def test_pdbqt_to_pdb(self):
"""Test that a PDBQT molecule can be converted back in to PDB."""
xyz, mol = rdkit_util.load_molecule(self.protein_file,
calc_charges=False,
add_hydrogens=False)
with tempfile.TemporaryDirectory() as tmp:
out_pdb = os.path.join(tmp, "mol.pdb")
out_pdbqt = os.path.join(tmp, "mol.pdbqt")
rdkit_util.write_molecule(mol, out_pdb, is_protein=True)
rdkit_util.write_molecule(mol, out_pdbqt, is_protein=True)
pdb_block = pdbqt_utils.pdbqt_to_pdb(out_pdbqt)
from rdkit import Chem
pdb_mol = Chem.MolFromPDBBlock(pdb_block,
sanitize=False,
removeHs=False)
xyz, pdbqt_mol = rdkit_util.load_molecule(out_pdbqt,
add_hydrogens=False,
calc_charges=False)
assert pdb_mol.GetNumAtoms() == pdbqt_mol.GetNumAtoms()
for atom_idx in range(pdb_mol.GetNumAtoms()):
atom1 = pdb_mol.GetAtoms()[atom_idx]
atom2 = pdbqt_mol.GetAtoms()[atom_idx]
assert atom1.GetAtomicNum() == atom2.GetAtomicNum()
def extract_active_site(protein_file, ligand_file, cutoff=4):
"""Extracts a box for the active site."""
protein_coords = rdkit_util.load_molecule(
protein_file, add_hydrogens=False)[0]
ligand_coords = rdkit_util.load_molecule(
ligand_file, add_hydrogens=True, calc_charges=True)[0]
num_ligand_atoms = len(ligand_coords)
num_protein_atoms = len(protein_coords)
pocket_inds = []
pocket_atoms = set([])
for lig_atom_ind in range(num_ligand_atoms):
lig_atom = ligand_coords[lig_atom_ind]
for protein_atom_ind in range(num_protein_atoms):
protein_atom = protein_coords[protein_atom_ind]
if np.linalg.norm(lig_atom - protein_atom) < cutoff:
if protein_atom_ind not in pocket_atoms:
pocket_atoms = pocket_atoms.union(set([protein_atom_ind]))
# Should be an array of size (n_pocket_atoms, 3)
pocket_atoms = list(pocket_atoms)
n_pocket_atoms = len(pocket_atoms)
pocket_coords = np.zeros((n_pocket_atoms, 3))
for ind, pocket_ind in enumerate(pocket_atoms):
pocket_coords[ind] = protein_coords[pocket_ind]
x_min = int(np.floor(np.amin(pocket_coords[:, 0])))
x_max = int(np.ceil(np.amax(pocket_coords[:, 0])))
y_min = int(np.floor(np.amin(pocket_coords[:, 1])))
y_max = int(np.ceil(np.amax(pocket_coords[:, 1])))
z_min = int(np.floor(np.amin(pocket_coords[:, 2])))
z_max = int(np.ceil(np.amax(pocket_coords[:, 2])))
return (((x_min, x_max), (y_min, y_max), (z_min, z_max)), pocket_atoms,
pocket_coords)
def extract_active_site(protein_file, ligand_file, cutoff=4):
"""Extracts a box for the active site."""
protein_coords = rdkit_util.load_molecule(
protein_file, add_hydrogens=False)[0]
ligand_coords = rdkit_util.load_molecule(
ligand_file, add_hydrogens=True, calc_charges=True)[0]
num_ligand_atoms = len(ligand_coords)
num_protein_atoms = len(protein_coords)
pocket_inds = []
pocket_atoms = set([])
for lig_atom_ind in range(num_ligand_atoms):
lig_atom = ligand_coords[lig_atom_ind]
for protein_atom_ind in range(num_protein_atoms):
protein_atom = protein_coords[protein_atom_ind]
if np.linalg.norm(lig_atom - protein_atom) < cutoff:
if protein_atom_ind not in pocket_atoms:
pocket_atoms = pocket_atoms.union(set([protein_atom_ind]))
# Should be an array of size (n_pocket_atoms, 3)
pocket_atoms = list(pocket_atoms)
n_pocket_atoms = len(pocket_atoms)
pocket_coords = np.zeros((n_pocket_atoms, 3))
for ind, pocket_ind in enumerate(pocket_atoms):
pocket_coords[ind] = protein_coords[pocket_ind]
x_min = int(np.floor(np.amin(pocket_coords[:, 0])))
x_max = int(np.ceil(np.amax(pocket_coords[:, 0])))
y_min = int(np.floor(np.amin(pocket_coords[:, 1])))
y_max = int(np.ceil(np.amax(pocket_coords[:, 1])))
z_min = int(np.floor(np.amin(pocket_coords[:, 2])))
z_max = int(np.ceil(np.amax(pocket_coords[:, 2])))
return (((x_min, x_max), (y_min, y_max), (z_min, z_max)), pocket_atoms,
pocket_coords)
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
try:
frag1_coords, frag1_mol = rdkit_util.load_molecule(
mol_pdb_file, is_protein=False, sanitize=True, add_hydrogens=False)
frag2_coords, frag2_mol = rdkit_util.load_molecule(
protein_pdb_file, is_protein=True, sanitize=True, add_hydrogens=False)
except MoleculeLoadException:
# Currently handles loading failures by returning None
# TODO: Is there a better handling procedure?
logging.warning("Some molecules cannot be loaded by Rdkit. Skipping")
return None
system_mol = rdkit_util.merge_molecules([frag1_mol, frag2_mol])
system_coords = rdkit_util.get_xyz_from_mol(system_mol)
frag1_coords, frag1_mol = self._strip_hydrogens(frag1_coords, frag1_mol)
frag2_coords, frag2_mol = self._strip_hydrogens(frag2_coords, frag2_mol)
system_coords, system_mol = self._strip_hydrogens(system_coords, system_mol)
try:
frag1_coords, frag1_neighbor_list, frag1_z = self.featurize_mol(
frag1_coords, frag1_mol, self.frag1_num_atoms)
frag2_coords, frag2_neighbor_list, frag2_z = self.featurize_mol(
frag2_coords, frag2_mol, self.frag2_num_atoms)
system_coords, system_neighbor_list, system_z = self.featurize_mol(
system_coords, system_mol, self.complex_num_atoms)
except ValueError as e:
logging.warning(
"max_atoms was set too low. Some complexes too large and skipped")
return None
return frag1_coords, frag1_neighbor_list, frag1_z, frag2_coords, frag2_neighbor_list, frag2_z, \
system_coords, system_neighbor_list, system_z
def extract_active_site(protein_file, ligand_file, cutoff=4):
"""Extracts a box for the active site.
