(m, params)
self.assertEqual(Chem.MolToSmiles(nm), "FCCCF")
# now with defaults
nm = rdMolStandardize.Normalize(m)
self.assertEqual(Chem.MolToSmiles(nm), "ClCCCBr")
m = Chem.MolFromSmiles('c1cc([O-])cc(C(=O)O)c1')
nm = rdMolStandardize.Reionize(m, params)
self.assertEqual(Chem.MolToSmiles(nm), "O=C([O-])c1cccc(O)c1")
# now with defaults
nm = rdMolStandardize.Reionize(m)
self.assertEqual(Chem.MolToSmiles(nm), "O=C([O-])c1cccc(O)c1")
m = Chem.MolFromSmiles('C1=C(C=CC(=C1)[S]([O-])=O)[S](O)(=O)=O')
nm = rdMolStandardize.Reionize(m, params)
self.assertEqual(Chem.MolToSmiles(nm), "O=S([O-])c1ccc(S(=O)(=O)O)cc1")
# now with defaults
nm = rdMolStandardize.Reionize(m)
self.assertEqual(Chem.MolToSmiles(nm), "O=S(O)c1ccc(S(=O)(=O)[O-])cc1")
m = Chem.MolFromSmiles('[F-].[Cl-].[Br-].CC')
nm = rdMolStandardize.RemoveFragments(m, params)
self.assertEqual(Chem.MolToSmiles(nm), "CC.[Br-]")
# now with defaults
nm = rdMolStandardize.RemoveFragments(m)
self.assertEqual(Chem.MolToSmiles(nm), "CC")
def test7Fragment(self):
fragremover = rdMolStandardize.FragmentRemover()
mol = Chem.MolFromSmiles("CN(C)C.Cl.Cl.Br")
nm = fragremover.remove(mol)
self.assertEqual(Chem.MolToSmiles(nm), "CN(C)C")
lfragchooser = rdMolStandardize.LargestFragmentChooser()
mol2 = Chem.MolFromSmiles("[N+](=O)([O-])[O-].[CH3+]")
nm2 = lfragchooser.choose(mol2)
self.assertEqual(Chem.MolToSmiles(nm2), "O=[N+]([O-])[O-]")
lfragchooser2 = rdMolStandardize.LargestFragmentChooser(
preferOrganic=True)
nm3 = lfragchooser2.choose(mol2)
self.assertEqual(Chem.MolToSmiles(nm3), "[CH3+]")
fragremover = rdMolStandardize.FragmentRemover(skip_if_all_match=True)
mol = Chem.MolFromSmiles("[Na+].Cl.Cl.Br")
nm = fragremover.remove(mol)
self.assertEqual(nm.GetNumAtoms(), mol.GetNumAtoms())
smi3 = "CNC[C@@H]([C@H]([C@@H]([C@@H](CO)O)O)O)O.c1cc2c(cc1C(=O)O)oc(n2)c3cc(cc(c3)Cl)Cl"
lfParams = rdMolStandardize.CleanupParameters()
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol3 = Chem.MolFromSmiles(smi3)
lfrag3 = lfrag_params.choose(mol3)
self.assertEqual(Chem.MolToSmiles(lfrag3),
"CNC[C@H](O)[C@@H](O)[C@H](O)[C@H](O)CO")
lfParams = rdMolStandardize.CleanupParameters()
lfParams.largestFragmentChooserCountHeavyAtomsOnly = True
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol3 = Chem.MolFromSmiles(smi3)
lfrag3 = lfrag_params.choose(mol3)
self.assertEqual(Chem.MolToSmiles(lfrag3),
"O=C(O)c1ccc2nc(-c3cc(Cl)cc(Cl)c3)oc2c1")
lfParams = rdMolStandardize.CleanupParameters()
lfParams.largestFragmentChooserUseAtomCount = False
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol3 = Chem.MolFromSmiles(smi3)
lfrag3 = lfrag_params.choose(mol3)
self.assertEqual(Chem.MolToSmiles(lfrag3),
"O=C(O)c1ccc2nc(-c3cc(Cl)cc(Cl)c3)oc2c1")
smi4 = "CC.O=[Pb]=O"
lfParams = rdMolStandardize.CleanupParameters()
