def main(argv=[__name__]):
if len(argv) != 3:
oechem.OEThrow.Usage("%s <infile> <outfile>" % argv[0])
ifs = oechem.oemolistream()
if not ifs.open(argv[1]):
oechem.OEThrow.Fatal("Unable to open %s for reading" % argv[1])
ofs = oechem.oemolostream()
if not ofs.open(argv[2]):
oechem.OEThrow.Fatal("Unable to open %s for writing" % argv[2])
for mol in ifs.GetOEMols():
oechem.OEDeleteEverythingExceptTheFirstLargestComponent(mol)
oechem.OEWriteMolecule(ofs, mol)
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData, argv)
# flag on command line indicates uncoloring option or not
bUncolor = itf.GetBool("-uncolor")
# input structure(s) to transform
ifsmols = oechem.oemolistream()
if not ifsmols.open(itf.GetString("-i")):
oechem.OEThrow.Fatal("Unable to open %s for reading" %
itf.GetString("-i"))
# save output structure(s) to this file
ofs = oechem.oemolostream()
if not ofs.open(itf.GetString("-o")):
oechem.OEThrow.Fatal("Unable to open %s for writing" %
itf.GetString("-o"))
if not oechem.OEIsSDDataFormat(ofs.GetFormat()):
oechem.OEThrow.Fatal("Unable to open %s for writing" %
itf.GetString("-o"))
irec = 0
ototal = 0
frag = oechem.OEGraphMol()
for mol in ifsmols.GetOEGraphMols():
irec += 1
oechem.OEDeleteEverythingExceptTheFirstLargestComponent(mol)
iter = oemedchem.OEGetBemisMurcko(mol)
if not iter.IsValid():
name = mol.GetTitle()
if not mol.GetTitle():
name = 'Record ' + str(irec)
oechem.OEThrow.Warning("%s: no perceived regions" % name)
continue
for bmregion in iter:
# create a fragment from the perceived region
oechem.OESubsetMol(frag, mol, bmregion, True)
if bUncolor:
# ignore 3D stereo parities
if (frag.GetDimension() == 3):
frag.SetDimension(0)
# uncolor the fragment
oechem.OEUncolorMol(frag)
smi = oechem.OEMolToSmiles(frag)
# annotate the input molecule with the role information
for role in bmregion.GetRoles():
oechem.OEAddSDData(mol, role.GetName(), smi)
ototal += 1
oechem.OEWriteMolecule(ofs, mol)
if not irec:
oechem.OEThrow.Fatal('No records in input structure file to perceive')
if not ototal:
oechem.OEThrow.Warning('No annotated structures generated')
print(
"Input molecules={0:d}, output annotated {1:s}molecules={2:d}".format(
irec, ("(uncolored) " if bUncolor else ""), ototal))
return 0
def MMPIndex(itf):
# output index file
mmpindexfile = itf.GetString("-output")
if not oemedchem.OEIsMatchedPairAnalyzerFileType(mmpindexfile):
oechem.OEThrow.Fatal("Output file is not a matched pair index type - \
needs .mmpidx extension: {}"
.format(mmpindexfile))
# create options class with defaults
mmpopts = oemedchem.OEMatchedPairAnalyzerOptions()
# set up options from command line
if not oemedchem.OESetupMatchedPairIndexOptions(mmpopts, itf):
oechem.OEThrow.Fatal("Error setting matched pair indexing options!")
# input structures to index
ifsindex = oechem.oemolistream()
if not ifsindex.open(itf.GetString("-input")):
oechem.OEThrow.Fatal("Unable to open {} for reading"
.format(itf.GetString("-input")))
# get requested verbosity setting
verbose = itf.GetBool("-verbose")
vverbose = itf.GetBool("-vverbose")
if vverbose:
verbose = vverbose
maxrec = max(itf.GetInt("-maxrec"), 0)
statusrec = itf.GetInt("-status")
if itf.GetBool("-exportcompress"):
if not mmpopts.SetOptions(mmpopts.GetOptions() |
oemedchem.OEMatchedPairOptions_ExportCompression):
oechem.OEThrow.Warning("Error enabling export compression!")
