def canonicalize(self, mol):
"""Return a canonical tautomer by enumerating and scoring all possible tautomers.
:param mol: The input molecule.
:type mol: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
:return: The canonical tautomer.
:rtype: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
"""
# TODO: Overload the mol parameter to pass a list of pre-enumerated tautomers
tautomers = self._enumerate_tautomers(mol)
if len(tautomers) == 1:
return tautomers[0]
# Calculate score for each tautomer
highest = None
for t in tautomers:
smiles = Chem.MolToSmiles(t, isomericSmiles=True)
logger.debug('Tautomer: %s', smiles)
score = 0
# Add aromatic ring scores
ssr = Chem.GetSymmSSSR(t)
for ring in ssr:
btypes = {
t.GetBondBetweenAtoms(*pair).GetBondType()
for pair in pairwise(ring)
}
elements = {
t.GetAtomWithIdx(idx).GetAtomicNum()
for idx in ring
}
if btypes == {BondType.AROMATIC}:
logger.debug('Score +100 (aromatic ring)')
score += 100
if elements == {6}:
logger.debug('Score +150 (carbocyclic aromatic ring)')
score += 150
# Add SMARTS scores
for tscore in self.scores:
for match in t.GetSubstructMatches(tscore.smarts):
logger.debug('Score %+d (%s)', tscore.score, tscore.name)
score += tscore.score
# Add (P,S,Se,Te)-H scores
for atom in t.GetAtoms():
if atom.GetAtomicNum() in {15, 16, 34, 52}:
hs = atom.GetTotalNumHs()
if hs:
logger.debug('Score %+d (%s-H bonds)', -hs,
atom.GetSymbol())
score -= hs
# Set as highest if score higher or if score equal and smiles comes first alphabetically
if not highest or highest['score'] < score or (
highest['score'] == score and smiles < highest['smiles']):
logger.debug('New highest tautomer: %s (%s)', smiles, score)
highest = {'smiles': smiles, 'tautomer': t, 'score': score}
return highest['tautomer']
def canonicalize(self, mol):
"""Return a canonical tautomer by enumerating and scoring all possible tautomers.
:param mol: The input molecule.
:type mol: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
:return: The canonical tautomer.
:rtype: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
"""
# TODO: Overload the mol parameter to pass a list of pre-enumerated tautomers
tautomers = self._enumerate_tautomers(mol)
if len(tautomers) == 1:
return tautomers[0]
# Calculate score for each tautomer
highest = None
for t in tautomers:
smiles = Chem.MolToSmiles(t, isomericSmiles=True)
logger.debug('Tautomer: %s', smiles)
score = 0
# Add aromatic ring scores
ssr = Chem.GetSymmSSSR(t)
for ring in ssr:
btypes = {t.GetBondBetweenAtoms(*pair).GetBondType() for pair in pairwise(ring)}
elements = {t.GetAtomWithIdx(idx).GetAtomicNum() for idx in ring}
if btypes == {BondType.AROMATIC}:
logger.debug('Score +100 (aromatic ring)')
score += 100
if elements == {6}:
logger.debug('Score +150 (carbocyclic aromatic ring)')
score += 150
# Add SMARTS scores
for tscore in self.scores:
for _ in t.GetSubstructMatches(tscore.smarts):
logger.debug('Score %+d (%s)', tscore.score, tscore.name)
score += tscore.score
# Add (P,S,Se,Te)-H scores
for atom in t.GetAtoms():
if atom.GetAtomicNum() in {15, 16, 34, 52}:
hs = atom.GetTotalNumHs()
if hs:
logger.debug('Score %+d (%s-H bonds)', -hs, atom.GetSymbol())
score -= hs
# Set as highest if score higher or if score equal and smiles comes first alphabetically
if not highest or highest['score'] < score or (highest['score'] == score and smiles < highest['smiles']):
logger.debug('New highest tautomer: %s (%s)', smiles, score)
highest = {'smiles': smiles, 'tautomer': t, 'score': score}
return highest['tautomer']
def compute_score(self, mol, returndetails=False, log=False):
"""
Return a canonical tautomer by enumerating and scoring all possible tautomers.
:param mol: The input molecule.
:type mol: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
:return: The canonical tautomer.
