def get_fragments(insmiles):
"""Get core and sidechains using Murcko fragmentation method
Args:
insmiles (str): SMILES for molecule
Returns:
core: scaffold SMILES
side: side chain SMILES
"""
mol_prev = Chem.MolFromSmiles(insmiles)
### get core using Murcko fragmentation ###
core = Murcko.MurckoScaffoldSmilesFromSmiles(insmiles)
if core != "":
mol_core = Chem.MolFromSmiles(core)
### get sidechains ###
mol_side = Chem.rdmolops.DeleteSubstructs(mol_prev, mol_core)
side = Chem.MolToSmiles(mol_side).split(".")
else:
side = [insmiles]
return core, side
def main(name, argv):
if len(argv) != 2:
print_usage(name)
return
with open(argv[0], 'r') as f:
smile = f.readline().split()[0]
mol = Chem.MolFromSmiles(smile)
size = mol.GetNumHeavyAtoms()
hierarch = Recap.RecapDecompose(mol)
children = []
for child in hierarch.GetAllChildren().keys() + [smile]:
new_smiles = child.replace('[*]', '[H]')
new = Chem.MolFromSmiles(new_smiles)
if not new == None:
new_size = new.GetNumHeavyAtoms()
if new_size > 7 and new_size <= 25:
if rdMolDescriptors.CalcNumRotatableBonds(new) <= 5:
children.append(Chem.MolToSmiles(new, isomericSmiles=True))
#children.append(new_smiles)
core_smile = MurckoScaffold.MurckoScaffoldSmilesFromSmiles(
new_smiles, includeChirality=True)
core = Chem.MolFromSmiles(core_smile)
if rdMolDescriptors.CalcNumRotatableBonds(
core) <= 5 and core.GetNumHeavyAtoms() > 7:
children.append(core_smile)
with open(argv[1], 'w') as f:
i = 1
for m in set(children):
if len(m) > 0:
f.write(m + '\t' + str(i) + '\n')
i += 1
def build_library(in_smile,
frags,
lib,
rules=os.environ["COVALIB"] +
"/Code/Covalentizer/numbered_reaction.re",
linker_lib=False,
linker_smiles=''):
argv = [in_smile, frags, lib]
with open(argv[0], 'r') as f:
smile = f.readline().split()[0]
mol = Chem.MolFromSmiles(smile)
if mol == None:
return
size = mol.GetNumHeavyAtoms()
hierarch = Recap.RecapDecompose(mol)
children = []
for child in hierarch.GetAllChildren().keys() + [smile]:
new_smiles = child.replace('[*]', '[H]')
new = Chem.MolFromSmiles(new_smiles)
if not new == None:
new_size = new.GetNumHeavyAtoms()
if new_size > 7:
if new_size <= 25 and rdMolDescriptors.CalcNumRotatableBonds(
new) <= 5:
children.append(Chem.MolToSmiles(new, isomericSmiles=True))
core_smile = MurckoScaffold.MurckoScaffoldSmilesFromSmiles(
new_smiles, includeChirality=True)
core = Chem.MolFromSmiles(core_smile)
if new_size <= 25 and rdMolDescriptors.CalcNumRotatableBonds(
core) <= 5 and core.GetNumHeavyAtoms() > 7:
children.append(core_smile)
with open(argv[1], 'w') as f:
i = 1
for m in set(children):
if len(m) > 0:
f.write(m + '\t' + str(i) + '\n')
i += 1
if not linker_lib:
multi_react([rules, argv[1], argv[2]])
else:
multi_linkers([rules, argv[1], argv[2]], linker_smiles)
def calculate_Murcko_scaffold_smiles(self, binding_sites):
"""
Given a number of binding sites with ligands in them, returns the Murcko Scaffolds of the ligands by binding site
:param binding_sites:
:return:
"""
from rdkit.Chem.Scaffolds import MurckoScaffold
# Calculate Murcko scaffolds for each of the binding site ligands
murcko_dict = []
# Note: next step assumes the binding sites have unique identifiers: the output of self.get_ensemble_binding_sites() would.
for b in binding_sites:
murckos = [
MurckoScaffold.MurckoScaffoldSmilesFromSmiles(l.smiles)
for l in b.ligands
]
murcko_dict[b.identifier] = murckos
return murcko_dict
def get_scaffold(self, smile: str) -> str:
return MS.MurckoScaffoldSmilesFromSmiles(smile)
