from openpharmacophore import Ligand, Pharmacophore
import openpharmacophore.pharmacophore.ligand_based.common_pharmacophore as cp
[docs]class LigandBasedPharmacophore:
""" Class to store and extract pharmacophores from a set of ligands.
Parameters
----------
ligands : list[Ligand]
"""
# TODO: add exclusion volumes to ligand based model
def __init__(self, ligands):
self._ligands = ligands
self._pharmacophores = [] # type: list[Pharmacophore]
@property
def pharmacophores(self):
return self._pharmacophores
[docs] def extract(self, n_points, min_actives=None, max_pharmacophores=None, *args, **kwargs):
""" Finds and scores common pharmacophores from a set of ligands.
Parameters
----------
n_points : int
Extracted pharmacophores will have this number of pharmacophoric
points.
min_actives : int, optional
Number of ligands that must match a common pharmacophore.
max_pharmacophores : int, optional
Maximum number of pharmacophores to return. If set to null
all found pharmacophores will be returned.
"""
if min_actives is None:
min_actives = len(self._ligands)
# TODO: use chemical features directionality
chem_feats = self._get_chem_feats(self._ligands)
extractor = cp.CommonPharmacophoreFinder(*args, **kwargs)
self._pharmacophores = extractor(chem_feats, n_points, min_actives, max_pharmacophores)
@staticmethod
def _get_chem_feats(ligands):
""" Get the chemical feature of all ligands in the pharmacophore.
Parameters
----------
list[Ligand]
Returns
-------
list[ChemFeatContainer]
"""
chem_feats = []
for ii, lig in enumerate(ligands):
ligand_feats = []
for conf in range(lig.n_conformers):
chem_f = lig.get_chem_feats(conf, indices=False)
ligand_feats.append(chem_f)
chem_feats.append(ligand_feats)
return chem_feats
def __len__(self):
return len(self._pharmacophores)
def __getitem__(self, frame):
"""
Returns
-------
Pharmacophore
"""
return self._pharmacophores[frame]
