def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
oechem.OEConfigureSplitMolComplexOptions(itf)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
if itf.GetBool("-verbose"):
oechem.OEThrow.SetLevel(oechem.OEErrorLevel_Verbose)
iname = itf.GetString("-in")
oname = itf.GetString("-out")
ims = oechem.oemolistream()
if not itf.GetUnsignedInt("-modelnum") == 1:
ims.SetFlavor(
oechem.OEFormat_PDB,
oechem.OEGetDefaultIFlavor(oechem.OEFormat_PDB)
& ~oechem.OEIFlavor_PDB_ENDM)
if not ims.open(iname):
oechem.OEThrow.Fatal("Cannot open input file!")
oms = oechem.oemolostream()
if not oms.open(oname):
oechem.OEThrow.Fatal("Cannot open output file!")
mol = oechem.OEGraphMol()
if not oechem.OEReadMolecule(ims, mol):
oechem.OEThrow.Fatal("Cannot read input file!")
if itf.GetBool("-verbose"):
oechem.OEThrow.SetLevel(oechem.OEErrorLevel_Verbose)
opt = oechem.OESplitMolComplexOptions()
oechem.OESetupSplitMolComplexOptions(opt, itf)
# @ <SNIPPET-SPLIT-MOL-COMPLEX-LOWLEVEL-FRAGMENT>
# for input molecule mol and options opt ...
frags = oechem.OEAtomBondSetVector()
if oechem.OEGetMolComplexFragments(frags, mol, opt):
# ...
# @ </SNIPPET-SPLIT-MOL-COMPLEX-LOWLEVEL-FRAGMENT>
oechem.OEThrow.Verbose("Able to fragment mol complex from %s" % iname)
else:
oechem.OEThrow.Fatal("Unable to fragment mol complex from %s" % iname)
# @ <SNIPPET-SPLIT-MOL-COMPLEX-LOWLEVEL-COMBINE>
# for options opt and OEAtomBondSetVector frags produced earlier ...
numSites = oechem.OECountMolComplexSites(frags)
oechem.OEThrow.Verbose("sites %d" % numSites)
lig = oechem.OEGraphMol()
ligfilter = opt.GetLigandFilter()
if not oechem.OECombineMolComplexFragments(lig, frags, opt, ligfilter):
oechem.OEThrow.Warning("Unable to combine ligand frags from %s" %
mol.GetTitle())
protComplex = oechem.OEGraphMol()
p = opt.GetProteinFilter()
w = opt.GetWaterFilter()
if not oechem.OECombineMolComplexFragments(protComplex, frags, opt,
oechem.OEOrRoleSet(p, w)):
oechem.OEThrow.Warning("Unable to combine complex frags from %s" %
mol.GetTitle())
# @ </SNIPPET-SPLIT-MOL-COMPLEX-LOWLEVEL-COMBINE>
if not lig.NumAtoms() == 0:
oechem.OEThrow.Verbose(" lig %s" % lig.GetTitle())
oechem.OEWriteMolecule(oms, lig)
if not protComplex.NumAtoms() == 0:
oechem.OEThrow.Verbose(" prot %s" % protComplex.GetTitle())
oechem.OEWriteMolecule(oms, protComplex)
oms.close()
def split(system, ligand_res_name='LIG'):
"""
This function splits the passed molecule in components and tracks the
mapping between the original molecule and the split components. The
mapping is created as separated atom component index sets.
