def createOeBinaryDatabaseAndIndex(self, oebMolFilePath, oeMolDbFilePath):
"""Create OE binary database file and associated index from the input serial
binary data file.
Args:
oebMolFilePath (str): input OeMol stream binary file path
oeMolDbFilePath (str): output OeMolDatabase file path
Returns:
int: number of molecules processed in the database.
"""
molCount = 0
try:
startTime = time.time()
moldb = oechem.OEMolDatabase()
if not moldb.Open(oebMolFilePath):
logger.error("Read fails for %r", oebMolFilePath)
return molCount
#
logger.info(
"Opened database in format %r num mols %d max index %d",
moldb.GetFormat(), moldb.NumMols(), moldb.GetMaxMolIdx())
moldb.Save(oeMolDbFilePath)
tL = list(moldb.GetTitles())
logger.info("First and last titles: %r %r", tL[0], tL[-1])
molCount = moldb.NumMols()
endTime = time.time()
logger.info("Completed operation at %s (%.4f seconds)",
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - startTime)
except Exception as e:
logger.exception("Failing with %s", str(e))
return molCount
def _sort_by_rotbond(ifs, outdir):
"""
Parameters
----------
ifs: str
absolute path to molecule database
outdir: str
absolute path to where output files should be written.
"""
nrotors_map = {}
moldb = oechem.OEMolDatabase(ifs)
mol = oechem.OEGraphMol()
for idx in range(moldb.GetMaxMolIdx()):
if moldb.GetMolecule(mol, idx):
nrotors = sum([bond.IsRotor() for bond in mol.GetBonds()])
if nrotors not in nrotors_map:
nrotors_map[nrotors] = []
nrotors_map[nrotors].append(idx)
# Write out a separate database for each num rotors
for nrotor in nrotors_map:
size = len(nrotors_map[nrotor])
ofname = os.path.join(outdir, 'nrotor_{}.smi'.format(nrotor))
ofs = new_output_stream(ofname)
write_oedatabase(moldb, ofs, nrotors_map[nrotor], size)
def initialize(self, dbdict):
self('DELETE FROM db', commit=True)
for name, path in dbdict.items():
db = oechem.OEMolDatabase(path)
nmols = db.NumMols()
self('INSERT INTO db (name, path, molcount) VALUES (?,?,?)',
(name, path, nmols),
commit=True)
def SplitChunk(ifs, chunksize, outbase, ext):
moldb = oechem.OEMolDatabase(ifs)
chunk, count = 1, chunksize
for idx in range(moldb.GetMaxMolIdx()):
if count == chunksize:
ofs = NewOutputStream(outbase, ext, chunk)
chunk, count = chunk + 1, 0
count += 1
moldb.WriteMolecule(ofs, idx)
def MolSort(ifs, ofs):
moldb = oechem.OEMolDatabase(ifs)
titles = [(t, i) for i, t in enumerate(moldb.GetTitles())]
titles.sort()
indices = [i for t, i in titles]
moldb.Order(indices)
moldb.Save(ofs)
def MolCount(fname):
ifs = oechem.oemolistream()
if not ifs.open(fname):
oechem.OEThrow.Warning("Unable to open %s for reading" % fname)
return 0
moldb = oechem.OEMolDatabase(ifs)
nummols = moldb.NumMols()
print("%s contains %d molecule(s)." % (fname, nummols))
return nummols
def main():
if not oefastrocs.OEFastROCSIsGPUReady():
oechem.OEThrow.Info("No supported GPU available!")