Params
------
protein_file: str
Location of protein PDB
ligand_file: str
Location of ligand input file
cutoff: int, optional
The distance in angstroms from the protein pocket to
consider for featurization.
Returns
-------
A tuple of `(CoordinateBox, np.ndarray)` where the second entry is
of shape `(N, 3)` with `N` the number of atoms in the active site.
"""
protein = rdkit_util.load_molecule(protein_file, add_hydrogens=False)
ligand = rdkit_util.load_molecule(
ligand_file, add_hydrogens=True, calc_charges=True)
protein_contacts, ligand_contacts = get_contact_atom_indices(
[protein, ligand], cutoff=cutoff)
protein_coords = protein[0]
pocket_coords = protein_coords[protein_contacts]
x_min = int(np.floor(np.amin(pocket_coords[:, 0])))
x_max = int(np.ceil(np.amax(pocket_coords[:, 0])))
y_min = int(np.floor(np.amin(pocket_coords[:, 1])))
y_max = int(np.ceil(np.amax(pocket_coords[:, 1])))
z_min = int(np.floor(np.amin(pocket_coords[:, 2])))
z_max = int(np.ceil(np.amax(pocket_coords[:, 2])))
box = box_utils.CoordinateBox((x_min, x_max), (y_min, y_max), (z_min, z_max))
return (box, pocket_coords)
def test_pdbqt_to_pdb(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir,
"../../dock/tests/1jld_protein.pdb")
xyz, mol = rdkit_util.load_molecule(protein_file,
calc_charges=False,
add_hydrogens=False)
out_pdb = "/tmp/mol.pdb"
out_pdbqt = "/tmp/mol.pdbqt"
rdkit_util.write_molecule(mol, out_pdb)
rdkit_util.write_molecule(mol, out_pdbqt, is_protein=True)
pdb_block = rdkit_util.pdbqt_to_pdb(out_pdbqt)
from rdkit import Chem
pdb_mol = Chem.MolFromPDBBlock(pdb_block,
sanitize=False,
removeHs=False)
xyz, pdbqt_mol = rdkit_util.load_molecule(out_pdbqt,
add_hydrogens=False,
calc_charges=False)
assert_equal(pdb_mol.GetNumAtoms(), pdbqt_mol.GetNumAtoms())
for atom_idx in range(pdb_mol.GetNumAtoms()):
atom1 = pdb_mol.GetAtoms()[atom_idx]
atom2 = pdbqt_mol.GetAtoms()[atom_idx]
assert_equal(atom1.GetAtomicNum(), atom2.GetAtomicNum())
os.remove(out_pdb)
os.remove(out_pdbqt)
def test_pdbqt_to_pdb(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir,
"../../dock/tests/1jld_protein.pdb")
xyz, mol = rdkit_util.load_molecule(
protein_file, calc_charges=False, add_hydrogens=False)
out_pdb = "/tmp/mol.pdb"
out_pdbqt = "/tmp/mol.pdbqt"
rdkit_util.write_molecule(mol, out_pdb)
rdkit_util.write_molecule(mol, out_pdbqt, is_protein=True)
pdb_block = rdkit_util.pdbqt_to_pdb(out_pdbqt)
pdb_mol = Chem.MolFromPDBBlock(pdb_block, sanitize=False, removeHs=False)
xyz, pdbqt_mol = rdkit_util.load_molecule(
out_pdbqt, add_hydrogens=False, calc_charges=False)
assert_equal(pdb_mol.GetNumAtoms(), pdbqt_mol.GetNumAtoms())
for atom_idx in range(pdb_mol.GetNumAtoms()):
atom1 = pdb_mol.GetAtoms()[atom_idx]
atom2 = pdbqt_mol.GetAtoms()[atom_idx]
assert_equal(atom1.GetAtomicNum(), atom2.GetAtomicNum())
os.remove(out_pdb)
os.remove(out_pdbqt)
def test_load_molecule(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir, "../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(
ligand_file, calc_charges=False, add_hydrogens=False)
assert_true(xyz is not None)
assert_true(mol is not None)
def test_merge_molecules(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
num_mol_atoms = mol.GetNumAtoms()
# self.ligand_file is for 3ws9_ligand.sdf
oth_xyz, oth_mol = rdkit_util.load_molecule(self.ligand_file,
calc_charges=False,
add_hydrogens=False)
num_oth_mol_atoms = oth_mol.GetNumAtoms()
merged = rdkit_util.merge_molecules([mol, oth_mol])
merged_num_atoms = merged.GetNumAtoms()
assert merged_num_atoms == num_mol_atoms + num_oth_mol_atoms
def find_pockets(self, protein_file, ligand_file):
"""Find list of suitable binding pockets on protein."""
protein_coords = rdkit_util.load_molecule(
protein_file, add_hydrogens=False, calc_charges=False)[0]
ligand_coords = rdkit_util.load_molecule(
ligand_file, add_hydrogens=False, calc_charges=False)[0]
boxes = get_all_boxes(protein_coords, self.pad)
mapping = boxes_to_atoms(protein_coords, boxes)
pockets, pocket_atoms_map = merge_overlapping_boxes(mapping, boxes)
pocket_coords = []
for pocket in pockets:
atoms = pocket_atoms_map[pocket]
coords = np.zeros((len(atoms), 3))
for ind, atom in enumerate(atoms):
coords[ind] = protein_coords[atom]
pocket_coords.append(coords)
return pockets, pocket_atoms_map, pocket_coords
def find_pockets(self, protein_file, ligand_file):
"""Find list of suitable binding pockets on protein."""
protein_coords = rdkit_util.load_molecule(
protein_file, add_hydrogens=False, calc_charges=False)[0]
ligand_coords = rdkit_util.load_molecule(
ligand_file, add_hydrogens=False, calc_charges=False)[0]
boxes = get_all_boxes(protein_coords, self.pad)
mapping = boxes_to_atoms(protein_coords, boxes)
pockets, pocket_atoms_map = merge_overlapping_boxes(mapping, boxes)
pocket_coords = []
for pocket in pockets:
atoms = pocket_atoms_map[pocket]
coords = np.zeros((len(atoms), 3))
for ind, atom in enumerate(atoms):
coords[ind] = protein_coords[atom]
pocket_coords.append(coords)
return pockets, pocket_atoms_map, pocket_coords
def test_load_molecule2(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
assert xyz is not None
assert mol is not None
def test_write_molecule(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir, "../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(
ligand_file, calc_charges=False, add_hydrogens=False)
outfile = "/tmp/mol.sdf"
rdkit_util.write_molecule(mol, outfile)
xyz, mol2 = rdkit_util.load_molecule(
outfile, calc_charges=False, add_hydrogens=False)
assert_equal(mol.GetNumAtoms(), mol2.GetNumAtoms())
for atom_idx in range(mol.GetNumAtoms()):
atom1 = mol.GetAtoms()[atom_idx]
atom2 = mol.GetAtoms()[atom_idx]
assert_equal(atom1.GetAtomicNum(), atom2.GetAtomicNum())
os.remove(outfile)
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
"""
Compute neighbor list for complex.
