def __searchSubStructure(self,
oeQueryMol,
idxList=None,
reverseFlag=False,
matchOpts="graph-relaxed"):
"""Perform a graph match for the input query molecule on the binary
database of molecules. The search optionally restricted to the input index
list. The sense of the search may be optionally reversed.
Args:
oeQueryMol (object): query molecule OeGraphMol or OeQmol
idxList ([type], optional): [description]. Defaults to None.
reverseFlag (bool, optional): [description]. Defaults to False.
matchOpts (str, optional): graph match criteria type (graph-strict|graph-relaxed|graph-relaxed-stereo). Defaults to "graph-relaxed".
Returns:
[type]: [description]
"""
hL = []
retStatus = True
startTime = time.time()
try:
# logger.info("Query mol type %r", type(oeQueryMol))
atomexpr, bondexpr = OeCommonUtils.getAtomBondExprOpts(matchOpts)
ss = oechem.OESubSearch(oeQueryMol, atomexpr, bondexpr)
if not ss.IsValid():
retStatus = False
logger.error("Unable to initialize substructure search!")
return retStatus, hL
#
searchType = "exhaustive-substructure"
if idxList:
searchType = "prefilterd-substructure"
idxIt = idxList if idxList else range(
self.__oeMolDb.GetMaxMolIdx())
for idx in idxIt:
mol = oechem.OEGraphMol()
ccId = self.__oeMolDb.GetTitle(idx)
if not self.__oeMolDb.GetMolecule(mol, idx):
logger.error("Unable to read molecule %r at index %r",
ccId, idx)
continue
oechem.OEPrepareSearch(mol, ss)
if ss.SingleMatch(mol) != reverseFlag:
score = float(oeQueryMol.NumAtoms()) / float(
oeQueryMol.NumAtoms())
hL.append(
MatchResults(ccId=ccId,
searchType=searchType,
matchOpts=matchOpts,
fpScore=score))
retStatus = True
except Exception as e:
retStatus = False
logger.exception("Failing with %s", str(e))
#
logger.info("Substructure search returns %d (%.4f seconds)", len(hL),
time.time() - startTime)
return retStatus, hL
def keep_molecule(mol, remove_smirks = list()):
"""
Determines if the molecule will be stored.
Parameters
----------
mol - OEMol
remove_smirks - list of SMIRKS strings you don't want in your molecules
Returns
-------
boolean - True (molecule meets the requirements below)
- has no metal atoms
- no more than 200 heavy atoms
- has none of the SMIRKS in remove_smirks list
- molecule has appropriate valency
"""
# Check number of metal atoms
if oechem.OECount(mol, oechem.OEIsMetal()) > 0:
return False
# Check number of heavy atoms
if oechem.OECount(mol, oechem.OEIsHeavy()) > 200:
return False
# Check for patterns in remove smirks list
for smirks in remove_smirks:
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smirks):
continue
ss = oechem.OESubSearch(qmol)
matches = [match for match in ss.Match(mol, False)]
if len(matches) > 0:
return False
# check valency
return check_valence(mol)
def get_covalent_warhead_atom(molecule, covalent_warhead_type):
"""
Get tagged atom index in provided tagged SMARTS string, or None if no match found.