return 0
args = getargs()
dbname = args.d
# read in database
ifs = oechem.oemolistream()
if not ifs.open(dbname):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
print("Opening database file %s ..." % dbname)
timer = oechem.OEWallTimer()
opts = oefastrocs.OEShapeDatabaseOptions()
opts.SetLimit(1)
dbase = oefastrocs.OEShapeDatabase()
moldb = oechem.OEMolDatabase()
if not moldb.Open(ifs):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
dots = oechem.OEThreadedDots(10000, 200, "conformers")
if not dbase.Open(moldb, dots):
oechem.OEThrow.Fatal("Unable to initialize OEShapeDatabase on '%s'" %
dbname)
dots.Total()
print("%f seconds to load database" % timer.Elapsed())
df = pd.read_csv(args.i)
res = []
for smile in tqdm(df.loc[:, 'smiles'].tolist()):
resn = len(res)
try:
q = FromString(smile)[0]
for score in dbase.GetSortedScores(q, 1):
res.append(score.GetTanimotoCombo())
break
except KeyboardInterrupt:
print("caught")
exit()
except:
res.append(np.nan)
if len(res) == resn:
res.append(np.nan)
df['fastroc'] = res
print(df.head)
df.to_csv(args.o, sep=',', index=False)
return 0
def loadOeBinaryDatabaseAndIndex(self, oeMolDbFilePath):
molDb = None
try:
molDb = oechem.OEMolDatabase()
if not molDb.Open(oeMolDbFilePath):
logger.error("Unable to open %r", oeMolDbFilePath)
molCount = molDb.NumMols()
logger.info("Loaded OE database file containing %d molecules",
molCount)
except Exception as e:
logger.exception("Loading %r failing with %s", oeMolDbFilePath,
str(e))
return molDb
def test_sort_nrotors(self):
"""Tests sorting fragments by nrotors"""
ifs = get_fn('frags.smi')
out_dir = self.get_writes_dir()
fragment._sort_by_rotbond(ifs, outdir=out_dir)
mol = oechem.OEGraphMol()
for n in range(1, 6):
db = get_fn('nrotor_{}.smi'.format(n), written=True)
moldb = oechem.OEMolDatabase(db)
for idx in range(moldb.GetMaxMolIdx()):
if moldb.GetMolecule(mol, idx):
nrotor = sum([bond.IsRotor() for bond in mol.GetBonds()])
self.assertEquals(n, nrotor)
def main(argv=[__name__]):
itf = oechem.OEInterface()
oechem.OEConfigure(itf, InterfaceData)
oegraphsim.OEConfigureFingerPrint(
itf, oegraphsim.OEGetFPType(oegraphsim.OEFPType_Tree))
if not oechem.OEParseCommandLine(itf, argv):
return 1
ifname = itf.GetString("-in")
ffname = itf.GetString("-fpdb")
if oechem.OEGetFileExtension(ffname) != "fpbin":
oechem.OEThrow.Fatal(
"Fingerprint database file should have '.fpbin' file extension!")
idxfname = oechem.OEGetMolDatabaseIdxFileName(ifname)
if not os.path.exists(idxfname):
if not oechem.OECreateMolDatabaseIdx(ifname):
oechem.OEThrow.Warning("Unable to create %s molecule index file" %
idxfname)
oechem.OEThrow.Info("Using %s index molecule file" % idxfname)
moldb = oechem.OEMolDatabase()
if not moldb.Open(ifname):
oechem.OEThrow.Fatal("Cannot open molecule database file!")
nrmols = moldb.GetMaxMolIdx()
fptype = oegraphsim.OESetupFingerPrint(itf)
oechem.OEThrow.Info("Using fingerprint type %s" % fptype.GetFPTypeString())
opts = oegraphsim.OECreateFastFPDatabaseOptions(fptype)
opts.SetTracer(oechem.OEDots(100000, 1000, "fingerprints"))
oechem.OEThrow.Info("Generating fingerprints with %d threads" %
opts.GetNumProcessors())
timer = oechem.OEWallTimer()
if not oegraphsim.OECreateFastFPDatabaseFile(ffname, ifname, opts):
oechem.OEThrow.Fatal("Cannot create fingerprint database file!")