stripstereo = itf.GetBool("-stripstereo")
stripsalts = itf.GetBool("-stripsalts")
keepFields = []
if itf.HasString("-keepSD"):
for field in itf.GetStringList("-keepSD"):
keepFields.append(field)
if verbose:
oechem.OEThrow.Info('Retaining SD data fields: {}'.format(' '.join(keepFields)))
alldata = itf.GetBool("-allSD")
cleardata = itf.GetBool("-clearSD")
if keepFields:
if verbose and (alldata or cleardata):
oechem.OEThrow.Info("Option -keepSD overriding -allSD, -clearSD")
alldata = False
cleardata = False
elif cleardata:
alldata = False
if verbose:
oechem.OEThrow.Info("Forced clearing of all input SD data")
elif alldata:
if verbose:
oechem.OEThrow.Info("Retaining all input SD data")
cleardata = False
elif verbose:
oechem.OEThrow.Info("No SD data handling option specified, -allSD assumed")
if cleardata:
keepFields = ['-CLEARSD']
elif alldata or not keepFields:
keepFields = ['-ALLSD']
if verbose:
if not mmpopts.HasIndexableFragmentHeavyAtomRange():
oechem.OEThrow.Info("Indexing all fragments")
else:
oechem.OEThrow.Info("Limiting fragment cores to {0:.2f}-{1:.2f}% of input molecules"
.format(mmpopts.GetIndexableFragmentRangeMin(),
mmpopts.GetIndexableFragmentRangeMax()))
if statusrec:
oechem.OEThrow.Info("Status output after every {0} records".format(statusrec))
if maxrec:
oechem.OEThrow.Info("Indexing a maximum of {0} records".format(maxrec))
if itf.GetBool("-exportcompress"):
oechem.OEThrow.Info("Removing singleton index nodes from index")
if stripstereo:
oechem.OEThrow.Info("Stripping stereo")
if stripsalts:
oechem.OEThrow.Info("Stripping salts")
if itf.GetBool("-clearSD"):
oechem.OEThrow.Info("Clearing all input SD data")
elif alldata:
oechem.OEThrow.Info("Retaining all input SD data")
elif keepFields:
oechem.OEThrow.Info('Retaining floating point SD data fields: {}'
.format(''.join(keepFields)))
# create indexing engine
mmp = oemedchem.OEMatchedPairAnalyzer(mmpopts)
# interpret SD fields as floating point data
validdata = FilterSDData(keepFields, True)
# add molecules to be indexed
record = 0
unindexed = 0
for mol in ifsindex.GetOEGraphMols():
if not alldata:
# filter the input molecule SD data based on allowed fields
validdata.FilterMolData(mol)
if stripsalts:
oechem.OEDeleteEverythingExceptTheFirstLargestComponent(mol)
if stripstereo:
oechem.OEUncolorMol(mol,
(oechem.OEUncolorStrategy_RemoveAtomStereo |
oechem.OEUncolorStrategy_RemoveBondStereo |
oechem.OEUncolorStrategy_RemoveGroupStereo))
status = mmp.AddMol(mol, record)
if status != record:
unindexed += 1
if vverbose:
oechem.OEThrow.Info('Input structure not added to index, record=%d status=%s' %
(record, oemedchem.OEMatchedPairIndexStatusName(status)))
record += 1
if maxrec and record >= maxrec:
break
if statusrec and (record % statusrec) == 0:
oechem.OEThrow.Info("Records: {} Indexed: {} Unindexed: {}"
.format(record, (record - unindexed), unindexed))
if not mmp.NumMols():
oechem.OEThrow.Fatal('No records in index structure file')
if not mmp.NumMatchedPairs():
oechem.OEThrow.Fatal('No matched pairs found from indexing, ' +
'use -fragGe,-fragLe options to extend indexing range')
if not oemedchem.OEWriteMatchedPairAnalyzer(mmpindexfile, mmp):
oechem.OEThrow.Fatal('Error serializing MMP index: {}'
.format(mmpindexfile))
# return some status information
oechem.OEThrow.Info("Records: {}, Indexed: {}, matched pairs: {:,d}"
.format(record,
mmp.NumMols(),
mmp.NumMatchedPairs()))
return 0
def FindSimpleMatchedPairs(itf):
ims = oechem.oemolistream()
if not ims.open(itf.GetString("-input")):
oechem.OEThrow.Fatal("Unable to open %s for reading: " +
itf.GetString("-input"))
maxrecs = itf.GetInt("-maxrec")
oechem.OEThrow.SetLevel(oechem.OEErrorLevel_Warning)
# @ <SNIPPET-FINDSIMPLEMATCHEDPAIRS-EXAMPLE>
# create options class with defaults
mmpOpts = oemedchem.OEMatchedPairAnalyzerOptions()
# for 'simple' pairs, alter default indexing options
# - single cuts only, heavy atom substituents only (HMember indexing off)
mmpOpts.SetOptions(oemedchem.OEMatchedPairOptions_SingleCuts
| oemedchem.OEMatchedPairOptions_ComboCuts
| oemedchem.OEMatchedPairOptions_UniquesOnly)
# - limit substituent size to no more than 20% of input structure
mmpOpts.SetIndexableFragmentRange(80., 100.)
