def atom_count(str_data, in_format):
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.ReadString(mol, str_data)
return mol.NumAtoms()
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 cjson_to_ob_molecule(cjson):
cjson_str = json.dumps(cjson)
sdf_str = avo_convert_str(cjson_str, 'cjson', 'sdf')
conv = OBConversion()
conv.SetInFormat('sdf')
conv.SetOutFormat('sdf')
mol = OBMol()
conv.ReadString(mol, sdf_str)
return mol
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 LoadMolFromSmiles(smiles):
"""Returns an OBMol construcetd from an SMILES code"""
smiles = sorted(smiles.split("."), key=len)[-1] ## Strip salts
mol=OBMol()
loader=OBConversion()
loader.SetInAndOutFormats("smi","smi")
if not loader.ReadString(mol, smiles):
return None
mol.smilesCode=smiles
return mol
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 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)
def get_formula(str_data, in_format):
# Inchi must start with 'InChI='
if in_format == 'inchi' and not str_data.startswith('InChI='):
str_data = 'InChI=' + str_data
validate_start_of_inchi(str_data)
# Get the molecule using the "Hill Order" - i. e., C first, then H,
# and then alphabetical.
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.ReadString(mol, str_data)
return mol.GetFormula()
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())
def properties(str_data, in_format, add_hydrogens=False):
# Returns a dict with the atom count, formula, heavy atom count,
# mass, and spaced formula.
if in_format == 'inchi' and not str_data.startswith('InChI='):
# Inchi must start with 'InChI='
str_data = 'InChI=' + str_data
validate_start_of_inchi(str_data)
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.ReadString(mol, str_data)
if add_hydrogens:
mol.AddHydrogens()
props = {}
props['atomCount'] = mol.NumAtoms()
props['formula'] = mol.GetFormula()
props['heavyAtomCount'] = mol.NumHvyAtoms()
props['mass'] = mol.GetMolWt()
props['spacedFormula'] = mol.GetSpacedFormula()
return props
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()