def process():
if len(sys.argv) < 4:
print('Usage:', sys.argv[0], 'training-set.sdf logP-data regression-coeff-file', file=sys.stderr)
sys.exit(2)
struct_is = Base.FileIOStream(sys.argv[1], 'r')
exp_logp_is = Base.FileIOStream(sys.argv[2], 'r')
coeff_os = Base.FileIOStream(sys.argv[3], 'w')
def process():
if len(sys.argv) < 4:
print('Usage:', sys.argv[0], '[training-set.sdf] [logS-data] [regression coeff. output file]', file=sys.stderr)
sys.exit(2)
struct_is = Base.FileIOStream(sys.argv[1], 'r')
exp_logs_is = Base.FileIOStream(sys.argv[2], 'r')
coeff_os = Base.FileIOStream(sys.argv[3], 'w')
def process():
if len(sys.argv) < 4:
print('Usage:',
sys.argv[0],
'[input.sdf] [exclude-molecules.sdf] [output.sdf]',
file=sys.stderr)
sys.exit(2)
ifs = Base.FileIOStream(sys.argv[1], 'r')
xifs = Base.FileIOStream(sys.argv[2], 'r')
ofs = Base.FileIOStream(sys.argv[3], 'w')
reader = Chem.SDFMoleculeReader(ifs)
xreader = Chem.SDFMoleculeReader(xifs)
writer = Chem.SDFMolecularGraphWriter(ofs)
mol = Chem.BasicMolecule()
Chem.setMultiConfImportParameter(reader, False)
Chem.setMultiConfImportParameter(xreader, False)
Chem.setMultiConfExportParameter(writer, False)
stats = Stats()
stats.read = 0
stats.dropped = 0
xhashes = set()
while xreader.read(mol):
setupMolecule(mol)
hashcode = Chem.calcHashCode(mol)
xhashes.add(hashcode)
while reader.read(mol):
#print('Processing Molecule ' + str(stats.read)
setupMolecule(mol)
hashcode = Chem.calcHashCode(mol)
if hashcode in xhashes:
stats.dropped += 1
print('Dropped Molecule ' + str(stats.read) + ': ' +
Chem.generateSMILES(mol) + ' ' + Chem.getName(mol),
file=sys.stderr)
else:
writer.write(mol)
stats.read += 1
if stats.read % 10000 == 0:
print('Processed ' + str(stats.read) + ' Molecules...',
file=sys.stderr)
print('', file=sys.stderr)
print('-- Summary --', file=sys.stderr)
print('Molecules processed: ' + str(stats.read), file=sys.stderr)
print('Molecules dropped: ' + str(stats.dropped), file=sys.stderr)
def read_in_ph(ph_path, output_dir_path):
fr = Pharm.PMLPharmacophoreReader(Base.FileIOStream(ph_path))
ph = Pharm.BasicPharmacophore()
fr.read(ph)
ph.pml_path = ph_path
ph.dir_path = output_dir_path
return ph
def savePharmacophore(pha, path):
'''
Saves a particula pha at the target path.\n
Input:\n
pha (CDPL BasicPharmacophore): the pharmacophore to be saved as a pml file \n
path (String): path where to save the pml file (includes the filename.pml)
'''
Pharm.PMLFeatureContainerWriter(Base.FileIOStream(path, 'w')).write(pha)
return True
def write_ph_for_rpms(self, rpm_maps, output_directory):
for fv in rpm_maps:
directory = output_directory + '/' + str(fv)
if not os.path.exists(directory):
os.makedirs(directory)
for ph_key in rpm_maps[fv]:
ph_to_write = directory + '/ph_' + str(fv) + '_' + str(
ph_key) + '.pml'
print '- Writing pharmacophore: ' + str(ph_to_write)
Pharm.PMLFeatureContainerWriter(
Base.FileIOStream(ph_to_write,
'w')).write(rpm_maps[ph_key])
def CDPLphaFromPML(pml_path):
'''
reads a single CDPL BasicPharmacophore from an pml-file.
Input: \n
pml_path (string): path to the pml file \n
Return: \n
(CDPL BasicPharmacophore): the corresponding CDPL BasicPharmacophore
'''
pha = Pharm.BasicPharmacophore()
ifs = Base.FileIOStream(pml_path, 'r')
pml_reader = Pharm.PMLPharmacophoreReader(ifs)
if not pml_reader.read(pha):
log.error("COULD NOT READ PML", pml_path)
return False
return pha
def CDPLmolFromSdf(sdf_path, conformation):
'''
generates a single CDPL Molecule from an sdf-file. If conformations is true, then
one random conformation will be generated. \n
Input: \n
sdf_path (string): path to the sdf file \n
conformation (boolean): generates one 3d conformation according to MMFF94 \n
Return: \n
(CDPL BasicMolecule): the corresponding CDPL BasicMolecule
'''
mol = Chem.BasicMolecule()
ifs = Base.FileIOStream(sdf_path, 'r')
sdf_reader = Chem.SDFMoleculeReader(ifs)
if not sdf_reader.read(mol):
log.error("COULD NOT READ SDF", sdf_path)
return False
if conformation:
return _CDPLgenerateConformation(mol)
return mol
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 process(sdf_file, psd_file_path):
ifs = Base.FileIOStream(sdf_file, 'r')
reader = Chem.SDFMoleculeReader(ifs)
mol = Chem.BasicMolecule()
Chem.setMultiConfImportParameter(reader, True)
psd_creator = Pharm.PSDScreeningDBCreator(
psd_file_path, Pharm.PSDScreeningDBCreator.CREATE, True)
i = 0
t0 = time.clock()
while reader.read(mol):
setupMolecule(mol)
psd_creator.process(mol)
i += 1
if i % 100 == 0:
print 'Processed ' + str(i) + ' molecules (' + str(
time.clock() - t0), 's elapsed)...'
