def testCondense(self):
m = Chem.MolFromSmiles('FC(F)(F)CC(=O)O')
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.defaultAbbrevs, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*C* |$CF3;;CO2H$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (1, 4, 5))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (3, 4))
m = Chem.MolFromSmiles('CCC(F)(F)F')
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.defaultAbbrevs)
self.assertEqual(Chem.MolToCXSmiles(nm), '*C(F)(F)F |$Et;;;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (1, 2, 3, 4, 5))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (1, 2, 3, 4))
# make sure we don't mess up chirality
m = Chem.MolFromSmiles('FC(F)(F)[C@](Cl)(F)I')
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.defaultAbbrevs, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*[C@@](F)(Cl)I |$CF3;;;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (1, 4, 5, 6, 7))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (3, 4, 5, 6))
def testAbbreviationsAndLinkers(self):
m = Chem.MolFromSmiles('COC1CCC(C)CC1')
# wouldn't normally do this in this order:
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.defaultAbbrevs, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*C1CCC(C)CC1 |$OMe;;;;;;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (1, 2, 3, 4, 5, 6, 7, 8))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (1, 2, 3, 4, 5, 6, 7, 8))
nm = rdAbbreviations.CondenseMolAbbreviations(nm, self.customLinkers, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '**C |$OMe;Cy;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (1, 2, 6))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (1, 5))
# This is a more logical order
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.customLinkers, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'C*OC |$;Cy;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (0, 1, 2, 6))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (0, 1, 5))
nm = rdAbbreviations.CondenseMolAbbreviations(nm, self.defaultAbbrevs, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'C*OC |$;Cy;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (0, 1, 2, 6))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (0, 1, 5))
def testCondenseLinkers(self):
m = Chem.MolFromSmiles('FCOCCOCCOCCCCCCCCCCl')
nm = rdAbbreviations.CondenseMolAbbreviations(m,
self.defaultLinkers,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'FC**Cl |$;;PEG3;Hept;$|')
m = Chem.MolFromSmiles('COC1CCC(C)CC1')
nm = rdAbbreviations.CondenseMolAbbreviations(m,
self.customLinkers,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'C*OC |$;Cy;;$|')
def testCondense(self):
m = Chem.MolFromSmiles('FC(F)(F)CC(=O)O')
nm = rdAbbreviations.CondenseMolAbbreviations(m,
self.defaultAbbrevs,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*C* |$CF3;;CO2H$|')
m = Chem.MolFromSmiles('CCC(F)(F)F')
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.defaultAbbrevs)
self.assertEqual(Chem.MolToCXSmiles(nm), '*C(F)(F)F |$Et;;;;$|')
# make sure we don't mess up chirality
m = Chem.MolFromSmiles('FC(F)(F)[C@](Cl)(F)I')
nm = rdAbbreviations.CondenseMolAbbreviations(m,
self.defaultAbbrevs,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*[C@@](F)(Cl)I |$CF3;;;;$|')
def testCondenseLinkers(self):
m = Chem.MolFromSmiles('FCOCCOCCOCCCCCCCCCCl')
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.defaultLinkers, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'FC**Cl |$;;PEG3;Hept;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (0, 1, 2, 11, 18))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (0, 1, 10, 17))
m = Chem.MolFromSmiles('COC1CCC(C)CC1')
nm = rdAbbreviations.CondenseMolAbbreviations(m, self.customLinkers, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'C*OC |$;Cy;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (0, 1, 2, 6))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (0, 1, 5))
def testParsingAbbrevs(self):
defn = '''CO2Et C(=O)OCC
COOEt C(=O)OCC
OiBu OCC(C)C
tBu C(C)(C)C'''
abbrevs = rdAbbreviations.ParseAbbreviations(defn)
m = Chem.MolFromSmiles('CCC(=O)OCC')
nm = rdAbbreviations.CondenseMolAbbreviations(m,
abbrevs,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*CC |$CO2Et;;$|')
def testAbbreviationsAndLinkers(self):
m = Chem.MolFromSmiles('COC1CCC(C)CC1')
# wouldn't normally do this in this order:
nm = rdAbbreviations.CondenseMolAbbreviations(m,
self.defaultAbbrevs,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*C1CCC(C)CC1 |$OMe;;;;;;;$|')
nm = rdAbbreviations.CondenseMolAbbreviations(nm,
self.customLinkers,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '**C |$OMe;Cy;$|')
# This is a more logical order
nm = rdAbbreviations.CondenseMolAbbreviations(m,
