def test_AlignDepict(self):
mol = Chem.MolFromSmiles('CNC')
core = Chem.MolFromSmiles('CC')
pattern = Chem.MolFromSmarts('CCC')
self.assertRaises(ValueError, AlignDepict, mol, core, pattern)
pattern = Chem.MolFromSmarts('CN')
self.assertRaises(ValueError, AlignDepict, mol, core, pattern)
pattern = Chem.MolFromSmarts('CC')
self.assertRaises(ValueError, AlignDepict, mol, core, pattern)
pattern = Chem.MolFromSmarts('CC')
self.assertRaises(ValueError, AlignDepict, mol, core, pattern)
mol = Chem.MolFromSmiles('CCC')
Chem.rdDepictor.Compute2DCoords(core)
AlignDepict(mol, core, pattern)
mol = Chem.MolFromSmiles('CNC')
AlignDepict(mol, core, pattern, acceptFailure=True)
'name', 'alternate_names', 'mol', 'smiles',
'name_rt', 'alternate_names_rt', 'mol_rt', 'smiles_rt']]
# In[73]:
if len(name_conflicts) == 0:
display(HTML('<h2 style="color: green">No name conflicts!</h2>'))
for _, r in name_conflicts.iterrows():
mol1 = r['mol']
mol2 = r['mol_rt']
mcs_result = rdFMCS.FindMCS([mol1, mol2], ringMatchesRingOnly=True, matchValences=True)
mcs = AllChem.MolFromSmarts(mcs_result.smartsString)
AllChem.Compute2DCoords(mcs)
AlignDepict(mol1, mcs)
AlignDepict(mol2, mcs)
alt = '(' + ', '.join(r['alternate_names']) + ')' if r['alternate_names'] else ''
alt_rt = '(' + ', '.join(r['alternate_names_rt']) + ')' if r['alternate_names_rt'] else ''
mol1_match = mol1.GetSubstructMatch(mcs)
mol2_match = mol2.GetSubstructMatch(mcs)
std_args = {'highlightColor': [0, 0.6, 0.6], 'kekulize': False, 'size': (500, 300)}
display(Box([
VBox([Image(value=Draw.MolToImage(mol1, highlightAtoms=mol1_match, **std_args)._repr_png_()),
Label('{name} {alt} : {broad_id}'.format(alt=alt, **r))]),
VBox([Image(value=Draw.MolToImage(mol2, highlightAtoms=mol2_match, **std_args)._repr_png_()),
Label('{name_rt} {alt} : {lincs_id}'.format(alt=alt_rt, **r))]),
]))
print(r['smiles'])
print(r['smiles_rt'])
def _exploder(mol,
depth,
sidechains,
core,
chainIndices,
autoNames=True,
templateName='',
resetCounter=True,
do3D=False,
useTethers=False):
global nDumped
if resetCounter:
nDumped = 0
ourChains = sidechains[depth]
patt = '[%d*]' % (depth + 1)
patt = Chem.MolFromSmiles(patt)
for i, (chainIdx, chain) in enumerate(ourChains):
tchain = chainIndices[:]
tchain.append((i, chainIdx))
rs = Chem.ReplaceSubstructs(mol, patt, chain, replaceAll=True)
if rs:
r = rs[0]
if depth < len(sidechains) - 1:
for entry in _exploder(r,
depth + 1,
sidechains,
core,
tchain,
autoNames=autoNames,
templateName=templateName,
resetCounter=0,
do3D=do3D,
useTethers=useTethers):
yield entry
else:
try:
Chem.SanitizeMol(r)
except ValueError:
import traceback
traceback.print_exc()
continue
if not do3D:
if r.HasSubstructMatch(core):
try:
AlignDepict(r, core)
except:
import traceback
traceback.print_exc()
print(Chem.MolToSmiles(r), file=sys.stderr)
else:
print('>>> no match', file=sys.stderr)
AllChem.Compute2DCoords(r)
else:
r = Chem.AddHs(r)
AllChem.ConstrainedEmbed(r, core, useTethers)
Chem.Kekulize(r)
if autoNames:
tName = "TemplateEnum: Mol_%d" % (nDumped + 1)
else:
tName = templateName
for bbI, bb in enumerate(tchain):
bbMol = sidechains[bbI][bb[0]][1]
if bbMol.HasProp('_Name'):
bbNm = bbMol.GetProp('_Name')
else:
bbNm = str(bb[1])
tName += '_' + bbNm
r.SetProp("_Name", tName)
r.SetProp('seq_num', str(nDumped + 1))
r.SetProp('reagent_indices',
'_'.join([str(x[1]) for x in tchain]))
for bbI, bb in enumerate(tchain):
bbMol = sidechains[bbI][bb[0]][1]
if bbMol.HasProp('_Name'):
bbNm = bbMol.GetProp('_Name')
else:
bbNm = str(bb[1])
r.SetProp('building_block_%d' % (bbI + 1), bbNm)
for propN in bbMol.GetPropNames():
r.SetProp('building_block_%d_%s' % (bbI + 1, propN),
bbMol.GetProp(propN))
nDumped += 1
if not nDumped % 100:
logger.info('Done %d molecules' % nDumped)
yield r