oechem.OEThrow.Info("%5.2f secs to generate %d fingerprints" %
(timer.Elapsed(), nrmols))
return 0
def __init__(self, itf):
""" Create a MCMolShapeDatabase from the parameters specified by the OEInterface. """
self.rwlock = ReadWriteLock()
self.loadedEvent = Event()
self.dbname = itf.GetString("-dbase")
self.moldb = oechem.OEMolDatabase()
self.dbtype = GetDatabaseType(itf.GetBool("-shapeOnly"))
self.shapedb = oefastrocs.OEShapeDatabase(*GetShapeDatabaseArgs(itf))
# this thread is daemonic so a KeyboardInterupt
# during the load will cancel the process
self.loaderThread = DatabaseLoaderThread(self.shapedb, self.moldb,
self.dbname, self.loadedEvent)
self.loaderThread.setDaemon(True)
self.loaderThread.start()
def SplitNParts(ifs, nparts, outbase, ext):
moldb = oechem.OEMolDatabase(ifs)
molcount = moldb.NumMols()
chunksize, lft = divmod(molcount, nparts)
if lft != 0:
chunksize += 1
chunk, count = 1, 0
ofs = NewOutputStream(outbase, ext, chunk)
for idx in range(moldb.GetMaxMolIdx()):
count += 1
if count > chunksize:
if chunk == lft:
chunksize -= 1
ofs.close()
chunk, count = chunk + 1, 1
ofs = NewOutputStream(outbase, ext, chunk)
moldb.WriteMolecule(ofs, idx)
def main(argv=[__name__]):
parser = argparse.ArgumentParser()
# positional arguments retaining backward compatibility
parser.add_argument(
'database',
help='File containing the database molecules to be search \
(format not restricted to *.oeb).')
parser.add_argument(
'query',
default=[],
nargs='+',
help='File containing the query molecule(s) to be search \
(format not restricted to *.oeb).')
parser.add_argument(
'--nHits',
dest='nHits',
type=int,
default=100,
help='Number of hits to return (default = number of database mols).')
parser.add_argument('--cutoff',
dest='cutoff',
type=float,
default=argparse.SUPPRESS,
help='Specify a cutoff criteria for scores.')
parser.add_argument(
'--tversky',
dest='tversky',
action='store_true',
default=argparse.SUPPRESS,
help='Switch to Tversky similarity scoring (default = Tanimoto).')
args = parser.parse_args()
dbname = args.database
if not oefastrocs.OEFastROCSIsGPUReady():
oechem.OEThrow.Info("No supported GPU available!")
return 0
# set options
opts = oefastrocs.OEShapeDatabaseOptions()
opts.SetLimit(args.nHits)
print("Number of hits set to %u" % opts.GetLimit())
if hasattr(args, 'cutoff') is not False:
opts.SetCutoff(args.cutoff)
print("Cutoff set to %f" % args.cutoff)
if hasattr(args, 'tversky') is not False:
opts.SetSimFunc(args.tversky)
print("Tversky similarity scoring set.")
# read in database
ifs = oechem.oemolistream()
if not ifs.open(dbname):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
print("\nOpening database file %s ..." % dbname)
timer = oechem.OEWallTimer()
dbase = oefastrocs.OEShapeDatabase()
moldb = oechem.OEMolDatabase()
if not moldb.Open(ifs):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
dots = oechem.OEThreadedDots(10000, 200, "conformers")
if not dbase.Open(moldb, dots):
oechem.OEThrow.Fatal("Unable to initialize OEShapeDatabase on '%s'" %
dbname)
dots.Total()
print("%f seconds to load database\n" % timer.Elapsed())
for qfname in args.query:
# read in query
qfs = oechem.oemolistream()
if not qfs.open(qfname):
oechem.OEThrow.Fatal("Unable to open '%s'" % qfname)
mcmol = oechem.OEMol()
if not oechem.OEReadMolecule(qfs, mcmol):
oechem.OEThrow.Fatal("Unable to read query from '%s'" % qfname)
qfs.rewind()
ext = oechem.OEGetFileExtension(qfname)
qmolidx = 0
while oechem.OEReadMolecule(qfs, mcmol):
# write out to file name based on molecule title
ofs = oechem.oemolostream()
moltitle = mcmol.GetTitle()
if len(moltitle) == 0:
moltitle = str(qmolidx)
ofname = moltitle + "_results." + ext
if not ofs.open(ofname):
oechem.OEThrow.Fatal("Unable to open '%s'" % argv[4])
print("Searching for %s of %s (%s conformers)" %
(moltitle, qfname, mcmol.NumConfs()))
qconfidx = 0
for conf in mcmol.GetConfs():
for score in dbase.GetSortedScores(conf, opts):
dbmol = oechem.OEMol()
dbmolidx = score.GetMolIdx()
if not moldb.GetMolecule(dbmol, dbmolidx):
print(
"Unable to retrieve molecule '%u' from the database"
% dbmolidx)
continue
mol = oechem.OEGraphMol(
dbmol.GetConf(oechem.OEHasConfIdx(score.GetConfIdx())))
oechem.OESetSDData(mol, "QueryConfidx", "%s" % qconfidx)
oechem.OESetSDData(mol, "ShapeTanimoto",
"%.4f" % score.GetShapeTanimoto())
oechem.OESetSDData(mol, "ColorTanimoto",
"%.4f" % score.GetColorTanimoto())
oechem.OESetSDData(mol, "TanimotoCombo",
"%.4f" % score.GetTanimotoCombo())
score.Transform(mol)
oechem.OEWriteMolecule(ofs, mol)
qconfidx += 1
print("%s conformers processed" % qconfidx)
print("Wrote results to %s\n" % ofname)
qmolidx += 1
return 0
def main(argv=[__name__]):
if len(argv) < 3:
oechem.OEThrow.Usage("%s <database> [<queries> ... ]" % argv[0])
return 0
# check system
if not oefastrocs.OEFastROCSIsGPUReady():
oechem.OEThrow.Info("No supported GPU available!")