Parameters
----------
molecule : openeye.oechem.OEMol
The molecule to search
covalent_warhead : str
Covalent warhead name
Returns
-------
index : int or None
The atom index in molecule of the covalent atom, or None if SMARTS does not match
"""
smarts = covalent_warhead_smarts[covalent_warhead_type]
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smarts):
raise ValueError(f"Error parsing SMARTS '{smarts}'")
substructure_search = oechem.OESubSearch(qmol)
substructure_search.SetMaxMatches(1)
matches = list()
for match in substructure_search.Match(molecule):
# Compile list of atom indices that match the pattern tags
for matched_atom in match.GetAtoms():
if(matched_atom.pattern.GetMapIdx()==1):
return matched_atom.target.GetIdx()
return None
def _find_torsions_from_smarts(molecule, smarts):
"""
Do a substrcutre search on provided SMARTS to find torsions that match the SAMRTS
Parameters
----------
molecule: OEMol
molecule to search on
smarts: str
SMARTS pattern to search for
Returns
-------
tors: list
list of torsions that match the SMARTS string
"""
from openeye import oechem
#ToDO use MDL aromaticity model
qmol=oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smarts):
utils.logger().warning('OEParseSmarts failed')
ss = oechem.OESubSearch(qmol)
tors = []
oechem.OEPrepareSearch(molecule, ss)
unique = True
for match in ss.Match(molecule, unique):
tor = []
for ma in match.GetAtoms():
tor.append(ma.target)
tors.append(tor)
return tors
def GetCommonFragments(mollist, frags,
atomexpr=oechem.OEExprOpts_DefaultAtoms,
bondexpr=oechem.OEExprOpts_DefaultBonds):
corefrags = []
from rich.progress import track
#for frag in track(frags, description='Finding common fragments'):
for frag in frags:
ss = oechem.OESubSearch(frag, atomexpr, bondexpr)
if not ss.IsValid():
print('Is not valid')
continue
validcore = True
for mol in mollist:
oechem.OEPrepareSearch(mol, ss)
validcore = ss.SingleMatch(mol)
if not validcore:
break
if validcore:
corefrags.append(frag)
return corefrags
def __setupSubStructure(self, refmol):
"""Internal initialization for a substructure comparison."""
#
atomexpr, bondexpr = OeCommonUtils.getAtomBondExprOpts(self.__searchType)
self.__ss = oechem.OESubSearch(refmol, atomexpr, bondexpr)
if self.__verbose:
logger.info("Initialize SS (%r)", self.__searchType)
def keep_molecule(mol, max_heavy_atoms = 100,
remove_smirks = list(), max_metals = 0, elements = [], check_type = None):
if oechem.OECount(mol, oechem.OEIsMetal()) > max_metals:
return False
if oechem.OECount(mol, oechem.OEIsHeavy()) > max_heavy_atoms:
return False
# Remove very small molecules that are not interesting
if oechem.OECount(mol, oechem.OEIsHeavy()) < 5:
return False
for smirks in remove_smirks:
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smirks):
continue
ss = oechem.OESubSearch(qmol)
matches = [match for match in ss.Match(mol, False)]
if len(matches) > 0:
return False
if elements != None:
elements_list = read_Elements(elements)
if not check_element(mol, elements_list):
return False
if check_type != None:
types = check_type.split(",")
if not check_atomtype(mol, types):
return False
return check_valence(mol)
def OEAddLabel_OEMatch(disp):
subs = oechem.OESubSearch("a1aaaaa1")
unique = True
highlightstyle = oedepict.OEHighlightByBallAndStick(oechem.OELightGreen)
for match in subs.Match(disp.GetMolecule(), unique):
oedepict.OEAddHighlighting(disp, highlightstyle, match)
label = oedepict.OEHighlightLabel("aromatic", oechem.OELightGreen)
oedepict.OEAddLabel(disp, label, match)
def OEAddHighlighting_OEMatch(adisp):
ligand = adisp.GetDisplayedLigand()
subs = oechem.OESubSearch("a1aaaaa1")
colors = oechem.OEGetVividColors()
unique = True
for match, color in zip(subs.Match(ligand, unique), colors):
highlight = oedepict.OEHighlightByLasso(color)
highlight.SetConsiderAtomLabelBoundingBox(True)
oegrapheme.OEAddLigandHighlighting(adisp, highlight, match)
def substructure_search(smi, target_smiles='[!C;!c]-[F,Cl,Br]'):
from openeye import oechem
ss = oechem.OESubSearch(target_smiles)
oemol = oechem.OEGraphMol()
oechem.OESmilesToMol(oemol, smi)
oechem.OEAddExplicitHydrogens(oemol)
oechem.OEPrepareSearch(oemol, ss)
return ss.SingleMatch(oemol)
def match_subset(pattern: oechem.OEMol, target: oechem.OEMol):
"""Check if target is a subset of pattern."""
