Loading deepchem/utils/__init__.py +1 −0 Original line number Diff line number Diff line Loading @@ -84,6 +84,7 @@ from deepchem.utils.pdbqt_utils import convert_mol_to_pdbqt from deepchem.utils.vina_utils import write_vina_conf from deepchem.utils.vina_utils import load_docked_ligands from deepchem.utils.vina_utils import prepare_inputs from deepchem.utils.voxel_utils import convert_atom_to_voxel from deepchem.utils.voxel_utils import convert_atom_pair_to_voxel Loading deepchem/utils/test/test_vina_utils.py +21 −0 Original line number Diff line number Diff line Loading @@ -25,3 +25,24 @@ class TestVinaUtils(unittest.TestCase): xyz = rdkit_utils.get_xyz_from_mol(ligand) assert score < 0 # This is a binding free energy assert np.count_nonzero(xyz) > 0 def test_prepare_inputs(self): from rdkit import Chem pdbid = '3cyx' ligand_smiles = 'CC(C)(C)NC(O)C1CC2CCCCC2C[NH+]1CC(O)C(CC1CCCCC1)NC(O)C(CC(N)O)NC(O)C1CCC2CCCCC2N1' protein, ligand = vina_utils.prepare_inputs( pdbid, ligand_smiles, pdb_name=pdbid) assert protein.GetNumAtoms() == 1415 assert ligand.GetNumAtoms() == 124 protein, ligand = vina_utils.prepare_inputs(pdbid + '.pdb', 'ligand_' + pdbid + '.pdb') assert protein.GetNumAtoms() == 1415 assert ligand.GetNumAtoms() == 124 os.remove(pdbid + '.pdb') os.remove('ligand_' + pdbid + '.pdb') os.remove('tmp.pdb') deepchem/utils/vina_utils.py +69 −0 Original line number Diff line number Diff line Loading @@ -114,3 +114,72 @@ def load_docked_ligands( mol = Chem.MolFromPDBBlock(str(pdb_block), sanitize=False, removeHs=False) molecules.append(mol) return molecules, scores def prepare_inputs(protein: str, ligand: str, pdb_name: Optional[str] = None) -> Tuple[RDKitMol, RDKitMol]: """This prepares protein-ligand complexes for docking. Autodock Vina requires PDB files for proteins and ligands with sensible inputs. This function uses PDBFixer and RDKit to ensure that inputs are reasonable and ready for docking. Parameters ---------- protein: str Filename for protein PDB file or a PDBID. ligand: str Either a filename for a ligand PDB file or a SMILES string. pdb_name: Optional[str] If given, write sanitized protein and ligand to files called "pdb_name.pdb" and "ligand_pdb_name.pdb" Returns ------- Tuple[RDKitMol, RDKitMol] Tuple of `protein_molecule, ligand_molecule` with 3D information. Notes ----- This function requires RDKit and OpenMM to be installed. Read more about PDBFixer here: https://github.com/openmm/pdbfixer. """ try: from rdkit import Chem from pdbfixer import PDBFixer from simtk.openmm.app import PDBFile except ModuleNotFoundError: raise ImportError( "This function requires RDKit and OpenMM to be installed.") if protein.endswith('.pdb'): fixer = PDBFixer(protein) else: fixer = PDBFixer(url='https://files.rcsb.org/download/%s.pdb' % (protein)) if ligand.endswith('.pdb'): m = Chem.MolFromPDBFile(ligand) else: m = Chem.MolFromSmiles(ligand, sanitize=True) # Apply common fixes to PDB files fixer.findMissingResidues() fixer.findNonstandardResidues() fixer.replaceNonstandardResidues() fixer.removeHeterogens(False) # remove water fixer.addMissingHydrogens(7.0) PDBFile.writeFile(fixer.topology, fixer.positions, open('tmp.pdb', 'w')) p = Chem.MolFromPDBFile('tmp.pdb', sanitize=True) # Optimize ligand m = Chem.AddHs(m) # need hydrogens for optimization Chem.AllChem.EmbedMolecule(m) Chem.AllChem.MMFFOptimizeMolecule(m) if pdb_name: Chem.rdmolfiles.MolToPDBFile(p, '%s.pdb' % (pdb_name)) Chem.rdmolfiles.MolToPDBFile(m, 'ligand_%s.pdb' % (pdb_name)) return (p, m) Loading