Parameters
----------
mol_pdb: list
Should be a list of lines of the PDB file.
complex_pdb: list
Should be a list of lines of the PDB file.
"""
mol_coords, ob_mol = rdkit_util.load_molecule(mol_pdb_file)
protein_coords, protein_mol = rdkit_util.load_molecule(protein_pdb_file)
system_coords = rdkit_util.merge_molecules_xyz(mol_coords, protein_coords)
system_neighbor_list = compute_neighbor_list(
system_coords, self.neighbor_cutoff, self.max_num_neighbors, None)
return (system_coords, system_neighbor_list)
def find_all_pockets(self, protein_file):
"""Find list of binding pockets on protein.
Parameters
----------
protein_file: str
Protein to load in.
"""
coords, _ = rdkit_util.load_molecule(protein_file)
return box_utils.get_face_boxes(coords, self.pad)
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
"""
Compute neighbor list for complex.
Parameters
----------
mol_pdb_file: Str
Filename for ligand pdb file.
protein_pdb_file: Str
Filename for protein pdb file.
"""
mol_coords, ob_mol = rdkit_util.load_molecule(mol_pdb_file)
protein_coords, protein_mol = rdkit_util.load_molecule(protein_pdb_file)
system_coords = rdkit_util.merge_molecules_xyz([mol_coords, protein_coords])
system_neighbor_list = compute_neighbor_list(
system_coords, self.neighbor_cutoff, self.max_num_neighbors, None)
return (system_coords, system_neighbor_list)
def test_get_xyz_from_mol(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir, "../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(
ligand_file, calc_charges=False, add_hydrogens=False)
xyz2 = rdkit_util.get_xyz_from_mol(mol)
equal_array = np.all(xyz == xyz2)
assert_true(equal_array)
def test_convert_protein_to_pdbqt(self):
"""Test a protein in a PDB can be converted to PDBQT."""
from rdkit import Chem
xyz, mol = rdkit_util.load_molecule(self.protein_file,
calc_charges=False,
add_hydrogens=False)
with tempfile.TemporaryDirectory() as tmp:
outfile = os.path.join(tmp, "mol.pdbqt")
writer = Chem.PDBWriter(outfile)
writer.write(mol)
writer.close()
pdbqt_utils.convert_protein_to_pdbqt(mol, outfile)
pdbqt_xyz, pdbqt_mol = rdkit_util.load_molecule(
outfile, add_hydrogens=False, calc_charges=False)
assert pdbqt_mol.GetNumAtoms() == pdbqt_mol.GetNumAtoms()
for atom_idx in range(pdbqt_mol.GetNumAtoms()):
atom1 = pdbqt_mol.GetAtoms()[atom_idx]
atom2 = pdbqt_mol.GetAtoms()[atom_idx]
assert atom1.GetAtomicNum() == atom2.GetAtomicNum()
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
"""
Compute neighbor list for complex.
Parameters
----------
mol_pdb: list
Should be a list of lines of the PDB file.
complex_pdb: list
Should be a list of lines of the PDB file.
"""
mol_coords, ob_mol = rdkit_util.load_molecule(mol_pdb_file)
protein_coords, protein_mol = rdkit_util.load_molecule(protein_pdb_file)
system_coords = rdkit_util.merge_molecules_xyz(mol_coords, protein_coords)
system_neighbor_list = compute_neighbor_list(
system_coords, self.neighbor_cutoff, self.max_num_neighbors, None)
return (system_coords, system_neighbor_list)
def _featurize_complex(self, frag1_pdb_file, frag2_pdb_file):
"""Featurize fragments and complex.
Parameters
----------
frag1_pdb_file: string
Location of frag1_pdb_file.
frag2_pdb_file: string
Location of frag2_pdb_file.
Returns
-------
retval: tuple
Tuple containing coordinates, neighbor list, and atomic number for
fragment 1, fragment 2, and complex
"""
try:
frag1_mol = rdkit_util.load_molecule(frag1_pdb_file,
add_hydrogens=False,
calc_charges=False)[1]
frag2_mol = rdkit_util.load_molecule(frag2_pdb_file,
add_hydrogens=False,
calc_charges=False)[1]
except:
frag1_mol = None
frag2_mol = None
if frag1_mol and frag2_mol:
frag1_coords, frag1_neighbor_list, frag1_z = self.frag1_featurizer._featurize(
frag1_mol)
frag2_coords, frag2_neighbor_list, frag2_z = self.frag2_featurizer._featurize(
frag2_mol)
complex_mol = Chem.rdmolops.CombineMols(frag1_mol, frag2_mol)
complex_coords, complex_neighbor_list, complex_z = self.complex_featurizer._featurize(
complex_mol)
return (frag1_coords, frag1_neighbor_list, frag1_z, frag2_coords,
frag2_neighbor_list, frag2_z, complex_coords,
complex_neighbor_list, complex_z)
else:
print("failed to featurize")
return (None, None, None, None, None, None, None, None, None)
def test_merge_molecules_xyz(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir, "../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(
ligand_file, calc_charges=False, add_hydrogens=False)
merged = rdkit_util.merge_molecules_xyz(xyz, xyz)
for i in range(len(xyz)):
first_atom_equal = np.all(xyz[i] == merged[i])
second_atom_equal = np.all(xyz[i] == merged[i + len(xyz)])
assert_true(first_atom_equal)
assert_true(second_atom_equal)
def test_load_molecule(self):
# adding hydrogens and charges is tested in dc.utils
from rdkit.Chem.AllChem import Mol
for add_hydrogens in (True, False):
for calc_charges in (True, False):
mol_xyz, mol_rdk = rdkit_util.load_molecule(
self.ligand_file, add_hydrogens, calc_charges)
num_atoms = mol_rdk.GetNumAtoms()
self.assertIsInstance(mol_xyz, np.ndarray)
self.assertIsInstance(mol_rdk, Mol)
self.assertEqual(mol_xyz.shape, (num_atoms, 3))
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
frag1_coords, frag1_mol = rdkit_util.load_molecule(mol_pdb_file)
frag2_coords, frag2_mol = rdkit_util.load_molecule(protein_pdb_file)
system_mol = rdkit_util.merge_molecules(frag1_mol, frag2_mol)
system_coords = rdkit_util.get_xyz_from_mol(system_mol)
frag1_coords, frag1_mol = self._strip_hydrogens(frag1_coords, frag1_mol)
frag2_coords, frag2_mol = self._strip_hydrogens(frag2_coords, frag2_mol)
system_coords, system_mol = self._strip_hydrogens(system_coords, system_mol)
frag1_coords, frag1_neighbor_list, frag1_z = self.featurize_mol(
frag1_coords, frag1_mol, self.frag1_num_atoms)
frag2_coords, frag2_neighbor_list, frag2_z = self.featurize_mol(
frag2_coords, frag2_mol, self.frag2_num_atoms)
system_coords, system_neighbor_list, system_z = self.featurize_mol(
system_coords, system_mol, self.complex_num_atoms)
return frag1_coords, frag1_neighbor_list, frag1_z, frag2_coords, frag2_neighbor_list, frag2_z, \
system_coords, system_neighbor_list, system_z
def test_get_face_boxes_for_protein(self):
"""Tests that binding pockets are detected."""