# Atoms are equal if they have same atomic number (so explicit Hydrogens are needed as well for a match)
atomexpr = oechem.OEExprOpts_AtomicNumber
# single or double bonds are considered identical (resonance,chirality fix)
bondexpr = oechem.OEExprOpts_EqSingleDouble
ss = oechem.OESubSearch(pattern, atomexpr, bondexpr)
oechem.OEPrepareSearch(target, ss)
return ss.SingleMatch(target)
def find_ortho_substituents(frag_smi):
from openeye import oechem
oemol = oechem.OEGraphMol()
oechem.OESmilesToMol(oemol, frag_smi)
oechem.OEAddExplicitHydrogens(oemol)
ortho = '[!#1]~!@[*;r]~;@[*;r]~!@[!#1]'
ss = oechem.OESubSearch(ortho)
oechem.OEPrepareSearch(oemol, ss)
return ss.SingleMatch(oemol)
def main(argv=[__name__]):
itf = oechem.OEInterface(Interface, argv)
if not ((itf.HasString("-smarts1") and itf.HasString("-smarts2")) ^
(itf.HasString("-atom1") and itf.HasString("-atom2"))):
oechem.OEThrow.Fatal(
"-smarts1 and -smarts2 or -atom1 and -atom2 must be set")
ifs = oechem.oemolistream()
if not ifs.open(itf.GetString("-i")):
oechem.OEThrow.Fatal("Unable to open %s for reading" %
itf.GetString("-i").rstrip())
ofs = oechem.oemolostream()
if itf.HasString("-o"):
if not ofs.open(itf.GetString("-o")):
oechem.OEThrow.Fatal("Unable to open %s for writing" %
itf.GetString("-o"))
if itf.HasString("-smarts1") and itf.HasString("-smarts2"):
ss1 = oechem.OESubSearch()
smarts1 = itf.GetString("-smarts1")
if not ss1.Init(smarts1):
oechem.OEThrow.Fatal("Unable to parse SMARTS1: %s" %
smarts1.rstrip())
ss2 = oechem.OESubSearch()
smarts2 = itf.GetString("-smarts2")
if not ss2.Init(smarts2):
oechem.OEThrow.Fatal("Unable to parse SMARTS2: %s" %
smarts2.rstrip())
SmartsPathLength(ifs, ofs, itf, ss1, ss2)
else:
atom1 = itf.GetString("-atom1")
atom2 = itf.GetString("-atom2")
AtomPathLength(ifs, ofs, itf, atom1, atom2)
def __init__(self, molComplex, fprintDef=False):
self.molComplex = molComplex
self.fprint = []
self.fprintConsensus = []
if fprintDef == None:
self.fprintDef = "11111111111"
else:
self.fprintDef = fprintDef
#---------------#
# OESubSearches #
#---------------#
# Definition of a Hydrophobe atom
# Any atom (Carbon, or Sulfur, or Fluor, or Chloride, or Bromine,
# or Iodine)
self.ssHydrophobe = oechem.OESubSearch('[C,S,F,Cl,Br,I]')
# Definition of a H-Bond donor
# All atoms (Oxygen, or Nitrogen, or Sulfur, or Fluor)
# connected to (a Hydrogen)
self.ssDonor = oechem.OESubSearch('[O,N,S,F][H]')
# Definition of a H-Bond acceptor
# All atoms (Oxygen, or Nitrogen, or any negatively charged atom)
# AND (not a positively charged atom)
self.ssAcceptor = oechem.OESubSearch('[O,N,*-;!+]')
# Definition of a weak H-bond Acceptor
# Any 2 Aromatic atoms connected by aromatic bond,
# or 2 Aliphatic atoms connected by double bond,
# or 2 Aliphatic atoms connected by triple bond,
# self.ssWkAcceptor = OESubSearch("[a:a, A=A, A#A]")
self.ssWkAcceptor = oechem.OESubSearch('[A,a]=,#,:[A,a]')
# Definition of a weak H-bond Donor
# Any (Aromatic carbon atom, or Aliphatic carbon with 3 total bonds
# or Aliphatic carbon with 2 total bonds) connected to (a Hydrogen)
self.ssWkDonor = oechem.OESubSearch('[c,CX3,CX2][H]')
# Definition of a Cation :: Any negatively charged atom
self.ssCation = oechem.OESubSearch('[*+]')
# Definition of a Anion :: Any positively charged atom
self.ssAnion = oechem.OESubSearch('[*-]')
