def test_sanitizeSMILES():
"""
Test SMILES sanitization.
"""
smiles_list = [
'CC', 'CCC', '[H][C@]1(NC[C@@H](CC1CO[C@H]2CC[C@@H](CC2)O)N)[H]'
]
sanitized_smiles_list = sanitizeSMILES(smiles_list, mode='drop')
if len(sanitized_smiles_list) != 2:
raise Exception(
"Molecules with undefined stereochemistry are not being properly dropped (size=%d)."
% len(sanitized_smiles_list))
sanitized_smiles_list = sanitizeSMILES(smiles_list, mode='expand')
if len(sanitized_smiles_list) != 4:
raise Exception(
"Molecules with undefined stereochemistry are not being properly expanded (size=%d)."
% len(sanitized_smiles_list))
# Check that all molecules can be round-tripped.
from openeye.oechem import OEGraphMol, OESmilesToMol, OECreateIsoSmiString
for smiles in sanitized_smiles_list:
molecule = OEGraphMol()
OESmilesToMol(molecule, smiles)
isosmiles = OECreateIsoSmiString(molecule)
if (smiles != isosmiles):
raise Exception(
"Molecule '%s' was not properly round-tripped (result was '%s')"
% (smiles, isosmiles))
def __makeChemCompDescriptorCategory(self, ccId, oeMol):
'''
loop_
_pdbx_chem_comp_descriptor.comp_id
_pdbx_chem_comp_descriptor.type
_pdbx_chem_comp_descriptor.program
_pdbx_chem_comp_descriptor.program_version
_pdbx_chem_comp_descriptor.descriptor
ARG SMILES ACDLabs 10.04 "O=C(O)C(N)CCCN\\C(=[NH2+])N"
ARG SMILES_CANONICAL CACTVS 3.341 "N[C@@H](CCCNC(N)=[NH2+])C(O)=O"
ARG SMILES CACTVS 3.341 "N[CH](CCCNC(N)=[NH2+])C(O)=O"
ARG SMILES_CANONICAL "OpenEye OEToolkits" 1.5.0 "C(C[C@@H](C(=O)O)N)CNC(=[NH2+])N"
ARG SMILES "OpenEye OEToolkits" 1.5.0 "C(CC(C(=O)O)N)CNC(=[NH2+])N"
ARG InChI InChI 1.03 "InChI=1S/C6H14N4O2/c7-4(5(11)12)2-1-3-1..... "
ARG InChIKey InChI 1.03 ODKSFYDXXFIFQN-BYPYZUCNSA-O
#
'''
rowL = []
#
aRow = {}
aRow['comp_id'] = ccId
aRow['type'] = 'SMILES_CANONICAL'
aRow['program'] = "OpenEye OEToolkits"
aRow['program_version'] = '2016.2'
aRow['descriptor'] = OECreateIsoSmiString(oeMol)
rowL.append(aRow)
#
aRow = {}
aRow['comp_id'] = ccId
aRow['type'] = 'SMILES'
aRow['program'] = "OpenEye OEToolkits"
aRow['program_version'] = '2016.2'
aRow['descriptor'] = OECreateCanSmiString(oeMol)
rowL.append(aRow)
#
aRow = {}
aRow['comp_id'] = ccId
aRow['type'] = 'InChI'
aRow['program'] = "OpenEye OEToolkits"
aRow['program_version'] = '2016.2'
aRow['descriptor'] = OECreateInChI(oeMol)
rowL.append(aRow)
#
aRow = {}
aRow['comp_id'] = ccId
aRow['type'] = 'InChIKey'
aRow['program'] = "OpenEye OEToolkits"
aRow['program_version'] = '2016.2'
aRow['descriptor'] = OECreateInChIKey(oeMol)
rowL.append(aRow)
#
return rowL
def sanitizeSMILES(smiles_list, mode='drop', verbose=False):
"""
Sanitize set of SMILES strings by ensuring all are canonical isomeric SMILES.
Duplicates are also removed.
Parameters
----------
smiles_list : iterable of str
The set of SMILES strings to sanitize.
mode : str, optional, default='drop'
When a SMILES string that does not correspond to canonical isomeric SMILES is found, select the action to be performed.
'exception' : raise an `Exception`
'drop' : drop the SMILES string
'expand' : expand all stereocenters into multiple molecules
verbose : bool, optional, default=False
If True, print verbose output.
Returns
-------
sanitized_smiles_list : list of str
Sanitized list of canonical isomeric SMILES strings.
Examples
--------
Sanitize a simple list.
>>> smiles_list = ['CC', 'CCC', '[H][C@]1(NC[C@@H](CC1CO[C@H]2CC[C@@H](CC2)O)N)[H]']
Throw an exception if undefined stereochemistry is present.
>>> sanitized_smiles_list = sanitizeSMILES(smiles_list, mode='exception')
Traceback (most recent call last):
...
Exception: Molecule '[H][C@]1(NC[C@@H](CC1CO[C@H]2CC[C@@H](CC2)O)N)[H]' has undefined stereocenters
Drop molecules iwth undefined stereochemistry.
>>> sanitized_smiles_list = sanitizeSMILES(smiles_list, mode='drop')
>>> len(sanitized_smiles_list)
2
Expand molecules iwth undefined stereochemistry.
>>> sanitized_smiles_list = sanitizeSMILES(smiles_list, mode='expand')
>>> len(sanitized_smiles_list)
4
"""
from openeye.oechem import OEGraphMol, OESmilesToMol, OECreateIsoSmiString
sanitized_smiles_set = set()
for smiles in smiles_list:
molecule = OEGraphMol()
OESmilesToMol(molecule, smiles)
if verbose:
molecule.SetTitle(smiles)
oechem.OETriposAtomNames(molecule)
if has_undefined_stereocenters(molecule, verbose=verbose):
if mode == 'drop':
if verbose:
print("Dropping '%s' due to undefined stereocenters." % smiles)
continue
elif mode == 'exception':
raise Exception("Molecule '%s' has undefined stereocenters" % smiles)
elif mode == 'expand':
if verbose:
print('Expanding stereochemistry:')
print('original: %s', smiles)
molecules = enumerate_undefined_stereocenters(molecule, verbose=verbose)
for molecule in molecules:
isosmiles = OECreateIsoSmiString(molecule)
if verbose: print('expanded: %s', isosmiles)
sanitized_smiles_set.add(isosmiles)
else:
# Convert to OpenEye's canonical isomeric SMILES.
isosmiles = OECreateIsoSmiString(molecule)
sanitized_smiles_set.add(isosmiles)
sanitized_smiles_list = list(sanitized_smiles_set)
return sanitized_smiles_list
def getIsoSMILES(self):
""" Return the cannonical stereo SMILES string derived from the current OE molecule.
"""
return OECreateIsoSmiString(self.__oeMol)