jebus/enviPy-bayer
1
0
Fork 0
forked from enviPath/enviPy
Code Issues Pull Requests Actions Releases Activity

[Misc] Performance improvements, SMIRKS Coverage, Minor Bugfixes (#132)

Browse Source 
Bump Python Version to 3.12
Make use of "epauth" optional
Cache `srs` property of rules to speed up apply
Adjust view names for use of `reverse()`
Fix Views for Scenario Attachments
Added Simply Compare View/Template to identify differences between rdkit and ambit
Make migrations consistent with tests + compare
Fixes #76
Set default year for Scenario Modal
Fix html tags for package description
Added Tests for Pathway / Rule
Added remove stereo for apply

Co-authored-by: Tim Lorsbach <tim@lorsba.ch>
Reviewed-on: enviPath/enviPy#132
This commit is contained in:
jebus
2025-09-26 19:33:03 +12:00
parent b5c759d74e
commit b757a07f91
23 changed files with 671 additions and 463 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
utilities/chem.py
View File

@ -244,7 +244,7 @@ class FormatConverter(object):
@staticmethod
def apply(smiles: str, smirks: str, preprocess_smiles: bool = True, bracketize: bool = True,
standardize: bool = True, kekulize: bool = True) -> List['ProductSet']:
standardize: bool = True, kekulize: bool = True, remove_stereo: bool = True) -> List['ProductSet']:
logger.debug(f'Applying {smirks} on {smiles}')
# If explicitly wanted or rule generates multiple products add brackets around products to capture all
@ -274,7 +274,7 @@ class FormatConverter(object):
Chem.SanitizeMol(product)
product = GetMolFrags(product, asMols=True)
for p in product:
p = FormatConverter.standardize(Chem.MolToSmiles(p))
p = FormatConverter.standardize(Chem.MolToSmiles(p), remove_stereo=remove_stereo)
prods.append(p)
# if kekulize:
@ -314,30 +314,10 @@ class FormatConverter(object):
return pss
# @staticmethod
# def apply(reaction, smiles):
# rxn = AllChem.ReactionFromSmarts(reaction)
# return [Chem.MolToSmiles(x, 1) for x in rxn.RunReactants((Chem.MolFromSmiles(smiles),))[0]]
@staticmethod
def MACCS(smiles):
return MACCSkeys.GenMACCSKeys(FormatConverter.from_smiles(smiles))
@staticmethod
def neutralize_atoms(mol):
pattern = Chem.MolFromSmarts("[+1!h0!$([*]~[-1,-2,-3,-4]),-1!$([*]~[+1,+2,+3,+4])]")
at_matches = mol.GetSubstructMatches(pattern)
at_matches_list = [y[0] for y in at_matches]
if len(at_matches_list) > 0:
for at_idx in at_matches_list:
atom = mol.GetAtomWithIdx(at_idx)
chg = atom.GetFormalCharge()
hcount = atom.GetTotalNumHs()
atom.SetFormalCharge(0)
atom.SetNumExplicitHs(hcount - chg)
atom.UpdatePropertyCache()
return mol
@staticmethod
def sanitize_smiles(smiles_list: List):
parsed_smiles = []
@ -353,7 +333,8 @@ class FormatConverter(object):
# smi = smi.replace("@", "")
mol = Chem.MolFromSmiles(smi)
mol = FormatConverter.neutralize_atoms(mol)
mol = FormatConverter.neutralize_molecule(mol)
Chem.RemoveStereochemistry(mol)
mol = Chem.RemoveAllHs(mol)
Chem.Kekulize(mol)
smi_p = Chem.MolToSmiles(mol, kekuleSmiles=True)