# Definition of a metal atom
self.ssMetal = oechem.OESubSearch('[Ca,Cd,Co,Cu,Fe,Mg,Mn,Ni,Zn]')
def main(argv=[__name__]):
if len(argv) != 5:
oechem.OEThrow.Usage("%s <refmol> <fitmol> <outfile> <smarts>" %
argv[0])
reffs = oechem.oemolistream()
if not reffs.open(argv[1]):
oechem.OEThrow.Fatal("Unable to open %s for reading" % argv[1])
if not oechem.OEIs3DFormat(reffs.GetFormat()):
oechem.OEThrow.Fatal("Invalid input format: need 3D coordinates")
refmol = oechem.OEGraphMol()
if not oechem.OEReadMolecule(reffs, refmol):
oechem.OEThrow.Fatal("Unable to read molecule in %s" % argv[1])
if not refmol.GetDimension() == 3:
oechem.OEThrow.Fatal("%s doesn't have 3D coordinates" %
refmol.GetTitle())
fitfs = oechem.oemolistream()
if not fitfs.open(argv[2]):
oechem.OEThrow.Fatal("Unable to open %s for reading" % argv[2])
if not oechem.OEIs3DFormat(fitfs.GetFormat()):
oechem.OEThrow.Fatal("Invalid input format: need 3D coordinates")
ofs = oechem.oemolostream()
if not ofs.open(argv[3]):
oechem.OEThrow.Fatal("Unable to open %s for writing" % argv[3])
if not oechem.OEIs3DFormat(ofs.GetFormat()):
oechem.OEThrow.Fatal("Invalid output format: need 3D coordinates")
oechem.OEWriteConstMolecule(ofs, refmol)
ss = oechem.OESubSearch()
if not ss.Init(argv[4]):
oechem.OEThrow.Fatal("Unable to parse SMARTS: %s" % argv[4])
oechem.OEPrepareSearch(refmol, ss)
if not ss.SingleMatch(refmol):
oechem.OEThrow.Fatal("SMARTS fails to match refmol")
for fitmol in fitfs.GetOEGraphMols():
if not fitmol.GetDimension() == 3:
oechem.OEThrow.Warning("%s doesn't have 3D coordinates" %
fitmol.GetTitle())
continue
oechem.OEPrepareSearch(fitmol, ss)
if not ss.SingleMatch(fitmol):
oechem.OEThrow.Warning("SMARTS fails to match fitmol %s" %
fitmol.GetTitle())
continue
SmartsAlign(refmol, fitmol, ss, ofs)
def _tag_fgroups(mol, fgroups_smarts=None):
"""
This function tags atoms and bonds of functional groups defined in fgroup_smarts. fgroup_smarts is a dictionary
that maps functional groups to their smarts pattern. It can be user generated or from yaml file.
Parameters
----------
mol: Openeye OEMolGraph
frgroups_smarts: dictionary of functional groups mapped to their smarts pattern.
Default is None. It uses 'fgroup_smarts.yaml'
Returns
-------
fgroup_tagged: dict
a dictionary that maps indexed functional groups to corresponding atom and bond indices in mol
"""
if not fgroups_smarts:
# Load yaml file
fn = resource_filename('fragmenter', os.path.join('data', 'fgroup_smarts.yml'))
f = open(fn, 'r')
fgroups_smarts = yaml.safe_load(f)
f.close()
fgroup_tagged = {}
for f_group in fgroups_smarts:
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, fgroups_smarts[f_group]):
print('OEParseSmarts failed')
ss = oechem.OESubSearch(qmol)
oechem.OEPrepareSearch(mol, ss)
for i, match in enumerate(ss.Match(mol, True)):
fgroup_atoms = set()
for ma in match.GetAtoms():
fgroup_atoms.add(ma.target.GetIdx())
tag = oechem.OEGetTag('fgroup')
ma.target.SetData(tag, '{}_{}'.format(f_group, str(i)))
fgroup_bonds = set()
for ma in match.GetBonds():
#if not ma.target.IsInRing():
fgroup_bonds.add(ma.target.GetIdx())
tag =oechem.OEGetTag('fgroup')
ma.target.SetData(tag, '{}_{}'.format(f_group, str(i)))
fgroup_tagged['{}_{}'.format(f_group, str(i))] = (fgroup_atoms, fgroup_bonds)
return fgroup_tagged
def get_atom_map(tagged_smiles, molecule=None):
"""
Returns a dictionary that maps tag on SMILES to atom index in molecule.