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir, "1jld_protein.pdb")
ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = box_utils.get_face_boxes(coords)
assert isinstance(boxes, list)
# Pocket is of form ((x_min, x_max), (y_min, y_max), (z_min, z_max))
for pocket in boxes:
assert isinstance(pocket, box_utils.CoordinateBox)
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
frag1_coords, frag1_mol = rdkit_util.load_molecule(mol_pdb_file)
frag2_coords, frag2_mol = rdkit_util.load_molecule(protein_pdb_file)
system_mol = rdkit_util.merge_molecules(frag1_mol, frag2_mol)
system_coords = rdkit_util.get_xyz_from_mol(system_mol)
frag1_coords, frag1_mol = self._strip_hydrogens(frag1_coords, frag1_mol)
frag2_coords, frag2_mol = self._strip_hydrogens(frag2_coords, frag2_mol)
system_coords, system_mol = self._strip_hydrogens(system_coords, system_mol)
frag1_coords, frag1_neighbor_list, frag1_z = self.featurize_mol(
frag1_coords, frag1_mol, self.frag1_num_atoms)
frag2_coords, frag2_neighbor_list, frag2_z = self.featurize_mol(
frag2_coords, frag2_mol, self.frag2_num_atoms)
system_coords, system_neighbor_list, system_z = self.featurize_mol(
system_coords, system_mol, self.complex_num_atoms)
return frag1_coords, frag1_neighbor_list, frag1_z, frag2_coords, frag2_neighbor_list, frag2_z, \
system_coords, system_neighbor_list, system_z
def test_write_molecule(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
outfile = "/tmp/mol.sdf"
rdkit_util.write_molecule(mol, outfile)
xyz, mol2 = rdkit_util.load_molecule(outfile,
calc_charges=False,
add_hydrogens=False)
assert_equal(mol.GetNumAtoms(), mol2.GetNumAtoms())
for atom_idx in range(mol.GetNumAtoms()):
atom1 = mol.GetAtoms()[atom_idx]
atom2 = mol.GetAtoms()[atom_idx]
assert_equal(atom1.GetAtomicNum(), atom2.GetAtomicNum())
os.remove(outfile)
def test_get_xyz_from_mol(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
xyz2 = rdkit_util.get_xyz_from_mol(mol)
equal_array = np.all(xyz == xyz2)
assert equal_array
def test_write_molecule(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
with tempfile.TemporaryDirectory() as tmp:
outfile = os.path.join(tmp, "mol.sdf")
rdkit_util.write_molecule(mol, outfile)
xyz, mol2 = rdkit_util.load_molecule(outfile,
calc_charges=False,
add_hydrogens=False)
assert mol.GetNumAtoms() == mol2.GetNumAtoms()
for atom_idx in range(mol.GetNumAtoms()):
atom1 = mol.GetAtoms()[atom_idx]
atom2 = mol.GetAtoms()[atom_idx]
assert atom1.GetAtomicNum() == atom2.GetAtomicNum()
def test_merge_molecules_xyz(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
merged = rdkit_util.merge_molecules_xyz([xyz, xyz])
for i in range(len(xyz)):
first_atom_equal = np.all(xyz[i] == merged[i])
second_atom_equal = np.all(xyz[i] == merged[i + len(xyz)])
assert first_atom_equal
assert second_atom_equal
def test_compute_charges(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir, "../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(
ligand_file, calc_charges=False, add_hydrogens=True)
rdkit_util.compute_charges(mol)
has_a_charge = False
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
value = atom.GetProp(str("_GasteigerCharge"))
if value != 0:
has_a_charge = True
assert_true(has_a_charge)
def test_boxes_to_atoms(self, postfix_directory):
"""Test that mapping of protein atoms to boxes is meaningful."""
protein_file = os.path.join(postfix_directory, "PfATP4.pdb")
ligand_file = os.path.join(postfix_directory, "SJ733.pdb")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = dc.dock.binding_pocket.get_all_boxes(coords)
mapping = dc.dock.binding_pocket.boxes_to_atoms(coords, boxes)
assert isinstance(mapping, dict)
for box, box_atoms in mapping.items():
(x_min, x_max), (y_min, y_max), (z_min, z_max) = box
for atom_ind in box_atoms:
atom = coords[atom_ind]
assert x_min <= atom[0] and atom[0] <= x_max
assert y_min <= atom[1] and atom[1] <= y_max
assert z_min <= atom[2] and atom[2] <= z_max
def test_compute_charges(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=True)
rdkit_util.compute_charges(mol)
has_a_charge = False
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
value = atom.GetProp(str("_GasteigerCharge"))
if value != 0:
has_a_charge = True
assert has_a_charge
def test_boxes_to_atoms(self):
"""Test that mapping of protein atoms to boxes is meaningful."""
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir, "1jld_protein.pdb")
ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = dc.dock.binding_pocket.get_all_boxes(coords)
mapping = dc.dock.binding_pocket.boxes_to_atoms(coords, boxes)
assert isinstance(mapping, dict)
for box, box_atoms in mapping.items():
(x_min, x_max), (y_min, y_max), (z_min, z_max) = box
for atom_ind in box_atoms:
atom = coords[atom_ind]
assert x_min <= atom[0] and atom[0] <= x_max
assert y_min <= atom[1] and atom[1] <= y_max
assert z_min <= atom[2] and atom[2] <= z_max
def test_boxes_to_atoms(self):
"""Test that mapping of protein atoms to boxes is meaningful."""
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir, "1jld_protein.pdb")
ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = dc.dock.binding_pocket.get_all_boxes(coords)
mapping = dc.dock.binding_pocket.boxes_to_atoms(coords, boxes)
assert isinstance(mapping, dict)
for box, box_atoms in mapping.items():
(x_min, x_max), (y_min, y_max), (z_min, z_max) = box
for atom_ind in box_atoms:
atom = coords[atom_ind]
assert x_min <= atom[0] and atom[0] <= x_max
assert y_min <= atom[1] and atom[1] <= y_max
assert z_min <= atom[2] and atom[2] <= z_max
def hydrogenate_and_compute_partial_charges(input_file,
input_format,
hyd_output=None,
pdbqt_output=None,
protein=True,
verbose=True):
"""Outputs a hydrogenated pdb and a pdbqt with partial charges.