self.customLinkers,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'C*OC |$;Cy;;$|')
nm = rdAbbreviations.CondenseMolAbbreviations(nm,
self.defaultAbbrevs,
maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), 'C*OC |$;Cy;;$|')
def testParsingAbbrevs(self):
defn = '''CO2Et C(=O)OCC
COOEt C(=O)OCC
OiBu OCC(C)C
tBu C(C)(C)C'''
abbrevs = rdAbbreviations.ParseAbbreviations(defn)
m = Chem.MolFromSmiles('CCC(=O)OCC')
nm = rdAbbreviations.CondenseMolAbbreviations(m, abbrevs, maxCoverage=1.0)
self.assertEqual(Chem.MolToCXSmiles(nm), '*CC |$CO2Et;;$|')
self.assertTrue(nm.HasProp('_origAtomMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origAtomMapping')), (0, 1, 2))
self.assertTrue(nm.HasProp('_origBondMapping'))
self.assertEqual(self.parseMapping(nm.GetProp('_origBondMapping')), (0, 1))
def _drawHelper(mol, drawer, **kwargs):
tgt = request.get_json()
if tgt is None:
tgt = request.values
sanit = _stringtobool(tgt.get('sanitize', True))
kekulize = _stringtobool(tgt.get('kekulize', True))
for arg in ('highlightAtomColors', 'highlightBondColors'):
if arg not in kwargs:
tmp = _loadJSONParam(tgt.get(arg, None))
if tmp:
for k in list(tmp):
tmp[int(k)] = tuple(tmp[k])
del tmp[k]
kwargs[arg] = tmp
for arg in ('highlightAtoms', 'highlightBonds'):
if arg not in kwargs:
tmp = _loadJSONParam(tgt.get(arg, None))
if tmp:
kwargs[arg] = tmp
if 'legend' not in kwargs:
kwargs['legend'] = tgt.get('legend', '')
# if 'lw' not in kwargs and 'lw' in request.values:
# kwargs['lw'] = int(request.values['lw'])
if 'highlightSubstruct' not in kwargs:
if 'highlightColor' not in kwargs:
hcolor = _loadJSONParam(tgt.get('highlightColor', None))
if hcolor:
highlightColor = tuple(hcolor)
else:
highlightColor = None
else:
highlightColor = kwargs['highlightColor']
del kwargs['higlightColor']
if _stringtobool(tgt.get('highlightSubstruct', False)) or \
'smiles_highlight' in tgt or \
'smarts_highlight' in tgt or \
'mol_highlight' in tgt:
qmol = _queryfromrequest(suffix='_highlight')
if qmol is not None:
qMatches = mol.GetSubstructMatches(qmol)
highlights = kwargs.get('highlightAtoms', [])
if highlightColor is not None:
highlightBonds = kwargs.get('highlightBonds', [])
else:
highlightBonds = None
for entry in qMatches:
if highlightColor is not None:
had = kwargs.get('highlightAtomColors', {})
for v in entry:
had[v] = highlightColor
kwargs['highlightAtomColors'] = had
hbd = kwargs.get('highlightBondColors', {})
emap = dict(enumerate(entry))
for bond in qmol.GetBonds():
mbond = mol.GetBondBetweenAtoms(
emap[bond.GetBeginAtomIdx()],
emap[bond.GetEndAtomIdx()])
highlightBonds.append(mbond.GetIdx())
hbd[mbond.GetIdx()] = highlightColor
kwargs['highlightBondColors'] = hbd
highlights.extend(list(entry))
if highlights:
kwargs['highlightAtoms'] = highlights
if highlightBonds is not None:
kwargs['highlightBonds'] = highlightBonds
from rdkit.Chem.Draw import rdDepictor
rdDepictor.SetPreferCoordGen(True)
# print(Chem.MolToMolBlock(mc))
drawo = drawer.drawOptions()
for opt, default in (('dummiesAreAttachments', False),
('comicMode', False), ('addAtomIndices', False),
('addBondIndices', False), ('isotopeLabels', True),
('addStereoAnnotation',
False), ('explicitMethyl',
False), ('includeChiralFlagLabel', False),
('simplifiedStereoGroupLabel', False)):
setattr(drawo, opt, _stringtobool(tgt.get(opt, default)))
if drawo.comicMode:
drawo.fontFile = os.path.join(RDConfig.RDDataDir, "Fonts",
"ComicNeue-Regular.ttf")
if drawo.addStereoAnnotation:
Chem.FindMolChiralCenters(mol,
force=True,
useLegacyImplementation=False)
if 'backgroundColor' in tgt:
tmp = _loadJSONParam(tgt.get('backgroundColor', None))
drawo.SetBackgroundColor(tuple)
if _stringtobool(tgt.get('atomIndices', False)):
drawo.addAtomIndices = True
if _stringtobool(tgt.get('bondIndices', False)):
drawo.addBondIndices = True
if _stringtobool(tgt.get('useBW', False)):
drawo.useBWAtomPalette()
if _stringtobool(tgt.get('useGrid', '0')):
frags = []
for frag in Chem.GetMolFrags(mol, asMols=True):
frags.append(
rdMolDraw2D.PrepareMolForDrawing(frag,
kekulize=kekulize & sanit,
addChiralHs=sanit))
drawer.DrawMolecules(frags)
elif _stringtobool(tgt.get('asRxn', False)):
drawer.DrawReaction(mol)
else:
mc = rdMolDraw2D.PrepareMolForDrawing(mol,
kekulize=kekulize & sanit,
addChiralHs=sanit)
mc.SetProp("_Name", "temp")
if _stringtobool(tgt.get('collapseAbbreviations',False)) and \
len(Chem.GetMolSubstanceGroups(mc)):
mc = rdAbbreviations.CondenseAbbreviationSubstanceGroups(mc)
elif _stringtobool(tgt.get('findAbbreviations', False)):
mc = rdAbbreviations.CondenseMolAbbreviations(
mc,
rdAbbreviations.GetDefaultAbbreviations(),
maxCoverage=float(tgt.get('maxAbbreviationCoverage', 0.4)))
drawer.DrawMolecule(mc, **kwargs)
drawer.FinishDrawing()