Parameters
----------
tagged_smiles: str
index-tagged explicit hydrogen SMILES string
molecule: OEMol
molecule to generate map for. If None, a new OEMol will be generated from the tagged SMILES, the map will map to
this molecule and it will be returned.
Returns
-------
atom_map: dict
a dictionary that maps tag to atom index {tag:idx}
molecule: OEMol
If a molecule was not provided, the generated molecule will be returned.
"""
if molecule is None:
molecule = openeye.smiles_to_oemol(tagged_smiles)
ss = oechem.OESubSearch(tagged_smiles)
oechem.OEPrepareSearch(molecule, ss)
ss.SetMaxMatches(1)
atom_map = {}
t1 = time.time()
matches = [m for m in ss.Match(molecule)]
t2 = time.time()
seconds = t2 - t1
logger().info("Substructure search took {} seconds".format(seconds))
if not matches:
logger().info("MCSS failed for {}, smiles: {}".format(
molecule.GetTitle(), tagged_smiles))
return False
for match in matches:
for ma in match.GetAtoms():
atom_map[ma.pattern.GetMapIdx()] = ma.target.GetIdx()
# sanity check
mol = oechem.OEGraphMol()
oechem.OESubsetMol(mol, match, True)
logger().info("Match SMILES: {}".format(oechem.OEMolToSmiles(mol)))
if molecule is None:
return molecule, atom_map
return atom_map
def __subStructureSearch(self, oeQueryMol, idxList, reverseFlag=False):
"""Perform a graph match for the input query molecule on the binary
database of molecules. The search optionally restricted to the input index
list. The sense of the search may be optionally reversed.
Args:
oeQueryMol (object): query molecule OeGraphMol or OeQmol
idxList ([type], optional): [description]. Defaults to None.
reverseFlag (bool, optional): [description]. Defaults to False.
matchOpts (str, optional): graph match criteria type (graph-strict|graph-relaxed|graph-relaxed-stereo). Defaults to "graph-relaxed".
Returns:
[type]: [description]
"""
hL = []
sL = []
retStatus = True
try:
ss = oechem.OESubSearch(oeQueryMol, self.__atomexpr,
self.__bondexpr)
if not ss.IsValid():
retStatus = False
logger.error("Unable to initialize substructure search!")
return retStatus, hL, sL
#
for idx in idxList:
mol = oechem.OEGraphMol()
ccId = self.__oeMolDb.GetTitle(idx)
if not self.__oeMolDb.GetMolecule(mol, idx):
logger.error("Unable to read molecule %r at index %r",
ccId, idx)
continue
oechem.OEPrepareSearch(mol, ss)
if ss.SingleMatch(mol) != reverseFlag:
score = float(oeQueryMol.NumAtoms()) / float(
mol.NumAtoms())
hL.append(idx)
sL.append((idx, score))
# logger.info("%s queryAtoms %d molAtoms %d score %.4f", ccId, oeQueryMol.NumAtoms(), mol.NumAtoms(), score)
retStatus = True
except Exception as e:
retStatus = False
logger.exception("Failing with %s", str(e))
#
return retStatus, hL, sL
def match_smirks(smirks: str,
oe_molecule: OEMol,
unique: bool = False) -> List[Dict[int, int]]:
"""Attempt to find the indices (optionally unique) of all atoms which
match a particular SMIRKS pattern.
