def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
oemolprop.OEConfigureFilterParams(itf)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
iname = itf.GetString("-in")
ifs = oechem.oemolistream()
if not ifs.open(iname):
oechem.OEThrow.Fatal("Cannot open input file!")
ftype = oemolprop.OEGetFilterType(itf)
filt = oemolprop.OEFilter(ftype)
ver = itf.GetInt("-verbose")
oechem.OEThrow.SetLevel(ver)
ostr = oechem.oeosstream()
pwnd = False
filt.SetTable(ostr, pwnd)
headers = ostr.str().split(b'\t')
ostr.clear() # remove the header row from the stream
for mol in ifs.GetOEGraphMols():
filt(mol)
fields = ostr.str().decode("UTF-8").split('\t')
ostr.clear() # remove this row from the stream
tmpdct = dict(zip(headers, fields))
print(mol.GetTitle(), tmpdct[b"Lipinski violations"])
def filter_smiles(smis):
'''
:param smis: list of smiles
:return: (oe graph mols, smiles)
'''
ims = oechem.oemolistream()
ims.SetFormat(oechem.OEFormat_SMI)
smiles = "\n".join(list(smis))
ims.openstring(smiles)
oms = oechem.oemolostream()
oms.SetFormat(oechem.OEFormat_SMI)
oms.openstring()
filt = oemolprop.OEFilter(oemolprop.OEFilterType_BlockBuster)
goods = []
counter, good = 0, 0
for i, mol in enumerate(ims.GetOEGraphMols()):
counter += 1
if filt(mol):
oechem.OEWriteMolecule(oms, mol)
goods.append(i)
good += 1
print("FILTERD", counter, "ACCEPTED", good)
actions = str(oms.GetString().decode("utf-8"))
actions = actions.split("\n")
oms.close()
ims.close()
return [smis[i] for i in goods], actions
def from_smiles(in_smiles, outfile):
'''
:param in_smiles: a list of smile strings
:param outfile:
:return:
'''
in_smiles = "\n".join(in_smiles)
ims = oechem.oemolistream()
ims.SetFormat(oechem.OEFormat_SMI)
ims.openstring(in_smiles)
ofs = oechem.oemolostream()
ofs.SetFormat(oechem.OEFormat_OEB)
filt = oemolprop.OEFilter(oemolprop.OEFilterType_BlockBuster)
if not ofs.open(outfile):
oechem.OEThrow.Fatal("Unable to open %s for writing" % outfile)
for mol in ims.GetOEMols():
if filt(mol):
result = gen_conf(mol)
for res in result:
oechem.OEWriteMolecule(ofs, res)
ofs.close()
return 0
def from_single_smiles(in_smiles, outfile):
mol = oechem.OEMol()
oechem.OESmilesToMol(mol, in_smiles)
ofs = oechem.oemolostream()
ofs.SetFormat(oechem.OEFormat_OEB)
if not ofs.open(outfile):
oechem.OEThrow.Fatal("Unable to open %s for writing" % outfile)
filt = oemolprop.OEFilter(oemolprop.OEFilterType_BlockBuster)
if filt(mol):
result = gen_conf(mol)
for res in result:
oechem.OEWriteMolecule(ofs, res)
ofs.close()
return 0
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
oemolprop.OEConfigureFilterParams(itf)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
iname = itf.GetString("-in")
oname = itf.GetString("-out")
ifs = oechem.oemolistream()
if not ifs.open(iname):
oechem.OEThrow.Fatal("Cannot open input file!")
ftype = oemolprop.OEGetFilterType(itf)
filt = oemolprop.OEFilter(ftype)
ver = itf.GetInt("-verbose")
oechem.OEThrow.SetLevel(ver)
oemolprop.OEWritePropertyDataToCSV(oname, ifs, filt)
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
oemolprop.OEConfigureFilterParams(itf)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
iname = itf.GetString("-in")
oname = itf.GetString("-out")
ifs = oechem.oemolistream()
if not ifs.open(iname):
oechem.OEThrow.Fatal("Cannot open input file!")
ofs = oechem.oemolostream()
if not ofs.open(oname):
oechem.OEThrow.Fatal("Cannot create output file!")
fmt = ofs.GetFormat()
if fmt not in [
oechem.OEFormat_SDF, oechem.OEFormat_OEB, oechem.OEFormat_CSV
]:
oechem.OEThrow.Fatal("Only SD, OEB, and CSV formats preserve SD data")
ftype = oemolprop.OEGetFilterType(itf)
filt = oemolprop.OEFilter(ftype)
ver = itf.GetInt("-verbose")
oechem.OEThrow.SetLevel(ver)
pwnd = False
filt.SetTable(oechem.oenul, pwnd)
filt.SetSDTag(True)
for mol in ifs.GetOEGraphMols():
filt(mol)
oechem.OEWriteMolecule(ofs, mol)
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
oemolprop.OEConfigureFilterParams(itf)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
iname = itf.GetString("-in")
oname = itf.GetString("-out")
ifs = oechem.oemolistream()
if not ifs.open(iname):
oechem.OEThrow.Fatal("Cannot open input file!")
ofs = oechem.oemolostream()
if not ofs.open(oname):
oechem.OEThrow.Fatal("Cannot create output file!")