Takes an input file in specified format. Generates two outputs:
-) A pdb file that contains a hydrogenated (at pH 7.4) version of
original compound.
-) A pdbqt file that has computed Gasteiger partial charges. This pdbqt
file is build from the hydrogenated pdb.
TODO(rbharath): Can do a bit of refactoring between this function and
pdbqt_to_pdb.
Parameters
----------
input_file: String
Path to input file.
input_format: String
Name of input format.
"""
mol = rdkit_util.load_molecule(input_file,
add_hydrogens=True,
calc_charges=True)[1]
if verbose:
print("Create pdb with hydrogens added")
rdkit_util.write_molecule(mol, str(hyd_output), is_protein=protein)
if verbose:
print("Create a pdbqt file from the hydrogenated pdb above.")
rdkit_util.write_molecule(mol, str(pdbqt_output), is_protein=protein)
if protein:
print("Removing ROOT/ENDROOT/TORSDOF")
with open(pdbqt_output) as f:
pdbqt_lines = f.readlines()
filtered_lines = []
for line in pdbqt_lines:
filtered_lines.append(line)
with open(pdbqt_output, "w") as f:
f.writelines(filtered_lines)
def test_add_hydrogens_to_mol(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir, "../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(
ligand_file, calc_charges=False, add_hydrogens=False)
original_hydrogen_count = 0
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
if atom.GetAtomicNum() == 1:
original_hydrogen_count += 1
mol = rdkit_util.add_hydrogens_to_mol(mol)
after_hydrogen_count = 0
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
if atom.GetAtomicNum() == 1:
after_hydrogen_count += 1
assert_true(after_hydrogen_count >= original_hydrogen_count)
def hydrogenate_and_compute_partial_charges(input_file,
input_format,
hyd_output=None,
pdbqt_output=None,
protein=True,
verbose=True):
"""Outputs a hydrogenated pdb and a pdbqt with partial charges.
Takes an input file in specified format. Generates two outputs:
-) A pdb file that contains a hydrogenated (at pH 7.4) version of
original compound.
-) A pdbqt file that has computed Gasteiger partial charges. This pdbqt
file is build from the hydrogenated pdb.
TODO(rbharath): Can do a bit of refactoring between this function and
pdbqt_to_pdb.
Parameters
----------
input_file: String
Path to input file.
input_format: String
Name of input format.
"""
mol = rdkit_util.load_molecule(
input_file, add_hydrogens=True, calc_charges=True)[1]
if verbose:
logging.info("Create pdb with hydrogens added")
rdkit_util.write_molecule(mol, str(hyd_output), is_protein=protein)
if verbose:
logging.info("Create a pdbqt file from the hydrogenated pdb above.")
rdkit_util.write_molecule(mol, str(pdbqt_output), is_protein=protein)
if protein:
logging.info("Removing ROOT/ENDROOT/TORSDOF")
with open(pdbqt_output) as f:
pdbqt_lines = f.readlines()
filtered_lines = []
for line in pdbqt_lines:
filtered_lines.append(line)
with open(pdbqt_output, "w") as f:
f.writelines(filtered_lines)
def test_get_all_boxes(self):
"""Tests that binding pockets are detected."""
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir, "1jld_protein.pdb")
ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = dc.dock.binding_pocket.get_all_boxes(coords)
assert isinstance(boxes, list)
# Pocket is of form ((x_min, x_max), (y_min, y_max), (z_min, z_max))
for pocket in boxes:
assert len(pocket) == 3
assert len(pocket[0]) == 2
assert len(pocket[1]) == 2
assert len(pocket[2]) == 2
(x_min, x_max), (y_min, y_max), (z_min, z_max) = pocket
assert x_min < x_max
assert y_min < y_max
assert z_min < z_max
def test_get_all_boxes(self):
"""Tests that binding pockets are detected."""
current_dir = os.path.dirname(os.path.realpath(__file__))
protein_file = os.path.join(current_dir, "1jld_protein.pdb")
ligand_file = os.path.join(current_dir, "1jld_ligand.sdf")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = dc.dock.binding_pocket.get_all_boxes(coords)
assert isinstance(boxes, list)
# Pocket is of form ((x_min, x_max), (y_min, y_max), (z_min, z_max))
for pocket in boxes:
assert len(pocket) == 3
assert len(pocket[0]) == 2
assert len(pocket[1]) == 2
assert len(pocket[2]) == 2
(x_min, x_max), (y_min, y_max), (z_min, z_max) = pocket
assert x_min < x_max
assert y_min < y_max
assert z_min < z_max
def test_get_all_boxes(self, postfix_directory):
"""Tests that binding pockets are detected."""
print("Test_All_Boxes")
protein_file = os.path.join(postfix_directory, "PfATP4.pdb")
ligand_file = os.path.join(postfix_directory, "SJ733.pdb")
coords = rdkit_util.load_molecule(protein_file)[0]
boxes = dc.dock.binding_pocket.get_all_boxes(coords)
assert isinstance(boxes, list)
# Pocket is of form ((x_min, x_max), (y_min, y_max), (z_min, z_max))
for pocket in boxes:
assert len(pocket) == 3
assert len(pocket[0]) == 2
assert len(pocket[1]) == 2
assert len(pocket[2]) == 2
(x_min, x_max), (y_min, y_max), (z_min, z_max) = pocket
assert x_min < x_max
assert y_min < y_max
assert z_min < z_max
def test_add_hydrogens_to_mol(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
original_hydrogen_count = 0
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
if atom.GetAtomicNum() == 1:
original_hydrogen_count += 1
mol = rdkit_util.add_hydrogens_to_mol(mol)
after_hydrogen_count = 0
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
if atom.GetAtomicNum() == 1:
after_hydrogen_count += 1
assert_true(after_hydrogen_count >= original_hydrogen_count)
def find_pockets(self, macromolecule_file):
"""Find list of suitable binding pockets on protein.
This function computes putative binding pockets on this protein.
This class uses the `ConvexHull` to compute binding pockets. Each
face of the hull is converted into a coordinate box used for
binding.
Params
------
macromolecule_file: str
Location of the macromolecule file to load
Returns
-------
List of pockets. Each pocket is a `CoordinateBox`
"""
coords = rdkit_util.load_molecule(
macromolecule_file, add_hydrogens=False, calc_charges=False)[0]
boxes = box_utils.get_face_boxes(coords, self.pad)
boxes = box_utils.merge_overlapping_boxes(boxes)
return boxes
def featurize(self, protein_file, pockets):
"""
Calculate atomic coodinates.