Parameters
----------
smirks
The SMIRKS pattern to match.
oe_molecule
The molecule to match against.
unique
Whether to return back only unique matches.
Returns
-------
A list of all the matches where each match is stored as a dictionary of
the smirks indices and their corresponding matched atom indices.
"""
from openeye import oechem
query = oechem.OEQMol()
call_openeye(
oechem.OEParseSmarts,
query,
smirks,
exception_type=InvalidSmirksError,
exception_kwargs={"smirks": smirks},
)
substructure_search = oechem.OESubSearch(query)
substructure_search.SetMaxMatches(0)
matches = []
for match in substructure_search.Match(oe_molecule, unique):
matched_indices = {
atom_match.pattern.GetMapIdx() - 1: atom_match.target.GetIdx()
for atom_match in match.GetAtoms()
if atom_match.pattern.GetMapIdx() != 0
}
matches.append(matched_indices)
return matches
def __init__(self, infileName, tagname):
self.pattyTag = oechem.OEGetTag(tagname)
self.smartsList = []
ifs = open(infileName)
lines = ifs.readlines()
for line in lines:
# Strip trailing comments
index = line.find('%')
if index != -1:
line = line[0:index]
# Split into tokens.
toks = string.split(line)
if len(toks) == 2:
smarts, type = toks
pat = oechem.OESubSearch()
pat.Init(smarts)
pat.SetMaxMatches(0)
self.smartsList.append([pat, type, smarts])
def smirks_search(self, smirks):
matches = list()
ss = oechem.OESubSearch()
if not ss.Init(smirks):
# TODO: write custom exceptions?
raise ValueError("Error parsing SMIRKS %s" % smirks)
for match in ss.Match(self.mol, False):
d = dict()
for ma in match.GetAtoms():
smirks_idx = ma.pattern.GetMapIdx()
# if the map index is 0 then it isn't a "tagged" atom in the SMIRKS
if smirks_idx != 0:
d[smirks_idx] = self.get_atom_by_index(ma.target.GetIdx())
matches.append(d)
return matches
def fixAcid(self, mol):
"""
Dirty fix that adds a negative formal charge on both Oxygens of the
carboxylic acid from ASP and GLU residues. As the charge is attributed
to single atoms, some ionic interactions would be missed if proximity
is verified with the neutral O atom.
"""
# SMARTS definition for a carboxilic acid
carbox = oechem.OESubSearch("[CX3](=O)[O-]")
#print(mol)
# Look for pattern in ASP or GLU
for match in carbox.Match(mol):
for atom in match.GetTargetAtoms():
charge = atom.GetFormalCharge()
if charge == 0 and "O" in atom.GetName():
atom.SetFormalCharge(-1)
charge = atom.GetFormalCharge()
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
iname = itf.GetString("-in")
oname = itf.GetString("-out")
smarts = itf.GetString("-smarts")
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smarts):
oechem.OEThrow.Fatal("Invalid SMARTS: %s" % smarts)
oechem.OEGenerate2DCoordinates(qmol)
ss = oechem.OESubSearch(qmol)
if not ss.IsValid():
oechem.OEThrow.Fatal("Unable to initialize substructure search!")
ifs = oechem.oemolistream()
if not ifs.open(iname):
oechem.OEThrow.Fatal("Cannot open input molecule file!")
ofs = oechem.oemolostream()
if not ofs.open(oname):
oechem.OEThrow.Fatal("Cannot open output file!")
if not oechem.OEIs2DFormat(ofs.GetFormat()):
oechem.OEThrow.Fatal("Invalid output format for 2D coordinates")
for mol in ifs.GetOEGraphMols():
oechem.OEPrepareSearch(mol, ss)
alignres = oedepict.OEPrepareAlignedDepiction(mol, ss)
if not alignres.IsValid():
oechem.OEThrow.Warning(
"Substructure is not found in input molecule!")
oedepict.OEPrepareDepiction(mol)
oechem.OEWriteMolecule(ofs, mol)
return 0
def _check_nitro(molecule):
"""
Filter out nitro that is in ([NX3](=O)=O) form. OEGetReasonableTautomers generates this form.