return 0
# read in database
dbname = argv[1]
print("Opening database file %s ..." % dbname)
dbase = oefastrocs.OEShapeDatabase()
moldb = oechem.OEMolDatabase()
if not moldb.Open(dbname):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
dots = oechem.OEThreadedDots(10000, 200, "conformers")
if not dbase.Open(moldb, dots):
oechem.OEThrow.Fatal("Unable to initialize OEShapeDatabase on '%s'" %
dbname)
# customize search options
opts = oefastrocs.OEShapeDatabaseOptions()
opts.SetLimit(5)
for qfname in argv[2:]:
# read in query
qfs = oechem.oemolistream()
if not qfs.open(qfname):
oechem.OEThrow.Fatal("Unable to open '%s'" % qfname)
query = oechem.OEGraphMol()
if not oechem.OEReadMolecule(qfs, query):
oechem.OEThrow.Fatal("Unable to read query from '%s'" % qfname)
ext = oechem.OEGetFileExtension(qfname)
base = qfname[:-(len(ext) + 1)]
# write out everthing to a similary named file
ofs = oechem.oemolostream()
ofname = base + "_results." + ext
if not ofs.open(ofname):
oechem.OEThrow.Fatal("Unable to open '%s'" % argv[4])
oechem.OEWriteMolecule(ofs, query)
print("Searching for %s" % qfname)
for score in dbase.GetSortedScores(query, opts):
print("Score for mol %u(conf %u) %f shape %f color" %
(score.GetMolIdx(), score.GetConfIdx(),
score.GetShapeTanimoto(), score.GetColorTanimoto()))
dbmol = oechem.OEMol()
molidx = score.GetMolIdx()
if not moldb.GetMolecule(dbmol, molidx):
print("Unable to retrieve molecule '%u' from the database" %
molidx)
continue
mol = oechem.OEGraphMol(
dbmol.GetConf(oechem.OEHasConfIdx(score.GetConfIdx())))
oechem.OESetSDData(mol, "ShapeTanimoto",
"%.4f" % score.GetShapeTanimoto())
oechem.OESetSDData(mol, "ColorTanimoto",
"%.4f" % score.GetColorTanimoto())
oechem.OESetSDData(mol, "TanimotoCombo",
"%.4f" % score.GetTanimotoCombo())
score.Transform(mol)
oechem.OEWriteMolecule(ofs, mol)
print("Wrote results to %s" % ofname)
