def readPDBFromStream(stream: Base.IOStream): from Protein import Protein from MoleculeTools import sanitize_mol r = Biomol.PDBMoleculeReader(stream) mol = Chem.BasicMolecule() r.read(mol) sanitize_mol(mol, makeHydrogenComplete=True) return Protein(mol)
def cdfMol_pdb(pdb, output, name): initial_time = time.time() cdf_mol = Chem.BasicMolecule() pdb_mol = Chem.BasicMolecule() pdb_str = open(pdb, 'r').read().replace('WAT', 'HOH').replace('HIE', 'HIS') pdb_reader = Biomol.PDBMoleculeReader(Base.StringIOStream(pdb_str)) Biomol.setPDBApplyDictAtomBondingToNonStdResiduesParameter( pdb_reader, True) if not pdb_reader.read(pdb_mol): return None Chem.calcImplicitHydrogenCounts(pdb_mol, False) Chem.perceiveHybridizationStates(pdb_mol, False) Chem.setAtomSymbolsFromTypes(pdb_mol, False) Chem.perceiveSSSR(pdb_mol, False) Chem.setRingFlags(pdb_mol, False) Chem.setAromaticityFlags(pdb_mol, False) cdf_mol.assign(pdb_mol) for atom in cdf_mol.atoms: Chem.set3DCoordinatesArray(atom, Math.Vector3DArray()) i = 0 while i < cdf_mol.numAtoms: Chem.get3DCoordinatesArray(cdf_mol.getAtom(i)).addElement( Chem.get3DCoordinates(pdb_mol.getAtom(i))) i += 1 tmp_output = output + name + ".cdf" try: Chem.FileCDFMolecularGraphWriter(tmp_output).write(cdf_mol) except: print('> Cdf_mol writing failure.') raise residues = Biomol.ResidueList(cdf_mol) tmp_output = output + name + "_residue_info.txt" with open(tmp_output, 'w') as txt_writer: txt_writer.write('residue name_resid_chain\n') for res in residues: res_id = getResidueID(res) txt_writer.write('{}: \n'.format(res_id)) calc_time = time.time() - initial_time print('> Cdf and amino acid residue number list files generated in {}s'. format(int(calc_time)))
def _CDPLreadFromPDBFile(pdb_file): ''' PRIVAT METHOD reads a pdb file and is used by the CDPLreadProteinFile method. Input: \n pdb_file (string): the path to the pdb file \n Return: \n (CDPL BasicMolecule): the corresponding pdb molecule ''' ifs = Base.FileIOStream(pdb_file, 'r') pdb_reader = Biomol.PDBMoleculeReader(ifs) pdb_mol = Chem.BasicMolecule() Biomol.setPDBApplyDictAtomBondingToNonStdResiduesParameter( pdb_reader, False ) #TODO Should this be there for the pdb readin? or also in the config? if not pdb_reader.read(pdb_mol): log.error("COULD NOT READ PDB", pdb_file) return False return pdb_mol
def generate_ph(pdb, key): ifs = Base.FileIOStream(pdb, 'r') tlc = self.ligand_3_letter_code pdb_reader = Biomol.PDBMoleculeReader(ifs) pdb_mol = Chem.BasicMolecule() print '- Reading input: ', pdb, ' ...' if not pdb_reader.read(pdb_mol): print '!! Could not read input molecule' return print '- Processing macromolecule', pdb, ' ...' i = 0 while i < pdb_mol.getNumBonds(): bond = pdb_mol.getBond(i) if Chem.isMetal(bond.atoms[0]) or Chem.isMetal(bond.atoms[1]): pdb_mol.removeBond(i) else: i += 1 Chem.calcImplicitHydrogenCounts(pdb_mol, True) Chem.perceiveHybridizationStates(pdb_mol, True) Chem.makeHydrogenComplete(pdb_mol) Chem.setAtomSymbolsFromTypes(pdb_mol, False) Chem.calcImplicitHydrogenCounts(pdb_mol, True) Biomol.setHydrogenResidueSequenceInfo(pdb_mol, False) Chem.setRingFlags(pdb_mol, True) Chem.setAromaticityFlags(pdb_mol, True) Chem.generateHydrogen3DCoordinates(pdb_mol, True) ligand = Chem.Fragment() print '- Extracting ligand ', tlc, ' ...' for atom in pdb_mol.atoms: if Biomol.getResidueCode(atom) == tlc: Biomol.extractResidueSubstructure(atom, pdb_mol, ligand, False) break if ligand.numAtoms == 0: print '!! Could not find ligand', tlc, 'in input file' return Chem.perceiveSSSR(ligand, True) lig_env = Chem.Fragment() Biomol.extractEnvironmentResidues(ligand, pdb_mol, lig_env, 7.0) Chem.perceiveSSSR(lig_env, True) print '- Constructing pharmacophore ...' lig_pharm = Pharm.BasicPharmacophore() env_pharm = Pharm.BasicPharmacophore() pharm_gen = Pharm.DefaultPharmacophoreGenerator(False) pharm_gen.generate(ligand, lig_pharm) pharm_gen.generate(lig_env, env_pharm) analyzer = Pharm.DefaultInteractionAnalyzer() interactions = Pharm.FeatureMapping() analyzer.analyze(lig_pharm, env_pharm, interactions) #------------------------- XVOLS int_env_ftrs = Pharm.FeatureSet() Pharm.getFeatures(int_env_ftrs, interactions, False) int_core_ftrs = Pharm.FeatureSet() Pharm.getFeatures(int_core_ftrs, interactions, True) int_pharm = Pharm.BasicPharmacophore(int_core_ftrs) for ftr in int_env_ftrs: if Pharm.getType( ftr ) == Pharm.FeatureType.H_BOND_DONOR or Pharm.getType( ftr) == Pharm.FeatureType.H_BOND_ACCEPTOR: Pharm.setTolerance(ftr, 1.0) else: Pharm.setTolerance(ftr, 1.5) Pharm.createExclusionVolumes(int_pharm, int_env_ftrs, 0.0, 0.1, False) int_env_ftr_atoms = Chem.Fragment() Pharm.getFeatureAtoms(int_env_ftrs, int_env_ftr_atoms) int_residue_atoms = Chem.Fragment() Biomol.extractResidueSubstructures(int_env_ftr_atoms, lig_env, int_residue_atoms, True) Chem.makeHydrogenDeplete(int_residue_atoms) def isAlphaAtom(atom): return Biomol.getResidueAtomName(atom) == 'CA' Chem.removeAtomsIfNot(int_residue_atoms, isAlphaAtom) Pharm.createExclusionVolumes(int_pharm, int_residue_atoms, Chem.Atom3DCoordinatesFunctor(), 1.0, 2.0, False) features_in_ph = [] for int_ftr in int_pharm: if Pharm.hasSubstructure(int_ftr) == False: continue elif ftype_names[Pharm.getType(int_ftr)] == 'XV': continue feature_id = generate_key(int_ftr) features_in_ph.append(str(feature_id)) self.unique_feature_vector.add(str(feature_id)) int_pharm.fv = features_in_ph int_pharm.path_to_pdb = pdb return int_pharm
def generate_ph(pdb, args, df_constructor, ts): ifs = Base.FileIOStream(pdb, 'r') tlc = args.ligand_three_letter_code pdb_reader = Biomol.PDBMoleculeReader(ifs) pdb_mol = Chem.BasicMolecule() print '- Reading input: ', pdb, ' ...' if not pdb_reader.read(pdb_mol): print '!! Could not read input molecule' return print '- Processing macromolecule', pdb, ' ...' i = 0 while i < pdb_mol.getNumBonds(): bond = pdb_mol.getBond(i) if Chem.isMetal(bond.atoms[0]) or Chem.isMetal(bond.atoms[1]): pdb_mol.removeBond(i) else: i += 1 for a in pdb_mol.atoms: Chem.setImplicitHydrogenCount(a, 0) Chem.calcImplicitHydrogenCounts(pdb_mol, True) Chem.perceiveHybridizationStates(pdb_mol, True) Chem.makeHydrogenComplete(pdb_mol) Chem.setAtomSymbolsFromTypes(pdb_mol, False) Chem.calcImplicitHydrogenCounts(pdb_mol, True) Biomol.setHydrogenResidueSequenceInfo(pdb_mol, False) Chem.setRingFlags(pdb_mol, True) Chem.setAromaticityFlags(pdb_mol, True) Chem.generateHydrogen3DCoordinates(pdb_mol, True) Chem.calcFormalCharges(pdb_mol, True) ligand = Chem.Fragment() print '- Extracting ligand ', tlc, ' ...' for atom in pdb_mol.atoms: if Biomol.getResidueCode(atom) == tlc: Biomol.extractResidueSubstructure(atom, pdb_mol, ligand, False) break if ligand.numAtoms == 0: print '!! Could not find ligand', tlc, 'in input file' return Chem.perceiveSSSR(ligand, True) lig_env = Chem.Fragment() Biomol.extractEnvironmentResidues(ligand, pdb_mol, lig_env, 7.0) Chem.perceiveSSSR(lig_env, True) print '- Constructing pharmacophore ...' lig_pharm = Pharm.BasicPharmacophore() env_pharm = Pharm.BasicPharmacophore() pharm_gen = Pharm.DefaultPharmacophoreGenerator(True) pharm_gen.generate(ligand, lig_pharm) pharm_gen.generate(lig_env, env_pharm) #Pharm.FilePMLFeatureContainerWriter('./test/lig_ph_' + str(ts) + '.pml').write(lig_pharm) analyzer = Pharm.DefaultInteractionAnalyzer() interactions = Pharm.FeatureMapping() analyzer.analyze(lig_pharm, env_pharm, interactions) df_constructor, interaction_at_ts = outputInteractions( lig_pharm, env_pharm, interactions, df_constructor) #Chem.FileSDFMolecularGraphWriter('./test/ligand_' + str(ts) + '.sdf').write(ligand) return df_constructor, interaction_at_ts
def readPDBFromStream(stream: Base.IOStream): r = Biomol.PDBMoleculeReader(stream) mol = Chem.BasicMolecule() r.read(mol) return Protein(mol)