Params
------
protein_file: str
Location of PDB file. Will be loaded by MDTraj
pockets: list[CoordinateBox]
List of `dc.utils.CoordinateBox` objects.
Returns
-------
A numpy array of shale `(len(pockets), n_residues)`
"""
import mdtraj
protein_coords = rdkit_util.load_molecule(
protein_file, add_hydrogens=False, calc_charges=False)[0]
mapping = boxes_to_atoms(protein_coords, pockets)
protein = mdtraj.load(protein_file)
n_pockets = len(pockets)
n_residues = len(BindingPocketFeaturizer.residues)
res_map = dict(zip(BindingPocketFeaturizer.residues, range(n_residues)))
all_features = np.zeros((n_pockets, n_residues))
for pocket_num, pocket in enumerate(pockets):
pocket_atoms = mapping[pocket]
for ind, atom in enumerate(pocket_atoms):
atom_name = str(protein.top.atom(atom))
# atom_name is of format RESX-ATOMTYPE
# where X is a 1 to 4 digit number
residue = atom_name[:3]
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
def test_apply_pdbfixer(self):
current_dir = os.path.dirname(os.path.realpath(__file__))
ligand_file = os.path.join(current_dir,
"../../dock/tests/1jld_ligand.sdf")
xyz, mol = rdkit_util.load_molecule(ligand_file,
calc_charges=False,
add_hydrogens=False)
original_hydrogen_count = 0
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
if atom.GetAtomicNum() == 1:
original_hydrogen_count += 1
assert mol is not None
mol = rdkit_util.apply_pdbfixer(mol,
hydrogenate=True,
is_protein=False)
assert mol is not None
after_hydrogen_count = 0
for atom_idx in range(mol.GetNumAtoms()):
atom = mol.GetAtoms()[atom_idx]
if atom.GetAtomicNum() == 1:
after_hydrogen_count += 1
assert after_hydrogen_count >= original_hydrogen_count
def _featurize_complex(self, mol_pdb_file, protein_pdb_file):
"""Computes grid featurization of protein/ligand complex.
Takes as input filenames pdb of the protein, pdb of the ligand.
This function then computes the centroid of the ligand; decrements this
centroid from the atomic coordinates of protein and ligand atoms, and then
merges the translated protein and ligand. This combined system/complex is then
saved.
This function then computes a featurization with scheme specified by the user.
Parameters
----------
mol_pdb_file: Str
Filename for ligand pdb file.
protein_pdb_file: Str
Filename for protein pdb file.
"""
try:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_xyz, protein_rdk = load_molecule(
protein_pdb_file, calc_charges=True, sanitize=self.sanitize)
############################################################## TIMING
time2 = time.time()
log("TIMING: Loading protein coordinates took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
ligand_xyz, ligand_rdk = load_molecule(
mol_pdb_file, calc_charges=True, sanitize=self.sanitize)
############################################################## TIMING
time2 = time.time()
log("TIMING: Loading ligand coordinates took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
except MoleculeLoadException:
logging.warning("Some molecules cannot be loaded by Rdkit. Skipping")
return None
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
centroid = compute_centroid(ligand_xyz)
ligand_xyz = subtract_centroid(ligand_xyz, centroid)
protein_xyz = subtract_centroid(protein_xyz, centroid)
############################################################## TIMING
time2 = time.time()
log("TIMING: Centroid processing took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
pairwise_distances = compute_pairwise_distances(protein_xyz, ligand_xyz)
transformed_systems = {}
transformed_systems[(0, 0)] = [protein_xyz, ligand_xyz]
for i in range(self.nb_rotations):
rotated_system = rotate_molecules([protein_xyz, ligand_xyz])
transformed_systems[(i + 1, 0)] = rotated_system
features_dict = {}
for system_id, (protein_xyz, ligand_xyz) in transformed_systems.items():
feature_arrays = []
for is_flat, function_name in self.feature_types:
result = self._compute_feature(
function_name,
protein_xyz,
protein_rdk,
ligand_xyz,
ligand_rdk,
pairwise_distances,
)
feature_arrays += result
if self.flatten:
features_dict[system_id] = np.concatenate(
[feature_array.flatten() for feature_array in feature_arrays])
else:
features_dict[system_id] = np.concatenate(feature_arrays, axis=-1)
# TODO(rbharath): Is this squeeze OK?
features = np.squeeze(np.array(list(features_dict.values())))
return features
def find_all_pockets(self, protein_file): """Find list of binding pockets on protein.""" # protein_coords is (N, 3) tensor coords = rdkit_util.load_molecule(protein_file)[0] return get_all_boxes(coords, self.pad)
def _transform(self, protein_pdb, ligand_file):
"""Computes featurization of protein/ligand complex.
Takes as input files (strings) for pdb of the protein, pdb of the ligand,
and a directory to save intermediate files.
This function then computes the centroid of the ligand; decrements this
centroid from the atomic coordinates of protein and ligand atoms, and then
merges the translated protein and ligand. This combined system/complex is then
saved.
This function then computes a featurization with scheme specified by the user.
