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 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 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()
import CDPL.Base as Base
import CDPL.Chem as Chem
import CDPL.Math as Math
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')
mlr_model = Math.DMLRModel()
sdf_reader = Chem.SDFMoleculeReader(struct_is)
mol = Chem.BasicMolecule()
xlogp_calc = Chem.XLogPCalculator()
histo = Math.DVector()
histo.resize(Chem.XLogPCalculator.FEATURE_VECTOR_SIZE)
Chem.setMultiConfImportParameter(sdf_reader, False)
while sdf_reader.read(mol):
exp_logp = float(exp_logp_is.readline())
Chem.perceiveComponents(mol, False)
Chem.perceiveSSSR(mol, False)
Chem.setRingFlags(mol, False)
Chem.calcImplicitHydrogenCounts(mol, False)
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)