# create analyzer class with nondefault options
mmpAnalyzer = oemedchem.OEMatchedPairAnalyzer(mmpOpts)
# ignore common index status returns
sIgnoreStatus = 'FragmentRangeFilter,DuplicateStructure,'
sIgnoreStatus += 'FragmentationLimitFilter,HeavyAtomFilter'
# index the input structures
for recindex, mol in enumerate(ims.GetOEGraphMols(), start=1):
# consider only the largest input fragment
oechem.OEDeleteEverythingExceptTheFirstLargestComponent(mol)
# ignore stereochemistry
oechem.OEUncolorMol(mol, (oechem.OEUncolorStrategy_RemoveAtomStereo
| oechem.OEUncolorStrategy_RemoveBondStereo))
# explicitly provide a 1-based index to refer to indexed structures
# - to allow references back to external data elsewhere
status = mmpAnalyzer.AddMol(mol, recindex)
if status != recindex:
if not oemedchem.OEMatchedPairIndexStatusName(
status) in sIgnoreStatus:
oechem.OEThrow.Warning(
"{0}: molecule indexing error, status={1}".format(
recindex,
oemedchem.OEMatchedPairIndexStatusName(status)))
# if limiting input, quit after limit
if maxrecs and recindex >= maxrecs:
break
print("Index complete, matched pairs = {0}".format(
mmpAnalyzer.NumMatchedPairs()))
# specify how transforms are extracted (direction and allowed properties)
extractMode = (
oemedchem.OEMatchedPairTransformExtractMode_Sorted
| oemedchem.OEMatchedPairTransformExtractMode_NoSMARTS
| oemedchem.OEMatchedPairTransformExtractMode_AddMCSCorrespondence)
extractOptions = oemedchem.OEMatchedPairTransformExtractOptions()
# specify amount of chemical context at the site of the substituent change
# in the transform
extractOptions.SetContext(oemedchem.OEMatchedPairContext_Bond0)
extractOptions.SetOptions(extractMode)
# walk the transforms and print the matched pairs
xfmidx = 0
for mmpxform in oemedchem.OEMatchedPairGetTransforms(
mmpAnalyzer, extractOptions):
xfmidx += 1
print("{0:2} {1}".format(xfmidx, mmpxform.GetTransform()))
# dump matched molecular pairs and index identifiers
# (recindex from indexing loop above)
for mmppair in mmpxform.GetMatchedPairs():
print("\tmatched pair molecule indices=({0},{1})".format(
mmppair.FromIndex(), mmppair.ToIndex()))
# @ </SNIPPET-FINDSIMPLEMATCHEDPAIRS-EXAMPLE>
return True
def ChEMBLSolubilityUsage(itf):
ifs = oechem.oemolistream()
if not ifs.open(itf.GetString("-input")):
oechem.OEThrow.Fatal("Unable to open %s for reading: " +
itf.GetString("-input"))
ofs = oechem.oemolostream()
if not ofs.open(itf.GetString("-output")):
oechem.OEThrow.Fatal("Unable to open %s for writing: " +
ofs.GetString("-output"))
oechem.OEThrow.SetLevel(oechem.OEErrorLevel_Warning)
# @ <SNIPPET-OEAPPLYCHEMBLSOLUBILITY-EXAMPLE>
# number of bonds of chemistry context at site of change
# for the applied transforms
totalmols = 0
xformctxt = oemedchem.OEMatchedPairContext_Bond2
for molidx, mol in enumerate(ifs.GetOEGraphMols(), start=1):
# consider only the largest input fragment
oechem.OEDeleteEverythingExceptTheFirstLargestComponent(mol)
smolcnt = 0
# only consider solubility transforms having at least 5 matched pairs
for solMol in oemedchem.OEApplyChEMBL24SolubilityTransforms(
mol, xformctxt, 5):
# compute net change in solubility from MMP data
deltasol = []
if oechem.OEHasSDData(solMol, "OEMMP_normalized_value (uM)"):
for sditem in oechem.OEGetSDData(
solMol, "OEMMP_normalized_value (uM)").split('\n'):
# fromIndex,toIndex,fromValue,toValue
sdvalues = sditem.split(',')
if not sdvalues[2] or not sdvalues[3]:
continue
deltasol.append(float(sdvalues[3]) - float(sdvalues[2]))
if not len(deltasol):
continue
avgsol = deltasol[0]
if len(deltasol) > 1:
avgsol = average(deltasol)
# reject examples with net decrease in solubility
if avgsol < 0.0:
continue
sdev = stddev(deltasol)
# annotate with average,stddev,num
oechem.OEAddSDData(
solMol, "OEMMP_average_delta_normalized_value",
"{0:.1F},{1:.2F},{2}".format(avgsol, sdev, len(deltasol)))
# export solubility transformed molecule with SDData annotations
if oechem.OEWriteMolecule(
ofs, solMol) == oechem.OEWriteMolReturnCode_Success:
smolcnt += 1
oechem.OEThrow.Info("{0}: Exported molecule count, {1}".format(
molidx, smolcnt))
totalmols += smolcnt
# @ </SNIPPET-OEAPPLYCHEMBLSOLUBILITY-EXAMPLE>
print("Exported molecule count = {0}".format(totalmols))
return True