return 0
# current license or subscription to the applicable OpenEye offering.
# THE SAMPLE CODE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
# EXPRESS OR IMPLIED. OPENEYE DISCLAIMS ALL WARRANTIES, INCLUDING, BUT
# NOT LIMITED TO, WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
# PARTICULAR PURPOSE AND NONINFRINGEMENT. In no event shall OpenEye be
# liable for any damages or liability in connection with the Sample Code
# or its use.
# @ <SNIPPET>
from openeye import oechem
import sys
if len(sys.argv) != 4:
oechem.OEThrow.Usage("%s <input> <output> <index>" % sys.argv[0])
moldb = oechem.OEMolDatabase()
if not moldb.Open(sys.argv[1]):
oechem.OEThrow.Fatal("Unable to open %s for reading" % sys.argv[1])
ofs = oechem.oemolostream()
if not ofs.open(sys.argv[2]):
oechem.OEThrow.Fatal("Unable to open %s for writing" % sys.argv[2])
idx = int(sys.argv[3])
mol = oechem.OEMol()
if not moldb.GetMolecule(mol, idx):
oechem.OEThrow.Fatal("Unable to read a molecule from index %u" % idx)
oechem.OEWriteMolecule(ofs, mol)
# @ </SNIPPET>
def main(argv=[__name__]):
itf = oechem.OEInterface()
oechem.OEConfigure(itf, InterfaceData)
defopts = oegraphsim.OEFPDatabaseOptions(10,
oegraphsim.OESimMeasure_Tanimoto)
oegraphsim.OEConfigureFPDatabaseOptions(itf, defopts)
oegraphsim.OEConfigureFPDatabaseMemoryType(itf)
if not oechem.OEParseCommandLine(itf, argv):
return 0
qfname = itf.GetString("-query")
mfname = itf.GetString("-molfname")
ffname = itf.GetString("-fpdbfname")
ofname = itf.GetString("-out")
# initialize databases
timer = oechem.OEWallTimer()
timer.Start()
ifs = oechem.oemolistream()
if not ifs.open(qfname):
oechem.OEThrow.Fatal("Cannot open input file!")
query = oechem.OEGraphMol()
if not oechem.OEReadMolecule(ifs, query):
oechem.OEThrow.Fatal("Cannot read query molecule!")
moldb = oechem.OEMolDatabase()
if not moldb.Open(mfname):
oechem.OEThrow.Fatal("Cannot open molecule database!")
memtype = oegraphsim.OEGetFPDatabaseMemoryType(itf)
fpdb = oegraphsim.OEFastFPDatabase(ffname, memtype)
if not fpdb.IsValid():
oechem.OEThrow.Fatal("Cannot open fingerprint database!")
nrfps = fpdb.NumFingerPrints()
memtypestr = fpdb.GetMemoryTypeString()
ofs = oechem.oemolostream()
if not ofs.open(ofname):
oechem.OEThrow.Fatal("Cannot open output file!")
if not oegraphsim.OEAreCompatibleDatabases(moldb, fpdb):
oechem.OEThrow.Fatal("Databases are not compatible!")
oechem.OEThrow.Info("%5.2f sec to initialize databases" % timer.Elapsed())
fptype = fpdb.GetFPTypeBase()
oechem.OEThrow.Info("Using fingerprint type %s" % fptype.GetFPTypeString())
opts = oegraphsim.OEFPDatabaseOptions()
oegraphsim.OESetupFPDatabaseOptions(opts, itf)
# search fingerprint database
timer.Start()
scores = fpdb.GetSortedScores(query, opts)
oechem.OEThrow.Info("%5.2f sec to search %d fingerprints %s" %
(timer.Elapsed(), nrfps, memtypestr))
timer.Start()
nrhits = 0
hit = oechem.OEGraphMol()
for si in scores:
if moldb.GetMolecule(hit, si.GetIdx()):
nrhits += 1
oechem.OESetSDData(hit, "Similarity score", "%.2f" % si.GetScore())
oechem.OEWriteMolecule(ofs, hit)
oechem.OEThrow.Info("%5.2f sec to write %d hits" %
(timer.Elapsed(), nrhits))
return 0
def main(argv=[__name__]):
if len(argv) < 3:
oechem.OEThrow.Usage("%s <database> [<queries> ... ]" % argv[0])
dbname = argv[1]
# read in database
ifs = oechem.oemolistream()
if not ifs.open(dbname):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
print("Opening database file %s ..." % dbname)
timer = oechem.OEWallTimer()
dbase = oefastrocs.OEShapeDatabase()
moldb = oechem.OEMolDatabase()
if not moldb.Open(ifs):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
dots = oechem.OEThreadedDots(10000, 200, "conformers")
