def to_inchi(str_data, in_format):
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.ReadString(mol, str_data)
conv.SetOutFormat('inchi')
inchi = conv.WriteString(mol).rstrip()
conv.SetOptions('K', conv.OUTOPTIONS)
inchikey = conv.WriteString(mol).rstrip()
return (inchi, inchikey)
class Converter:
def __init__(self, can=False):
self._OBconverter = OBConversion()
self._OBconverter.SetOutFormat('smi')
options = 'in'
if can:
options += 'c'
self._OBconverter.SetOptions(options, OBConversion.OUTOPTIONS)
def getSmiles(self, mol):
smiles = self._OBconverter.WriteString(mol.OBMol)[:-1]
if '@' in smiles:
smiles = self._OBconverter.WriteString(mol.OBMol)[:-1]
return smiles
def convert_str(str_data, in_format, out_format):
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.SetOutFormat(out_format)
conv.ReadString(mol, str_data)
return (conv.WriteString(mol), conv.GetOutFormat().GetMIMEType())
def __init__(self, smile):
self._smile = smile.strip()
conv = OBConversion()
if not conv.SetInAndOutFormats('smi', 'inchi'):
raise 'Problem with openbabel'
mol = OBMol()
if not conv.ReadString(mol, self._smile):
raise TypeError, "No such smile: %s" % self._smile
self._inchi = conv.WriteString(mol).strip()
def autodetect_bonds(cjson):
mol = cjson_to_ob_molecule(cjson)
mol.ConnectTheDots()
mol.PerceiveBondOrders()
conv = OBConversion()
conv.SetInFormat('sdf')
conv.SetOutFormat('sdf')
sdf_str = conv.WriteString(mol)
cjson_str = avo_convert_str(sdf_str, 'sdf', 'cjson')
return json.loads(cjson_str)
def to_inchi(str_data, in_format):
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
# Hackish for now, convert to xyz first...
conv.SetOutFormat('xyz')
conv.ReadString(mol, str_data)
xyz = conv.WriteString(mol)
# Now convert to inchi and inchikey.
mol = OBMol()
conv.SetInFormat('xyz')
conv.ReadString(mol, xyz)
conv.SetOutFormat('inchi')
inchi = conv.WriteString(mol).rstrip()
conv.SetOptions("K", conv.OUTOPTIONS)
inchikey = conv.WriteString(mol).rstrip()
return (inchi, inchikey)
def ToCannonicalSmiles(self):
atomCache={}
mol=OBMol()
for sourceAtom in self.atoms:
atom=mol.NewAtom()
atom.SetAtomicNum(sourceAtom.GetAtomicNum())
if sourceAtom.IsAromatic():
atom.SetAromatic()
atom.SetSpinMultiplicity(2)
atomCache[sourceAtom]=atom
for sourceBond in self.bonds:
mol.AddBond(atomCache[sourceBond.atom1].GetIdx(), atomCache[sourceBond.atom2].GetIdx(), sourceBond.GetBondOrder())
writer=OBConversion()
writer.SetInAndOutFormats("smi", "can")
return writer.WriteString(mol).strip()
def write_ob_molecule(self, mol, format, f):
"""
Write an Open Babel molecule to file
:param mol: The molecule
:param format: The output format
:param f: The file to write output to
:param f_name: The file's name (for extension-finding purpose)
"""
conv = OBConversion()
if not conv.SetOutFormat(format):
raise ValueError("Error setting output format to " + format)
# write to file
try:
s = conv.WriteString(mol)
except (TypeError, ValueError, IOError):
raise ValueError("Error writing data file using OpenBabel")
if str.lower(format) == 'pdb':
s = s.replace("END", "ENDMDL")
f.write(s)
def convert_str(str_data,
in_format,
out_format,
gen3d=False,
add_hydrogens=False,
perceive_bonds=False,
out_options=None):
# Make sure that the start of InChI is valid before passing it to
# Open Babel, or Open Babel will crash the server.
if in_format.lower() == 'inchi':
validate_start_of_inchi(str_data)
if out_options is None:
out_options = {}
obMol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.SetOutFormat(out_format)
conv.ReadString(obMol, str_data)
if add_hydrogens:
obMol.AddHydrogens()
if gen3d:
# Generate 3D coordinates for the input
mol = pybel.Molecule(obMol)
mol.make3D()
if perceive_bonds:
obMol.ConnectTheDots()
obMol.PerceiveBondOrders()
for option, value in out_options.items():
conv.AddOption(option, conv.OUTOPTIONS, value)
return (conv.WriteString(obMol), conv.GetOutFormat().GetMIMEType())
kegg = Kegg.getInstance()
cids = kegg.get_all_cids()
print 'Found %d cids!' % len(cids)
smiles = open('smiles.txt', 'a')
nums = open('nums.txt', 'a')
if len(argv) < 2:
i = 0
else:
i = int(argv[1])
conv = OBConversion()
if not conv.SetInAndOutFormats('inchi', 'smi'):
raise 'Problem with openbabel'
for j, cid in enumerate(cids[i:]):
try:
nums.write(str(j + i) + '\n')
inchi = kegg.cid2inchi(cid)
mol = OBMol()
if not conv.ReadString(mol, inchi):
print cid
smiles.write(conv.WriteString(mol))
#smiles.write(kegg.cid2smiles(cid) + '\n')
except:
print cid
smiles.close()
nums.close()
# This script takes no arguments and should
# run from the src directory.
#
######################################
from pygibbs.kegg import Kegg
from openbabel import OBConversion, OBMol
from sys import argv
kegg = Kegg.getInstance()
cids = kegg.get_all_cids()
print 'Found %d cids!' % len(cids)
smiles = open('smiles.txt', 'w')
conv = OBConversion()
if not conv.SetInAndOutFormats('inchi', 'smi'):
raise 'Problem with openbabel'
for cid in cids:
try:
inchi = kegg.cid2inchi(cid)
mol = OBMol()
if not conv.ReadString(mol, inchi):
raise Exception
smiles.write(conv.WriteString(mol).strip() + '\n')
except:
print "C%05d cannot be converted to smiles" % cid
smiles.close()
def ToSmiles(self):
writer=OBConversion()
writer.SetInAndOutFormats("smi", "smi")
return writer.WriteString(self.ToOBMol()).strip()