deepchem/utils/__init__.py +1 −0 Original line number Diff line number Diff line Loading @@ -84,6 +84,7 @@ from deepchem.utils.pdbqt_utils import convert_mol_to_pdbqt from deepchem.utils.vina_utils import write_vina_conf from deepchem.utils.vina_utils import load_docked_ligands from deepchem.utils.vina_utils import prepare_inputs from deepchem.utils.voxel_utils import convert_atom_to_voxel from deepchem.utils.voxel_utils import convert_atom_pair_to_voxel Loading
deepchem/utils/test/test_vina_utils.py +21 −0 Original line number Diff line number Diff line Loading @@ -25,3 +25,24 @@ class TestVinaUtils(unittest.TestCase): xyz = rdkit_utils.get_xyz_from_mol(ligand) assert score < 0 # This is a binding free energy assert np.count_nonzero(xyz) > 0 def test_prepare_inputs(self): from rdkit import Chem pdbid = '3cyx' ligand_smiles = 'CC(C)(C)NC(O)C1CC2CCCCC2C[NH+]1CC(O)C(CC1CCCCC1)NC(O)C(CC(N)O)NC(O)C1CCC2CCCCC2N1' protein, ligand = vina_utils.prepare_inputs( pdbid, ligand_smiles, pdb_name=pdbid) assert protein.GetNumAtoms() == 1415 assert ligand.GetNumAtoms() == 124 protein, ligand = vina_utils.prepare_inputs(pdbid + '.pdb', 'ligand_' + pdbid + '.pdb') assert protein.GetNumAtoms() == 1415 assert ligand.GetNumAtoms() == 124 os.remove(pdbid + '.pdb') os.remove('ligand_' + pdbid + '.pdb') os.remove('tmp.pdb')
deepchem/utils/vina_utils.py +69 −0 Original line number Diff line number Diff line Loading @@ -114,3 +114,72 @@ def load_docked_ligands( mol = Chem.MolFromPDBBlock(str(pdb_block), sanitize=False, removeHs=False) molecules.append(mol) return molecules, scores def prepare_inputs(protein: str, ligand: str, pdb_name: Optional[str] = None) -> Tuple[RDKitMol, RDKitMol]: """This prepares protein-ligand complexes for docking. Autodock Vina requires PDB files for proteins and ligands with sensible inputs. This function uses PDBFixer and RDKit to ensure that inputs are reasonable and ready for docking. Parameters ---------- protein: str Filename for protein PDB file or a PDBID. ligand: str Either a filename for a ligand PDB file or a SMILES string. pdb_name: Optional[str] If given, write sanitized protein and ligand to files called "pdb_name.pdb" and "ligand_pdb_name.pdb" Returns ------- Tuple[RDKitMol, RDKitMol] Tuple of `protein_molecule, ligand_molecule` with 3D information. Notes ----- This function requires RDKit and OpenMM to be installed. Read more about PDBFixer here: https://github.com/openmm/pdbfixer. """ try: from rdkit import Chem from pdbfixer import PDBFixer from simtk.openmm.app import PDBFile except ModuleNotFoundError: raise ImportError( "This function requires RDKit and OpenMM to be installed.") if protein.endswith('.pdb'): fixer = PDBFixer(protein) else: fixer = PDBFixer(url='https://files.rcsb.org/download/%s.pdb' % (protein)) if ligand.endswith('.pdb'): m = Chem.MolFromPDBFile(ligand) else: m = Chem.MolFromSmiles(ligand, sanitize=True) # Apply common fixes to PDB files fixer.findMissingResidues() fixer.findNonstandardResidues() fixer.replaceNonstandardResidues() fixer.removeHeterogens(False) # remove water fixer.addMissingHydrogens(7.0) PDBFile.writeFile(fixer.topology, fixer.positions, open('tmp.pdb', 'w')) p = Chem.MolFromPDBFile('tmp.pdb', sanitize=True) # Optimize ligand m = Chem.AddHs(m) # need hydrogens for optimization Chem.AllChem.EmbedMolecule(m) Chem.AllChem.MMFFOptimizeMolecule(m) if pdb_name: Chem.rdmolfiles.MolToPDBFile(p, '%s.pdb' % (pdb_name)) Chem.rdmolfiles.MolToPDBFile(m, 'ligand_%s.pdb' % (pdb_name)) return (p, m)