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol4 = Chem.MolFromSmiles(smi4)
lfrag4 = lfrag_params.choose(mol4)
self.assertEqual(Chem.MolToSmiles(lfrag4), "CC")
lfParams = rdMolStandardize.CleanupParameters()
lfParams.largestFragmentChooserCountHeavyAtomsOnly = True
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol4 = Chem.MolFromSmiles(smi4)
lfrag4 = lfrag_params.choose(mol4)
self.assertEqual(Chem.MolToSmiles(lfrag4), "O=[Pb]=O")
lfParams = rdMolStandardize.CleanupParameters()
lfParams.largestFragmentChooserUseAtomCount = False
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol4 = Chem.MolFromSmiles(smi4)
lfrag4 = lfrag_params.choose(mol4)
self.assertEqual(Chem.MolToSmiles(lfrag4), "O=[Pb]=O")
lfParams = rdMolStandardize.CleanupParameters()
lfParams.largestFragmentChooserCountHeavyAtomsOnly = True
lfParams.preferOrganic = True
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol4 = Chem.MolFromSmiles(smi4)
lfrag4 = lfrag_params.choose(mol4)
self.assertEqual(Chem.MolToSmiles(lfrag4), "CC")
lfParams = rdMolStandardize.CleanupParameters()
lfParams.largestFragmentChooserUseAtomCount = False
lfParams.preferOrganic = True
lfrag_params = rdMolStandardize.LargestFragmentChooser(lfParams)
mol4 = Chem.MolFromSmiles(smi4)
lfrag4 = lfrag_params.choose(mol4)
self.assertEqual(Chem.MolToSmiles(lfrag4), "CC")
Trivalent O [*:1]=[O;X2;v3;+0:2]-[#6:3]>>[*:1]=[*+1:2]-[*:3]
Sulfoxide to -S+(O-) [!O:1][S+0;D3:2](=[O:3])[!O:4]>>[*:1][S+1:2]([O-:3])[*:4]
// this form addresses a pathological case that came up a few times in testing:
Sulfoxide to -S+(O-) 2 [!O:1][SH1+1;D3:2](=[O:3])[!O:4]>>[*:1][S+1:2]([O-:3])[*:4]
Trivalent S [O:1]=[S;D2;+0:2]-[#6:3]>>[*:1]=[*+1:2]-[*:3]
// Note that the next one doesn't work propertly because repeated appplications
// don't carry the cations from the previous rounds through. This should be
// fixed by implementing single-molecule transformations, but that's a longer-term
// project
//Alkaline oxide to ions [Li,Na,K;+0:1]-[O+0:2]>>([*+1:1].[O-:2])
Bad amide tautomer1 [C:1]([OH1;D1:2])=;!@[NH1:3]>>[C:1](=[OH0:2])-[NH2:3]
Bad amide tautomer2 [C:1]([OH1;D1:2])=;!@[NH0:3]>>[C:1](=[OH0:2])-[NH1:3]
Halogen with no neighbors [F,Cl,Br,I;X0;+0:1]>>[*-1:1]
Odd pyridine/pyridazine oxide structure [C,N;-;D2,D3:1]-[N+2;D3:2]-[O-;D1:3]>>[*-0:1]=[*+1:2]-[*-:3]
"""
_normalizer_params = rdMolStandardize.CleanupParameters()
_normalizer = rdMolStandardize.NormalizerFromData(_normalization_transforms,
_normalizer_params)
_alkoxide_pattern = Chem.MolFromSmarts('[Li,Na,K;+0]-[#7,#8;+0]')
def normalize_mol(m):
"""
"""
Chem.FastFindRings(m)
if m.HasSubstructMatch(_alkoxide_pattern):
m = Chem.RWMol(m)
for match in m.GetSubstructMatches(_alkoxide_pattern):
m.RemoveBond(match[0], match[1])
m.GetAtomWithIdx(match[0]).SetFormalCharge(1)
