:construction: wip commit (13259b56) · Commits · 钟慕尧 / deepchem

deepchem/feat/base_classes.py

+9 −23

Original line number	Diff line number	Diff line
		@@ -73,24 +73,6 @@ class Featurizer(object):
		raise NotImplementedError('Featurizer is not defined.')


		def _featurize_callback(
		featurizer,
		mol_pdb_file,
		protein_pdb_file,
		log_message,
		):
		"""Callback function for apply_async in ComplexFeaturizer.

		This callback function must be defined globally
		because `apply_async` doesn't execute a nested function.

		See the details from the following link.
		https://stackoverflow.com/questions/56533827/pool-apply-async-nested-function-is-not-executed
		"""
		logging.info(log_message)
		return featurizer._featurize(mol_pdb_file, protein_pdb_file)


		class ComplexFeaturizer(object):
		""""
		Abstract class for calculating features for mol/protein complexes.
		@@ -121,7 +103,7 @@ class ComplexFeaturizer(object):
		for i, (mol_file, protein_pdb) in enumerate(zip(mol_files, protein_pdbs)):
		log_message = "Featurizing %d / %d" % (i, len(mol_files))
		results.append(
		pool.apply_async(_featurize_callback,
		pool.apply_async(ComplexFeaturizer._featurize_callback,
		(self, mol_file, protein_pdb, log_message)))
		pool.close()
		features = []
		@@ -149,6 +131,12 @@ class ComplexFeaturizer(object):
		"""
		raise NotImplementedError('Featurizer is not defined.')

		@staticmethod
		def _featurize_callback(featurizer, mol_pdb_file, protein_pdb_file,
		log_message):
		logging.info(log_message)
		return featurizer._featurize(mol_pdb_file, protein_pdb_file)


		class MolecularFeaturizer(Featurizer):
		"""Abstract class for calculating a set of features for a
		@@ -183,8 +171,8 @@ class MolecularFeaturizer(Featurizer):

		Returns
		-------
		A numpy array containing a featurized representation of
		`datapoints`.
		features: np.ndarray
		A numpy array containing a featurized representation of `datapoints`.
		"""
		try:
		from rdkit import Chem
		@@ -266,7 +254,6 @@ class MaterialStructureFeaturizer(Featurizer):
		features: np.ndarray
		A numpy array containing a featurized representation of
		`structures`.

		"""

		structures = list(structures)
		@@ -332,7 +319,6 @@ class MaterialCompositionFeaturizer(Featurizer):
		features: np.ndarray
		A numpy array containing a featurized representation of
		`compositions`.

		"""

		compositions = list(compositions)

deepchem/feat/binding_pocket_features.py

+0 −1

Original line number	Diff line number	Diff line
		@@ -101,6 +101,5 @@ class BindingPocketFeaturizer(Featurizer):
		if residue not in res_map:
		logger.info("Warning: Non-standard residue in PDB file")
		continue
		atomtype = atom_name.split("-")[1]
		all_features[pocket_num, res_map[residue]] += 1
		return all_features

deepchem/feat/complex_featurizers/init.py

0 → 100644

+3 −0

# flake8: noqa

from deepchem.feat.complex_featurizers.atomic_coordinates import NeighborListComplexAtomicCoordinates

from deepchem.feat.complex_featurizers.atomic_coordinates import ComplexNeighborListFragmentAtomicCoordinates

deepchem/feat/molecule_featurizers/init.py

+4 −3

Original line number	Diff line number	Diff line
		# flake8: noqa
		from deepchem.feat.molecule_featurizers.adjacency_fingerprint import AdjacencyFingerprint
		from deepchem.feat.molecule_featurizers.coulomb_matrices import BPSymmetryFunctionInput
		from deepchem.feat.molecule_featurizers.bp_symmetry_function_input import BPSymmetryFunctionInput
		from deepchem.feat.molecule_featurizers.morgan_fingerprint import CircularFingerprint
		from deepchem.feat.molecule_featurizers.coulomb_matrices import CoulombMatrix
		from deepchem.feat.molecule_featurizers.coulomb_matrices import CoulombMatrixEig
		from deepchem.feat.molecule_featurizers.atom_coordinates import NeighborListAtomicCoordinates
		from deepchem.feat.molecule_featurizers.one_hot_featurizer import OneHotFeaturizer
		from deepchem.feat.molecule_featurizers.raw_featurizer import RawFeaturizer
		from deepchem.feat.molecule_featurizers.rdkit_descriptors import RDKitDescriptors
		from deepchem.feat.molecule_featurizers.smiles_featurizers import SmilesToSeq
		from deepchem.feat.molecule_featurizers.smiles_featurizers import SmilesToImage
		from deepchem.feat.molecule_featurizers.smiles_to_image import SmilesToImage
		from deepchem.feat.molecule_featurizers.smiles_to_seq import SmilesToSeq

deepchem/feat/molecule_featurizers/bp_symmetry_function_input.py

0 → 100644

+42 −0

Original line number	Diff line number	Diff line
		import numpy as np

		from deepchem.utils.typing import RDKitMol
		from deepchem.utils.rdkit_utils import get_coordinates_from_mol
		from deepchem.feat.base_classes import MolecularFeaturizer


		class BPSymmetryFunctionInput(MolecularFeaturizer):
		"""Calculate Symmetry Function for each atom in the molecules

		This method is described in [1]_

		References
		----------
		.. [1] Behler, Jörg, and Michele Parrinello. "Generalized neural-network
		representation of high-dimensional potential-energy surfaces." Physical
		review letters 98.14 (2007): 146401.

		Notes
		-----
		This class requires RDKit to be installed.
		"""

		def __init__(self, max_atoms: int):
		"""Initialize this featurizer.

		Parameters
		----------
		max_atoms: int
		The maximum number of atoms expected for molecules this featurizer will
		process.
		"""
		self.max_atoms = max_atoms

		def _featurize(self, mol: RDKitMol) -> np.ndarray:
		coordinates = get_coordinates_from_mol(mol, unit='bohr')
		atom_numbers = np.array([atom.GetAtomicNum() for atom in mol.GetAtoms()])
		atom_numbers = np.expand_dims(atom_numbers, axis=1)
		assert atom_numbers.shape[0] == coordinates.shape[0]
		n_atoms = atom_numbers.shape[0]
		features = np.concatenate([atom_numbers, coordinates], axis=1)
		return np.pad(features, ((0, self.max_atoms - n_atoms), (0, 0)), 'constant')

Admin message