ftype = oemolprop.OEGetFilterType(itf)
filt = oemolprop.OEFilter(ftype)
for mol in ifs.GetOEGraphMols():
if filt(mol):
oechem.OEWriteMolecule(ofs, mol)
def compute_conformers(smiles=None, smiles_file=None, start_index=0, batch_size=0, out_file=None, bad_file=None, save_csv=False, overwrite=False, save_gzip=False, license=None, timeout=0, max_failures=2):
import csv
import os
from openeye import oechem
from openeye import oeomega
from openeye import oemolprop
import signal
os.environ['OE_LICENSE'] = license
if save_gzip or save_csv:
raise Exception("GZip and CSV not supported")
if not overwrite and os.path.exists(out_file):
raise Exception("File exists: %s" % out_file)
if smiles_file:
with open(smiles_file) as current:
current.seek(start_index)
smiles = [current.readline() for i in range(batch_size)]
if len(smiles) == 0:
return ""
# function to compute enantiomers
# separated out so we can use an alarm to timeout
def get_enan(omega, enan):
enan = oechem.OEMol(enan)
ret = omega.Build(enan)
return enan, ret
def alarm_handler(signum, frame):
#print("ALARM signal received")
raise Exception()
#ofs = oechem.oemolostream()
#ofs.SetFormat(oechem.OEFormat_OEB)
#if not ofs.open(out_file):
# oechem.OEThrow.Fatal("Unable to open %s for writing" % out_file)
sep = ","
bad = []
mols = []
for s in smiles:
mol = s.split(sep)
if len(mol) > 1:
mols.append(mol[2].rstrip())
# put all mols in a string as we're told it is faster to process this way
in_smiles = "\n".join(mols)
ims = oechem.oemolistream()
ims.SetFormat(oechem.OEFormat_SMI)
ims.openstring(in_smiles)
# Turn off logging except errors
oechem.OEThrow.SetLevel(5)
filt = oemolprop.OEFilter(oemolprop.OEFilterType_BlockBuster)
#ofs.open(out_file)
signal.signal(signal.SIGALRM, alarm_handler)
oe_results = []
for mol in ims.GetOEMols():
if filt(mol):
oemols = []
ret_code = None
omegaOpts = oeomega.OEOmegaOptions(oeomega.OEOmegaSampling_FastROCS)
omega = oeomega.OEOmega(omegaOpts)
failures = 0
for enantiomer in oeomega.OEFlipper(mol.GetActive(), 6, True):
if max_failures > 0 and failures >= max_failures:
break
if len(oemols) >= 10:
break
ret_code = None
error = False
signal.alarm(timeout)
try:
enantiomer, ret_code = get_enan(omega, enantiomer)
except:
print("Timeout %s" % out_file)
failures += 1
error = True
signal.alarm(0)
if not error and ret_code == oeomega.OEOmegaReturnCode_Success:
halfMol = oechem.OEMol(mol, oechem.OEMCMolType_HalfFloatCartesian)
oemols.append(halfMol)
#else:
#oechem.OEThrow.Warning("%s: %s" %
# (enantiomer.GetTitle(), oeomega.OEGetOmegaError(ret_code)))
oe_results.append(oemols)
ofs = oechem.oemolostream()
ofs.SetFormat(oechem.OEFormat_OEB)
ofs.open(out_file)
for r in oe_results:
for res in r:
oechem.OEWriteMolecule(ofs, res)
ofs.close()
return out_file
#!/usr/bin/env python
"""
Extract SMILES from SMIRNOFF_sub3rot.tar.gz to generate input.smi
"""
from openeye import oechem
from openeye import oemolprop
import gzip
filterfile = oechem.oeifstream('oechem-filterfile')
filter = oemolprop.OEFilter(filterfile)
ifs = oechem.oemolistream('nci-250k.smi.gz')
ofs = oechem.oemolostream('input.smi')
for mol in ifs.GetOEMols():
if filter(mol):
smiles = oechem.OEMolToSmiles(mol)
oechem.OEWriteMolecule(ofs, mol)
ifs.close()
ofs.close()
def compute_druggables(smiles=None,
smiles_file=None,
start_index=0,
batch_size=0,
out_file=None,
bad_file=None,
save_csv=False,
overwrite=False,
save_gzip=False,
license=None,
timeout=0,
max_failures=2):
import csv
import os
from openeye import oechem
from openeye import oeomega
from openeye import oemolprop
import signal
os.environ['OE_LICENSE'] = license
if save_gzip or not save_csv:
raise Exception("GZip and PKL not supported")
if not overwrite and os.path.exists(out_file):
raise Exception("File exists: %s" % out_file)
if smiles_file:
with open(smiles_file) as current:
current.seek(start_index)
smiles = [current.readline() for i in range(batch_size)]
if len(smiles) == 0:
return ""
sep = ","
mols = []
full_mols = []
for s in smiles:
mol = s.split(sep)
if len(mol) > 1:
mols.append(mol[2].rstrip())
full_mols.append(mol)
# put all mols in a string as we're told it is faster to process this way
in_smiles = "\n".join(mols)
ims = oechem.oemolistream()
ims.SetFormat(oechem.OEFormat_SMI)
ims.openstring(in_smiles)
# Turn off logging except errors
oechem.OEThrow.SetLevel(5)
filt = oemolprop.OEFilter(oemolprop.OEFilterType_BlockBuster)
bad = []
results = []
for (m, mol) in zip(ims.GetOEMols(), full_mols):
if len(mol) > 1:
results.append((mol[0], mol[1], filt(m)))
with open(out_file, 'w') as output_file:
writer = csv.writer(output_file, delimiter=',')
writer.writerows(results)
return out_file