"""
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_name = str(protein_pdb).split("/")[len(str(protein_pdb).split("/"))
- 2]
if not self.ligand_only:
protein_xyz, protein_ob = rdkit_util.load_molecule(
protein_pdb, calc_charges=True)
############################################################## TIMING
time2 = time.time()
log("TIMING: Loading protein coordinates took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
ligand_xyz, ligand_ob = rdkit_util.load_molecule(
ligand_file, calc_charges=True)
############################################################## TIMING
time2 = time.time()
log("TIMING: Loading ligand coordinates took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "ecfp" in self.feature_types:
ecfp_array = compute_ecfp_features(ligand_ob, self.ecfp_degree,
self.ecfp_power)
return ({(0, 0): ecfp_array})
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
centroid = compute_centroid(ligand_xyz)
ligand_xyz = subtract_centroid(ligand_xyz, centroid)
if not self.ligand_only:
protein_xyz = subtract_centroid(protein_xyz, centroid)
############################################################## TIMING
time2 = time.time()
log("TIMING: Centroid processing took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "splif" in self.feature_types:
splif_array = self._featurize_splif(protein_xyz, protein_ob, ligand_xyz,
ligand_ob)
return ({(0, 0): splif_array})
if "flat_combined" in self.feature_types:
return (self._compute_flat_features(protein_xyz, protein_ob, ligand_xyz,
ligand_ob))
pairwise_distances = compute_pairwise_distances(protein_xyz, ligand_xyz)
if "ecfp" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_ecfp_dict, ligand_ecfp_dict = (featurize_binding_pocket_ecfp(
protein_xyz,
protein_ob,
ligand_xyz,
ligand_ob,
pairwise_distances,
cutoff=4.5,
ecfp_degree=self.ecfp_degree))
############################################################## TIMING
time2 = time.time()
log("TIMING: ecfp voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "splif" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
splif_dicts = featurize_splif(protein_xyz, protein_ob, ligand_xyz,
ligand_ob, self.contact_bins,
pairwise_distances, self.ecfp_degree)
############################################################## TIMING
time2 = time.time()
log("TIMING: splif voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "hbond" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
hbond_list = compute_hydrogen_bonds(
protein_xyz, protein_ob, ligand_xyz, ligand_ob, pairwise_distances,
self.hbond_dist_bins, self.hbond_angle_cutoffs, self.ecfp_degree)
############################################################## TIMING
time2 = time.time()
log("TIMING: hbond voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "sybyl" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_sybyl_dict, ligand_sybyl_dict = featurize_binding_pocket_sybyl(
protein_xyz,
protein_ob,
ligand_xyz,
ligand_ob,
pairwise_distances,
cutoff=7.0)
############################################################## TIMING
time2 = time.time()
log("TIMING: sybyl voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "pi_stack" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_pi_t, protein_pi_parallel, ligand_pi_t, ligand_pi_parallel = (
compute_pi_stack(protein_xyz, protein_ob, ligand_xyz, ligand_ob,
pairwise_distances))
############################################################## TIMING
time2 = time.time()
log("TIMING: pi_stack voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "cation_pi" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_cation_pi, ligand_cation_pi = (compute_binding_pocket_cation_pi(
protein_xyz, protein_ob, ligand_xyz, ligand_ob))
############################################################## TIMING
time2 = time.time()
log("TIMING: cation_pi voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
if "salt_bridge" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
salt_bridge_list = compute_salt_bridges(
protein_xyz, protein_ob, ligand_xyz, ligand_ob, pairwise_distances)
############################################################## TIMING
time2 = time.time()
log("TIMING: salt_bridge voxel computataion took %0.3f s" %
(time2 - time1), self.verbose)
############################################################## TIMING
if "charge" in self.voxel_feature_types:
############################################################## TIMING
time1 = time.time()
############################################################## TIMING
protein_charge_dictionary = compute_charge_dictionary(protein_ob)
ligand_charge_dictionary = compute_charge_dictionary(ligand_ob)
############################################################## TIMING
time2 = time.time()
log("TIMING: charge voxel computataion took %0.3f s" % (time2 - time1),
self.verbose)
############################################################## TIMING
transformed_systems = {}
transformed_systems[(0, 0)] = [protein_xyz, ligand_xyz]
for i in range(0, int(self.nb_rotations)):
rotated_system = rotate_molecules([protein_xyz, ligand_xyz])
transformed_systems[(i + 1, 0)] = rotated_system
for j in range(0, int(self.nb_reflections)):
reflected_system = self._reflect_molecule(rotated_system)
transformed_systems[(i + 1, j + 1)] = reflected_system
if "voxel_combined" in self.feature_types:
features = {}
for system_id, system in transformed_systems.items():
protein_xyz = system[0]
ligand_xyz = system[1]
feature_tensors = []
if "ecfp" in self.voxel_feature_types:
ecfp_tensor = self._voxelize(
convert_atom_to_voxel,
hash_ecfp,
protein_xyz,
feature_dict=protein_ecfp_dict,
channel_power=self.ecfp_power)
ecfp_tensor += self._voxelize(
convert_atom_to_voxel,
hash_ecfp,
ligand_xyz,
feature_dict=ligand_ecfp_dict,
channel_power=self.ecfp_power)
feature_tensors.append(ecfp_tensor)
print("Completed ecfp tensor")
if "splif" in self.voxel_feature_types:
feature_tensors += [
self._voxelize(
convert_atom_pair_to_voxel,
hash_ecfp_pair, (protein_xyz, ligand_xyz),
feature_dict=splif_dict,
channel_power=self.splif_power) for splif_dict in splif_dicts
]
print("Completed splif tensor")
if "hbond" in self.voxel_feature_types:
feature_tensors += [
self._voxelize(
convert_atom_pair_to_voxel,
None, (protein_xyz, ligand_xyz),
feature_list=hbond,
channel_power=0) for hbond in hbond_list
]
print("Completed hbond tensor")
if "sybyl" in self.voxel_feature_types:
sybyl_partial = partial(hash_sybyl, sybyl_types=self.sybyl_types)
sybyl_tensor = self._voxelize(
convert_atom_to_voxel,
hash_sybyl,
protein_xyz,
feature_dict=protein_sybyl_dict,
nb_channel=len(self.sybyl_types))
sybyl_tensor += self._voxelize(
convert_atom_to_voxel,
hash_sybyl,
ligand_xyz,
feature_dict=ligand_sybyl_dict,
nb_channel=len(self.sybyl_types))
feature_tensors.append(sybyl_tensor)
print("Completed sybyl tensor")
if "pi_stack" in self.voxel_feature_types:
pi_parallel_tensor = self._voxelize(
convert_atom_to_voxel,
None,
protein_xyz,
feature_dict=protein_pi_parallel,
nb_channel=1)
pi_parallel_tensor += self._voxelize(
convert_atom_to_voxel,
None,
ligand_xyz,
feature_dict=ligand_pi_parallel,
nb_channel=1)
feature_tensors.append(pi_parallel_tensor)
pi_t_tensor = self._voxelize(
convert_atom_to_voxel,
None,
protein_xyz,
feature_dict=protein_pi_t,
nb_channel=1)
pi_t_tensor += self._voxelize(
convert_atom_to_voxel,
None,
ligand_xyz,
feature_dict=ligand_pi_t,
nb_channel=1)
feature_tensors.append(pi_t_tensor)
print("Completed pi_stack tensor")
if "cation_pi" in self.voxel_feature_types:
cation_pi_tensor = self._voxelize(
convert_atom_to_voxel,
None,
protein_xyz,
feature_dict=protein_cation_pi,
nb_channel=1)
cation_pi_tensor += self._voxelize(
convert_atom_to_voxel,
None,
ligand_xyz,
feature_dict=ligand_cation_pi,
nb_channel=1)
feature_tensors.append(cation_pi_tensor)
print("Completed cation_pi tensor.")
if "salt_bridge" in self.voxel_feature_types:
salt_bridge_tensor = self._voxelize(
convert_atom_pair_to_voxel,
None, (protein_xyz, ligand_xyz),
feature_list=salt_bridge_list,
nb_channel=1)
feature_tensors.append(salt_bridge_tensor)
print("Completed salt_bridge tensor.")