:rtype: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
"""
t_scores = []
scores_detail = []
t_depict = []
if isinstance(mol, SmallMol):
mol = mol._mol
# TODO: Overload the mol parameter to pass a list of pre-enumerated tautomers
tautomers = self._enumerate_tautomers(mol)
for t in tautomers:
tmp_score_details = {
'ArRing': 0,
'CarbArRing': 0,
'MatchFeature': [0, []],
'Penalty': [0, []],
'Conjugate': 0
}
smiles = Chem.MolToSmiles(t, isomericSmiles=True)
if log:
print('Tautomer: %s', smiles)
score = 0
# Add aromatic ring scores
ssr = Chem.GetSymmSSSR(t)
for ring in ssr:
btypes = {
t.GetBondBetweenAtoms(*pair).GetBondType()
for pair in pairwise(ring)
}
elements = {
t.GetAtomWithIdx(idx).GetAtomicNum()
for idx in ring
}
if btypes == {BondType.AROMATIC}:
if log:
print('Score +100 (aromatic ring)')
score += 100
tmp_score_details['ArRing'] += 1
if elements == {6}:
if log:
print('Score +150 (carbocyclic aromatic ring)')
score += 150
tmp_score_details['CarbArRing'] += 1
# Add SMARTS scores, Chem.MolToSmiles(t))
for tscore in self.scores:
for match in t.GetSubstructMatches(tscore.smarts):
if log:
print('Score %+d (%s)' % (tscore.score, tscore.name))
score += tscore.score
tmp_score_details['MatchFeature'][0] += 1
tmp_score_details['MatchFeature'][1].append(tscore.name)
# Add (P,S,Se,Te)-H scores
for atom in t.GetAtoms():
if atom.GetAtomicNum() in {15, 16, 34, 52}:
hs = atom.GetTotalNumHs()
if hs:
if log:
print('Score %+d (%s-H bonds)' %
(-hs, atom.GetSymbol()))
score -= hs
tmp_score_details['Penalty'][0] += 1
tmp_score_details['Penalty'][1].append(
atom.GetSymbol())
# compute the conjuggate system
n_conjugate, depictionatoms = self.get_conjugate(t)
t_depict.append(depictionatoms)
tmp_score_details['Conjugate'] = n_conjugate
score += n_conjugate * 2
scores_detail.append(tmp_score_details)
t_scores.append(score)
if returndetails:
return [SmallMol(tautomer) for tautomer in tautomers
], t_scores, t_depict, scores_detail
return tautomers, t_scores
def compute_score(self, mol, returndetails=False, log=False):
"""
Return a canonical tautomer by enumerating and scoring all possible tautomers.
:param mol: The input molecule.
:type mol: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
:return: The canonical tautomer.
:rtype: :rdkit:`Mol <Chem.rdchem.Mol-class.html>`
"""
t_scores = []
scores_detail = []
t_depict = []
if isinstance(mol, SmallMol):
mol = mol._mol
# TODO: Overload the mol parameter to pass a list of pre-enumerated tautomers
tautomers = self._enumerate_tautomers(mol)
for t in tautomers:
tmp_score_details = {'ArRing': 0,
'CarbArRing': 0,
'MatchFeature': [0, []],
'Penalty': [0, []],
'Conjugate': 0}
smiles = Chem.MolToSmiles(t, isomericSmiles=True)
if log:
print('Tautomer: %s', smiles)
score = 0
# Add aromatic ring scores
ssr = Chem.GetSymmSSSR(t)
for ring in ssr:
btypes = {t.GetBondBetweenAtoms(*pair).GetBondType() for pair in pairwise(ring)}
elements = {t.GetAtomWithIdx(idx).GetAtomicNum() for idx in ring}
if btypes == {BondType.AROMATIC}:
if log:
print('Score +100 (aromatic ring)')
score += 100
tmp_score_details['ArRing'] += 1
if elements == {6}:
if log:
print('Score +150 (carbocyclic aromatic ring)')
score += 150
tmp_score_details['CarbArRing'] += 1
# Add SMARTS scores, Chem.MolToSmiles(t))
for tscore in self.scores:
for match in t.GetSubstructMatches(tscore.smarts):
if log:
print('Score %+d (%s)' % (tscore.score, tscore.name))
score += tscore.score
tmp_score_details['MatchFeature'][0] += 1
tmp_score_details['MatchFeature'][1].append(tscore.name)
# Add (P,S,Se,Te)-H scores
for atom in t.GetAtoms():
if atom.GetAtomicNum() in {15, 16, 34, 52}:
hs = atom.GetTotalNumHs()
if hs:
if log:
print('Score %+d (%s-H bonds)' % (-hs, atom.GetSymbol()))
score -= hs
tmp_score_details['Penalty'][0] += 1
tmp_score_details['Penalty'][1].append(atom.GetSymbol())
# compute the conjuggate system
n_conjugate, depictionatoms = self.get_conjugate(t)
t_depict.append(depictionatoms)
tmp_score_details['Conjugate'] = n_conjugate
score += n_conjugate * 2
scores_detail.append(tmp_score_details)
t_scores.append(score)
if returndetails:
return [SmallMol(tautomer) for tautomer in tautomers], t_scores, t_depict, scores_detail
return tautomers, t_scores