Parameters
----------
molecule :
Returns
-------
"""
from openeye import oechem
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, '([NX3](=O)=O)'):
print('OEParseSmarts failed')
ss = oechem.OESubSearch(qmol)
oechem.OEPrepareSearch(molecule, ss)
matches = [m for m in ss.Match(molecule)]
return bool(matches)
def smirks_search(self, smirks):
cmol = oechem.OEMol(self.mol)
matches = list()
ss = oechem.OESubSearch()
if not ss.Init(smirks):
raise ValueError("Error parsing SMIRKS %s" % smirks)
# set maximum matches in substructure search to infinite (0 in API)
ss.SetMaxMatches(0)
for match in ss.Match(cmol, False):
d = dict()
for ma in match.GetAtoms():
smirks_idx = ma.pattern.GetMapIdx()
# if the map index is 0 then it isn't a "tagged" atom in the SMIRKS
if smirks_idx !=0:
d[smirks_idx] = self.get_atom_by_index(ma.target.GetIdx())
matches.append(d)
return matches
def align2d(file1, file2):
atomexpr = oechem.OEExprOpts_AtomicNumber | oechem.OEExprOpts_RingMember
bondexpr = oechem.OEExprOpts_RingMember
ifs1 = oechem.oemolistream(file1)
ifs2 = oechem.oemolistream(file2)
ifs1.SetConfTest(oechem.OEAbsCanonicalConfTest())
ifs2.SetConfTest(oechem.OEAbsCanonicalConfTest())
popts, dopts, report = prep_pdf_writer()
for mol1, mol2 in zip(ifs1.GetOEMols(), ifs2.GetOEMols()):
oechem.OESuppressHydrogens(mol1)
oechem.OESuppressHydrogens(mol2)
oechem.OEGenerate2DCoordinates(mol2)
ss = oechem.OESubSearch(mol2, atomexpr, bondexpr)
oechem.OEPrepareSearch(mol1, ss)
alignres = oedepict.OEPrepareAlignedDepiction(mol1, ss)
if not alignres.IsValid():
oechem.OEThrow.Error(
"Substructure is not found in input molecule!")
cell1 = report.NewCell()
cell2 = report.NewCell()
oedepict.OEPrepareDepiction(mol1, popts)
oedepict.OEPrepareDepiction(mol2, popts)
disp1 = oedepict.OE2DMolDisplay(mol1, dopts)
disp2 = oedepict.OE2DMolDisplay(mol2, dopts)
oedepict.OERenderMolecule(cell1, disp1)
oedepict.OERenderMolecule(cell2, disp2)
ofs = oechem.oeofstream()
if not ofs.open('output.pdf'):
oechem.OEThrow.Fatal("Cannot open output file!")
oedepict.OEWriteReport(ofs, "pdf", report)
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData, argv)
if not (itf.GetBool("-c") ^ itf.HasString("-o")):
oechem.OEThrow.Fatal("Counting (-c) or output (-o) must be \
specified and are mutually exclusive.")
ifs = oechem.oemolistream()
filename = itf.GetString("-i")
if not ifs.open(filename):
oechem.OEThrow.Fatal("Unable to open %s for reading" % filename)
ofs = oechem.oemolostream()
if not itf.GetBool("-c"):
filename = itf.GetString("-o")
if not ofs.open(filename):
oechem.OEThrow.Fatal("Unable to open %s for writing" % filename)
smarts = itf.GetString("-p")
ss = oechem.OESubSearch()
if not ss.Init(smarts):
oechem.OEThrow.Fatal("Unable to parse SMARTS: %s" % smarts)
SubSearch(itf, ss, ifs, ofs)
def smirks_search(self, smirks):
"""
Performs a substructure search on the molecule with the provided
SMIRKS pattern. Note - this function expects SMIRKS patterns with indexed atoms
that is with :n for at least some atoms.