if not dbase.Open(moldb, dots):
oechem.OEThrow.Fatal("Unable to initialize OEShapeDatabase on '%s'" %
dbname)
dots.Total()
print("%s seconds to load database" % timer.Elapsed())
opts = oefastrocs.OEShapeDatabaseOptions()
opts.SetSimFunc(oefastrocs.OEShapeSimFuncType_Tversky)
numHits = moldb.NumMols()
opts.SetLimit(numHits)
for qfname in argv[2:]:
# read in query
qfs = oechem.oemolistream()
if not qfs.open(qfname):
oechem.OEThrow.Fatal("Unable to open '%s'" % argv[1])
query = oechem.OEGraphMol()
if not oechem.OEReadMolecule(qfs, query):
oechem.OEThrow.Fatal("Unable to read query from '%s'" % argv[1])
ext = oechem.OEGetFileExtension(qfname)
base = qfname[:-(len(ext) + 1)]
# write out everthing to a similary named file
ofs = oechem.oemolostream()
ofname = base + "_results." + ext
if not ofs.open(ofname):
oechem.OEThrow.Fatal("Unable to open '%s'" % argv[4])
print("Searching for %s" % qfname)
for score in dbase.GetSortedScores(query, opts):
dbmol = oechem.OEMol()
molidx = score.GetMolIdx()
if not moldb.GetMolecule(dbmol, molidx):
print("Unable to retrieve molecule '%u' from the database" %
molidx)
continue
mol = oechem.OEGraphMol(
dbmol.GetConf(oechem.OEHasConfIdx(score.GetConfIdx())))
oechem.OESetSDData(mol, "ShapeTversky",
"%.4f" % score.GetShapeTversky())
oechem.OESetSDData(mol, "ColorTversky",
"%.4f" % score.GetColorTversky())
oechem.OESetSDData(mol, "TverskyCombo",
"%.4f" % score.GetTverskyCombo())
score.Transform(mol)
oechem.OEWriteMolecule(ofs, mol)
print("Wrote results to %s" % ofname)
return 0
def main(argv=[__name__]):
itf = oechem.OEInterface()
oechem.OEConfigure(itf, InterfaceData)
defopts = oegraphsim.OEFPDatabaseOptions(10,
oegraphsim.OESimMeasure_Tanimoto)
oegraphsim.OEConfigureFPDatabaseOptions(itf, defopts)
oegraphsim.OEConfigureFingerPrint(
itf, oegraphsim.OEGetFPType(oegraphsim.OEFPType_Tree))
if not oechem.OEParseCommandLine(itf, argv):
return 0
qfname = itf.GetString("-query")
mfname = itf.GetString("-molfname")
ofname = itf.GetString("-out")
# initialize databases
timer = oechem.OEWallTimer()
timer.Start()
ifs = oechem.oemolistream()
if not ifs.open(qfname):
oechem.OEThrow.Fatal("Cannot open input file!")
query = oechem.OEGraphMol()
if not oechem.OEReadMolecule(ifs, query):
oechem.OEThrow.Fatal("Cannot read query molecule!")
moldb = oechem.OEMolDatabase()
if not moldb.Open(mfname):
oechem.OEThrow.Fatal("Cannot open molecule database!")
ofs = oechem.oemolostream()
if not ofs.open(ofname):
oechem.OEThrow.Fatal("Cannot open output file!")
fptype = oegraphsim.OESetupFingerPrint(itf)
oechem.OEThrow.Info("Using fingerprint type %s" % fptype.GetFPTypeString())
fpdb = oegraphsim.OEFPDatabase(fptype)
emptyfp = oegraphsim.OEFingerPrint()
emptyfp.SetFPTypeBase(fptype)
nrmols = moldb.GetMaxMolIdx()
mol = oechem.OEGraphMol()
for idx in range(0, nrmols):
if moldb.GetMolecule(mol, idx):
fpdb.AddFP(mol)
else:
fpdb.AddFP(emptyfp)
nrfps = fpdb.NumFingerPrints()
oechem.OEThrow.Info("%5.2f sec to initialize databases" % timer.Elapsed())
opts = oegraphsim.OEFPDatabaseOptions()
oegraphsim.OESetupFPDatabaseOptions(opts, itf)
# search fingerprint database
timer.Start()
scores = fpdb.GetSortedScores(query, opts)
oechem.OEThrow.Info("%5.2f sec to search %d fingerprints" %
(timer.Elapsed(), nrfps))
timer.Start()
hit = oechem.OEGraphMol()
for si in scores:
if moldb.GetMolecule(hit, si.GetIdx()):
oechem.OEWriteMolecule(ofs, hit)
oechem.OEThrow.Info("%5.2f sec to write %d hits" %
(timer.Elapsed(), opts.GetLimit()))
return 0
def main(argv=[__name__]):
if len(argv) < 3:
oechem.OEThrow.Usage("%s <database> [<queries> ... ]" % argv[0])
return 0
# check system
if not oefastrocs.OEFastROCSIsGPUReady():
oechem.OEThrow.Info("No supported GPU available!")