Parameters:
-----------
system: OEMol
The system to split in components. The components are:
the protein atoms,
the protein carbon alpha atoms
the water atoms,
the ion atoms,
the cofactor atoms
Returns:
--------
dic_set: python dictionary
The sysetm is splitted in a dictionary with token words as keys
and for value the related atom set. The token keywords are:
protein,
ca_protein,
ligand,
water,
ions,
cofactors,
system
"""
# Define Empty sets
lig_set = set()
prot_set = set()
ca_prot_set = set()
wat_set = set()
excp_set = set()
ion_set = set()
# cofactor_set = set()
# system_set = set()
# Atom Bond Set vector used to contains the whole system
frags = oechem.OEAtomBondSetVector()
# Define Options for the Filter
opt = oechem.OESplitMolComplexOptions()
# The protein filter is set to avoid that multiple
# chains are separated during the splitting and peptide
# molecules are recognized as ligands
pf = oechem.OEMolComplexFilterFactory(
oechem.OEMolComplexFilterCategory_Protein)
peptide = oechem.OEMolComplexFilterFactory(
oechem.OEMolComplexFilterCategory_Peptide)
protein_filter = oechem.OEOrRoleSet(pf, peptide)
opt.SetProteinFilter(protein_filter)
# The ligand filter is set to recognize just the ligand
lf = oechem.OEMolComplexFilterFactory(
oechem.OEMolComplexFilterCategory_Ligand)
not_protein_filter = oechem.OENotRoleSet(protein_filter)
ligand_filter = oechem.OEAndRoleSet(lf, not_protein_filter)
opt.SetLigandFilter(ligand_filter)
# The water filter is set to recognize just water molecules
wf = oechem.OEMolComplexFilterFactory(
oechem.OEMolComplexFilterCategory_Water)
opt.SetWaterFilter(wf)
# Set Category
cat = oechem.OEMolComplexCategorizer()
cat.AddLigandName(ligand_res_name)
opt.SetCategorizer(cat)
# Define the system fragments
if not oechem.OEGetMolComplexFragments(frags, system, opt):
raise ValueError('Unable to generate the system fragments')
# Set empty OEMol containers
prot = oechem.OEMol()
lig = oechem.OEMol()
wat = oechem.OEMol()
excp = oechem.OEMol()
# Split the protein from the system
atommap = oechem.OEAtomArray(system.GetMaxAtomIdx())
if not oechem.OECombineMolComplexFragments(
prot, frags, opt, opt.GetProteinFilter(), atommap):
raise ValueError('Unable to split the Protein')
# Populate the protein set and the protein carbon alpha set
pred = oechem.OEIsAlphaCarbon()
for sys_at in system.GetAtoms():
sys_idx = sys_at.GetIdx()
at_idx = atommap[sys_idx]
if at_idx:
prot_set.add(sys_idx)
at = system.GetAtom(oechem.OEHasAtomIdx(sys_idx))
if pred(at):
ca_prot_set.add(sys_idx)
# print(sys_idx, '->', at_idx)
# Split the ligand from the system
atommap = oechem.OEAtomArray(system.GetMaxAtomIdx())
if not oechem.OECombineMolComplexFragments(
lig, frags, opt, opt.GetLigandFilter(), atommap):
raise ValueError('Unable to split the Ligand')
# Populate the ligand set
for sys_at in system.GetAtoms():
sys_idx = sys_at.GetIdx()
at_idx = atommap[sys_idx]
if at_idx:
lig_set.add(sys_idx)
# print(sys_idx, '->', at_idx)
# Split the water from the system
atommap = oechem.OEAtomArray(system.GetMaxAtomIdx())
if not oechem.OECombineMolComplexFragments(
wat, frags, opt, opt.GetWaterFilter(), atommap):
raise ValueError('Unable to split the Water')
# Populate the water set
for sys_at in system.GetAtoms():
sys_idx = sys_at.GetIdx()
at_idx = atommap[sys_idx]
if at_idx:
wat_set.add(sys_idx)
# print(sys_idx, '->', at_idx)
# Split the excipients from the system
atommap = oechem.OEAtomArray(system.GetMaxAtomIdx())