if "charge" in self.voxel_feature_types:
charge_tensor = self._voxelize(
convert_atom_to_voxel,
None,
protein_xyz,
feature_dict=protein_charge_dictionary,
nb_channel=1,
dtype="np.float16")
charge_tensor += self._voxelize(
convert_atom_to_voxel,
None,
ligand_xyz,
feature_dict=ligand_charge_dictionary,
nb_channel=1,
dtype="np.float16")
feature_tensors.append(charge_tensor)
print("Completed salt_bridge tensor.")
if "charge" in self.voxel_feature_types:
feature_tensor = np.concatenate(
feature_tensors, axis=3).astype(np.float16)
else:
feature_tensor = np.concatenate(
feature_tensors, axis=3).astype(np.int8)
if self.flatten:
feature_tensor = np.squeeze(feature_tensor)
features[system_id] = feature_tensor
return (features)
def dock_ligands_to_receptors(docking_dir,
worker_pool=None,
exhaustiveness=None,
chosen_receptor=None,
restrict_box=True):
subdirs = glob.glob(os.path.join(docking_dir, '*/'))
for subdir in subdirs:
subdir = subdir.rstrip('/')
receptor_name = os.path.basename(subdir)
if chosen_receptor is not None and chosen_receptor != receptor_name:
continue
print("receptor name = %s" % receptor_name)
receptor_filename = os.path.join(subdir, "%s.pdbqt" % receptor_name)
if not os.path.exists(receptor_filename):
continue
print("Examining %s" % receptor_filename)
receptor_mol = rdkit_util.load_molecule(
os.path.join(subdir, "%s.pdb" % receptor_name))
protein_centroid = mol_xyz_util.get_molecule_centroid(receptor_mol[0])
protein_range = mol_xyz_util.get_molecule_range(receptor_mol[0])
box_dims = protein_range + 5.0
ligands = sorted(glob.glob(os.path.join(subdir, '*_prepared.pdbqt')))
print("Num ligands = %d" % len(ligands))
dock_ligand_to_receptor_partial = partial(
dock_ligand_to_receptor,
receptor_filename=receptor_filename,
protein_centroid=protein_centroid,
box_dims=box_dims,
subdir=subdir,
exhaustiveness=exhaustiveness)
if restrict_box:
active_ligand = ""
for ligand in ligands:
if "CHEM" in ligand:
active_ligand = ligand
break
print("Docking to %s first to ascertain centroid and box dimensions" %
active_ligand)
out_pdb_qt = dock_ligand_to_receptor_partial(active_ligand)
ligand_pybel = rdkit_util.load_molecule(out_pdb_qt)
ligand_centroid = mol_xyz_util.get_molecule_centroid(ligand_pybel[0])
print("Protein centroid = %s" % (str(protein_centroid)))
print("Ligand centroid = %s" % (str(ligand_centroid)))
box_dims = np.array([20., 20., 20.])
dock_ligand_to_receptor_partial = partial(
dock_ligand_to_receptor,
receptor_filename=receptor_filename,
protein_centroid=ligand_centroid,
box_dims=box_dims,
subdir=subdir,
exhaustiveness=exhaustiveness)
print("Finished docking to %s, docking to remainder of ligands now." %
active_ligand)
if worker_pool is None:
for i, ligand_file in enumerate(ligands):
a = time.time()
dock_ligand_to_receptor_partial(ligand_file)
print("took %f seconds to dock single ligand." % (time.time() - a))
else:
print("parallelizing docking over worker pool")
worker_pool.map(dock_ligand_to_receptor_partial, ligands)
def load_molecule(molecule_file, add_hydrogens=True, calc_charges=False): return rdkit_util.load_molecule(molecule_file, add_hydrogens, calc_charges)
def generate_poses(self,
protein_file,
ligand_file,
centroid=None,
box_dims=None,
dry_run=False,
out_dir=None):
"""Generates the docked complex and outputs files for docked complex."""
if out_dir is None:
out_dir = tempfile.mkdtemp()
# Prepare receptor
receptor_name = os.path.basename(protein_file).split(".")[0]
protein_hyd = os.path.join(out_dir, "%s.pdb" % receptor_name)
protein_pdbqt = os.path.join(out_dir, "%s.pdbqt" % receptor_name)
hydrogenate_and_compute_partial_charges(
protein_file,
"pdb",
hyd_output=protein_hyd,
pdbqt_output=protein_pdbqt,
protein=True)
# Get protein centroid and range
# TODO(rbharath): Need to add some way to identify binding pocket, or this is
# going to be extremely slow!
if centroid is not None and box_dims is not None:
protein_centroid = centroid
else:
if not self.detect_pockets:
receptor_mol = rdkit_util.load_molecule(
protein_hyd, calc_charges=False, add_hydrogens=False)
protein_centroid = mol_xyz_util.get_molecule_centroid(receptor_mol[0])
protein_range = mol_xyz_util.get_molecule_range(receptor_mol[0])
box_dims = protein_range + 5.0
else:
logger.info("About to find putative binding pockets")
pockets, pocket_atoms_maps, pocket_coords = self.pocket_finder.find_pockets(
protein_file, ligand_file)
# TODO(rbharath): Handle multiple pockets instead of arbitrarily selecting
# first pocket.
logger.info("Computing centroid and size of proposed pocket.")
pocket_coord = pocket_coords[0]
protein_centroid = np.mean(pocket_coord, axis=1)
pocket = pockets[0]
(x_min, x_max), (y_min, y_max), (z_min, z_max) = pocket
x_box = (x_max - x_min) / 2.
y_box = (y_max - y_min) / 2.
z_box = (z_max - z_min) / 2.
box_dims = (x_box, y_box, z_box)
# Prepare receptor
ligand_name = os.path.basename(ligand_file).split(".")[0]
ligand_hyd = os.path.join(out_dir, "%s.pdb" % ligand_name)
ligand_pdbqt = os.path.join(out_dir, "%s.pdbqt" % ligand_name)
# TODO(rbharath): Generalize this so can support mol2 files as well.
hydrogenate_and_compute_partial_charges(
ligand_file,
"sdf",
hyd_output=ligand_hyd,
pdbqt_output=ligand_pdbqt,
protein=False)
# Write Vina conf file
conf_file = os.path.join(out_dir, "conf.txt")
write_conf(
protein_pdbqt,
ligand_pdbqt,
protein_centroid,
box_dims,
conf_file,
exhaustiveness=self.exhaustiveness)
# Define locations of log and output files
log_file = os.path.join(out_dir, "%s_log.txt" % ligand_name)
out_pdbqt = os.path.join(out_dir, "%s_docked.pdbqt" % ligand_name)
# TODO(rbharath): Let user specify the number of poses required.
if not dry_run:
logger.info("About to call Vina")
call(
"%s --config %s --log %s --out %s" % (self.vina_cmd, conf_file,
log_file, out_pdbqt),
shell=True)
# TODO(rbharath): Convert the output pdbqt to a pdb file.
# Return docked files
return protein_hyd, out_pdbqt