Parameters
----------
smirks: str
SMIRKS pattern with indexed atoms (:n)
Returns
-------
matches: list of dictionaries
dictionary for each match with the form {smirks index: atom index}
"""
cmol = oechem.OEMol(self.mol)
matches = list()
ss = oechem.OESubSearch()
if not ss.Init(smirks):
raise ValueError("Error parsing SMIRKS %s" % smirks)
# set maximum matches in substructure search to infinite (0 in API)
ss.SetMaxMatches(0)
for match in ss.Match(cmol, False):
d = dict()
for ma in match.GetAtoms():
smirks_idx = ma.pattern.GetMapIdx()
# if the map index is 0 then it isn't a "tagged" atom in the SMIRKS
if smirks_idx != 0:
d[smirks_idx] = self.get_atom_by_index(ma.target.GetIdx())
matches.append(d)
return matches
def get_atom_map(molecule, mapped_smiles, strict=True):
"""
Map tag in mapped SMILES to atom idx using a substructure search
A substructure search finds chemically equivalent matches so if atoms are symmetrical, they can flip. The mapped
SMILES used for the pattern is first used to generate a molecule with the map indices, the order is canonicalized
and then it is used for the substructure search pattern. This ensures that symmetrical do not flip but there is no
guarantee that it won't happen.
Parameters
----------
molecule: oechem.OEMOl
Must have explicit hydrogen
mapped_smiles: str
explicit hydrogen SMILES with map indices on every atom
Returns
-------
atom_map: dict
{map_idx:atom_idx}
"""
# check that smiles has explicit hydrogen and map indices
mapped_mol = oechem.OEMol()
oechem.OESmilesToMol(mapped_mol, mapped_smiles)
if not has_atom_map(mapped_mol):
raise ValueError(
"Mapped SMILES must have map indices for all atoms and hydrogens")
# Check molecule for explicit hydrogen
if not has_explicit_hydrogen(molecule) and strict:
raise ValueError("Molecule must have explicit hydrogens")
# canonical order mapped mol to ensure atom map is always generated in the same order
canonical_order_atoms(mapped_mol)
aopts = oechem.OEExprOpts_DefaultAtoms
bopts = oechem.OEExprOpts_DefaultBonds
ss = oechem.OESubSearch(mapped_mol, aopts, bopts)
oechem.OEPrepareSearch(molecule, ss)
ss.SetMaxMatches(1)
atom_map = {}
matches = [m for m in ss.Match(molecule)]
if not matches:
raise RuntimeError("MCSS failed for {}, smiles: {}".format(
oechem.OEMolToSmiles(molecule), mapped_smiles))
for match in matches:
for ma in match.GetAtoms():
atom_map[ma.pattern.GetMapIdx()] = ma.target.GetIdx()
# sanity check
mol = oechem.OEGraphMol()
oechem.OESubsetMol(mol, match, True)
matched_smiles = mol_to_smiles(mol,
isomeric=False,
explicit_hydrogen=False,
mapped=False)
molcopy = oechem.OEMol(molecule)
smiles = mol_to_smiles(molcopy,
isomeric=False,
explicit_hydrogen=False,
mapped=False)
pattern_smiles = mol_to_smiles(mapped_mol,
isomeric=False,
explicit_hydrogen=False,
mapped=False)
if not matched_smiles == smiles == pattern_smiles:
raise RuntimeError(
"Matched molecule, input molecule and mapped SMILES are not the same "
)
return atom_map
# TERMS FOR USE OF SAMPLE CODE The software below ("Sample Code") is
# provided to current licensees or subscribers of OpenEye products or
# SaaS offerings (each a "Customer").
# Customer is hereby permitted to use, copy, and modify the Sample Code,
# subject to these terms. OpenEye claims no rights to Customer's
# modifications. Modification of Sample Code is at Customer's sole and
# exclusive risk. Sample Code may require Customer to have a then
# 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 __future__ import print_function
from openeye import oechem
mol = oechem.OEGraphMol()
oechem.OESmilesToMol(mol, "c1ccccc1C")
# create a substructure search object
ss = oechem.OESubSearch("c1ccccc1")
oechem.OEPrepareSearch(mol, ss)
if ss.SingleMatch(mol):
print("benzene matches toluene")
else:
print("benzene does not match toluene")
# @ </SNIPPET>