if not oechem.OECombineMolComplexFragments(
excp, frags, opt, opt.GetOtherFilter(), atommap):
raise ValueError('Unable to split the Excipients')
# Populate the excipient set
for sys_at in system.GetAtoms():
sys_idx = sys_at.GetIdx()
at_idx = atommap[sys_idx]
if at_idx:
excp_set.add(sys_idx)
# print(sys_idx, '->', at_idx)
# Create the ions set
for exc_idx in excp_set:
atom = system.GetAtom(oechem.OEHasAtomIdx(exc_idx))
if atom.GetDegree() == 0:
ion_set.add(exc_idx)
# Create the cofactor set
cofactor_set = excp_set - ion_set
# Create the system set
system_set = prot_set | lig_set | excp_set | wat_set
if len(system_set) != system.NumAtoms():
raise ValueError("The total system atom number {} is different "
"from its set representation {}".format(
system.NumAtoms(), system_set))
# The dictionary is used to link the token keywords to the created molecule sets
dic_set = {
'ligand': lig_set,
'protein': prot_set,
'ca_protein': ca_prot_set,
'water': wat_set,
'ions': ion_set,
'cofactors': cofactor_set,
'system': system_set
}
return dic_set
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
oechem.OEConfigureSplitMolComplexOptions(itf)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
if itf.GetBool("-verbose"):
oechem.OEThrow.SetLevel(oechem.OEErrorLevel_Verbose)
iname = itf.GetString("-in")
oname = itf.GetString("-out")
ims = oechem.oemolistream()
if not itf.GetUnsignedInt("-modelnum") == 1:
ims.SetFlavor(
oechem.OEFormat_PDB,
oechem.OEGetDefaultIFlavor(oechem.OEFormat_PDB)
& ~oechem.OEIFlavor_PDB_ENDM)
if not ims.open(iname):
oechem.OEThrow.Fatal("Cannot open input file!")
oms = oechem.oemolostream()
if not oms.open(oname):
oechem.OEThrow.Fatal("Cannot open output file!")
inmol = oechem.OEGraphMol()
if not oechem.OEReadMolecule(ims, inmol):
oechem.OEThrow.Fatal("Cannot read input file!")
opts = oechem.OESplitMolComplexOptions()
oechem.OESetupSplitMolComplexOptions(opts, itf)
frags = oechem.OEAtomBondSetVector()
if not oechem.OEGetMolComplexFragments(frags, inmol, opts):
oechem.OEThrow.Fatal("Unable to split mol complex from %s" % iname)
numSites = oechem.OECountMolComplexSites(frags)
if itf.GetBool("-verbose"):
oechem.OEThrow.SetLevel(oechem.OEErrorLevel_Verbose)
oechem.OEThrow.Verbose("sites %d" % numSites)
lig = oechem.OEGraphMol()
prot = oechem.OEGraphMol()
wat = oechem.OEGraphMol()
other = oechem.OEGraphMol()
if not oechem.OECombineMolComplexFragments(lig, frags, opts,
opts.GetLigandFilter()):
oechem.OEThrow.Fatal("Unable to split ligand from %s" % iname)
if not oechem.OECombineMolComplexFragments(prot, frags, opts,
opts.GetProteinFilter()):
oechem.OEThrow.Fatal("Unable to split protein complex from %s" % iname)
if not oechem.OECombineMolComplexFragments(wat, frags, opts,
opts.GetWaterFilter()):
oechem.OEThrow.Fatal("Unable to split waters from %s" % iname)
if not oechem.OECombineMolComplexFragments(other, frags, opts,
opts.GetOtherFilter()):
oechem.OEThrow.Fatal("Unable to split other mols from %s" % iname)
if not lig.NumAtoms() == 0:
oechem.OEThrow.Verbose(" lig %s" % lig.GetTitle())
oechem.OEWriteMolecule(oms, lig)
if not prot.NumAtoms() == 0:
oechem.OEThrow.Verbose(" prot %s" % prot.GetTitle())
oechem.OEWriteMolecule(oms, prot)
if not wat.NumAtoms() == 0:
oechem.OEThrow.Verbose(" wat %s" % wat.GetTitle())
oechem.OEWriteMolecule(oms, wat)
if not other.NumAtoms() == 0:
oechem.OEThrow.Verbose("other %s" % other.GetTitle())
oechem.OEWriteMolecule(oms, other)
oms.close()