return 0
# read in database
dbname = argv[1]
print("Opening database file %s ..." % dbname)
dbase = oefastrocs.OEShapeDatabase()
moldb = oechem.OEMolDatabase()
if not moldb.Open(dbname):
oechem.OEThrow.Fatal("Unable to open '%s'" % dbname)
dots = oechem.OEThreadedDots(10000, 200, "conformers")
if not dbase.Open(moldb, dots):
oechem.OEThrow.Fatal("Unable to initialize OEShapeDatabase on '%s'" %
dbname)
# customize search options
opts = oefastrocs.OEShapeDatabaseOptions()
opts.SetInitialOrientation(
oefastrocs.OEFastROCSOrientation_UserInertialStarts)
opts.SetLimit(5)
for qfname in argv[2:]:
# read in query
qfs = oechem.oemolistream()
if not qfs.open(qfname):
oechem.OEThrow.Fatal("Unable to open '%s'" % qfname)
query = oechem.OEGraphMol()
if not oechem.OEReadMolecule(qfs, query):
oechem.OEThrow.Fatal("Unable to read query from '%s'" % qfname)
ext = oechem.OEGetFileExtension(qfname)
base = qfname[:-(len(ext) + 1)]
# write out everthing to a similary named file
ofs = oechem.oemolostream()
ofname = base + "_user_results." + ext
if not ofs.open(ofname):
oechem.OEThrow.Fatal("Unable to open '%s'" % argv[4])
oechem.OEWriteMolecule(ofs, query)
startsCoords = oechem.OEFloatVector()
atomIdx = 1
xyz = query.GetCoords()[atomIdx]
for x in xyz:
startsCoords.append(x)
if not len(startsCoords) % 3 == 0:
oechem.OEThrow.Fatal(
"Something went wrong whilst reading in user-starts coordinates"
)
opts.SetUserStarts(oechem.OEFloatVector(startsCoords),
int(len(startsCoords) / 3))
opts.SetMaxOverlays(opts.GetNumInertialStarts() *
opts.GetNumUserStarts())
if opts.GetInitialOrientation(
) == oefastrocs.OEFastROCSOrientation_UserInertialStarts:
numStarts = opts.GetNumUserStarts()
print("This example will use %u starts" % numStarts)
print("Searching for %s" % qfname)
for score in dbase.GetSortedScores(query, opts):
print("Score for mol %u(conf %u) %f shape %f color" %
(score.GetMolIdx(), score.GetConfIdx(),
score.GetShapeTanimoto(), score.GetColorTanimoto()))
dbmol = oechem.OEMol()
molidx = score.GetMolIdx()
if not moldb.GetMolecule(dbmol, molidx):
print("Unable to retrieve molecule '%u' from the database" %
molidx)
continue
mol = oechem.OEGraphMol(
dbmol.GetConf(oechem.OEHasConfIdx(score.GetConfIdx())))
oechem.OESetSDData(mol, "ShapeTanimoto",
"%.4f" % score.GetShapeTanimoto())
oechem.OESetSDData(mol, "ColorTanimoto",
"%.4f" % score.GetColorTanimoto())
oechem.OESetSDData(mol, "TanimotoCombo",
"%.4f" % score.GetTanimotoCombo())
score.Transform(mol)
oechem.OEWriteMolecule(ofs, mol)
print("Wrote results to %s" % ofname)
return 0
def get_molcount(path):
db = oechem.OEMolDatabase(path)
return db.NumMols()