t0 = time.clock()
mol.clear()
print ''
print '-- Summary --'
print 'Molecules processed: ' + str(psd_creator.numProcessed)
print 'Molecules rejected: ' + str(psd_creator.numRejected)
print 'Molecules deleted: ' + str(psd_creator.numDeleted)
print 'Molecules inserted: ' + str(psd_creator.numInserted)
psd_creator.close()
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 process():
if len(sys.argv) < 4:
print(
'Usage:',
sys.argv[0],
'[input torsion rules.xml] [structures.sdf] [output torsion histogram library.sdf]',
file=sys.stderr)
sys.exit(2)
tor_lib = ConfGen.TorsionLibrary()
try:
tor_lib.load(Base.FileIOStream(sys.argv[1], 'r'))
except:
print('Error while loading input torsion rules:',
sys.exc_info()[0],
file=sys.stderr)
sys.exit(2)
tor_matcher = ConfGen.TorsionRuleMatcher(tor_lib)
tor_matcher.findAllRuleMappings(True)
tor_matcher.findUniqueMappingsOnly(True)
tor_matcher.stopAtFirstMatchingRule(True)
mol = Chem.BasicMolecule()
mol_reader = Chem.FileSDFMoleculeReader(sys.argv[2])
Chem.setMultiConfImportParameter(mol_reader, False)
print('- Analyzing input structures...', file=sys.stderr)
i = 1
rule_to_angle_hists = {}
coords = Math.Vector3DArray()
while True:
try:
if not mol_reader.read(mol):
break
except IOError as e:
print('Error while reading input molecule',
i,
':',
e,
file=sys.stderr)
i += 1
continue
if i % 500 == 0:
print(' ... At input molecule', i, '...', file=sys.stderr)
Chem.initSubstructureSearchTarget(mol, False)
try:
Chem.get3DCoordinates(mol, coords)
except Base.ItemNotFound:
print('Could not get 3D-coordinates for molecule',
i,
file=sys.stderr)
i += 1
continue
for bond in mol.bonds:
if Chem.getRingFlag(bond):
continue
if Chem.isHydrogenBond(bond):
continue
if Chem.getExplicitBondCount(
bond.getBegin()) <= 1 or Chem.getExplicitBondCount(
bond.getEnd()) <= 1:
continue
tor_matcher.findMatches(bond, mol, False)
for match in tor_matcher:
processMatch(i, match, mol, coords, rule_to_angle_hists)
i += 1
print('- Processing torsion angle histograms...', file=sys.stderr)
processHistograms(tor_lib, rule_to_angle_hists)
print('- Writing output torsion library...', file=sys.stderr)
try:
tor_lib.save(Base.FileIOStream(sys.argv[3], 'w+'))
except:
print('Error while writing torsion library:',
sys.exc_info()[0],
file=sys.stderr)
sys.exit(2)
print('DONE!', file=sys.stderr)
def readPDBFromFile(path: str) -> Chem.BasicMolecule:
s = Base.FileIOStream(path)
protein = readPDBFromStream(s)
return protein
def writePDB(path: str, protein: Chem.BasicMolecule) -> None:
s = Base.FileIOStream(path)
w = Biomol.PDBMolecularGraphWriter(s)
w.write(protein)
w.close()
def cleanStructures():
if len(sys.argv) < 5:
print('Usage:',
sys.argv[0],
'[input.sdf] [output.sdf] [dropped.sdf] [start_index] [[count]]',
file=sys.stderr)
sys.exit(2)
ifs = Base.FileIOStream(sys.argv[1], 'r')
ofs = Base.FileIOStream(sys.argv[2], 'w')
dofs = Base.FileIOStream(sys.argv[3], 'w')
offset = int(sys.argv[4])
count = 0
if len(sys.argv) > 5:
count = int(sys.argv[5])
reader = Chem.SDFMoleculeReader(ifs)
writer = Chem.SDFMolecularGraphWriter(ofs)
dwriter = Chem.SDFMolecularGraphWriter(dofs)
mol = Chem.BasicMolecule()
#Chem.setSMILESRecordFormatParameter(reader, 'SN')
stats = Stats()
stats.read = 0
stats.dropped = 0
stats.modified = 0
Chem.setMultiConfImportParameter(reader, False)
Chem.setMultiConfExportParameter(writer, False)
Chem.setMultiConfExportParameter(dwriter, False)
if offset > 0:
print('Skipping Molecules to Start Index ' + str(offset),
file=sys.stderr)
reader.setRecordIndex(offset)
#print('Finished Setting Record Index', file=sys.stderr)
stats.read = offset
while reader.read(mol):
#print('Processing Molecule ' + str(stats.read)
proc_mol = processMolecule(mol, stats)
if proc_mol:
writer.write(proc_mol)
else:
stats.dropped += 1
dwriter.write(mol)
print('Dropped Molecule ' + str(stats.read) + ': ' +
generateSMILES(mol) + ' ' + Chem.getName(mol),
file=sys.stderr)
stats.read += 1
if stats.read % 10000 == 0:
print('Processed ' + str(stats.read - offset) + ' Molecules...',
file=sys.stderr)
if count > 0 and (stats.read - offset) >= count:
break
print('', file=sys.stderr)
print('-- Summary --', file=sys.stderr)
print('Molecules processed: ' + str(stats.read - offset), file=sys.stderr)
print('Molecules dropped: ' + str(stats.dropped), file=sys.stderr)
print('Molecules modified: ' + str(stats.modified), file=sys.stderr)