def LoadMolFromSmiles(smiles):
"""Returns an OBMol construcetd from an SMILES code"""
mol=OBMol()
loader=OBConversion()
loader.SetInAndOutFormats("smi","smi")
if not loader.ReadString(mol, smiles):
return None
mol.smilesCode=smiles
return mol
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 get_atomic_fragments_from_mol(xyz):
"""Get the fragments around each atom
Args:
xyz (string): XYZ file for the
"""
# Read in the molecule
mol = OBMol()
converter.ReadString(mol, xyz)
# Determine ring types
OBRingTyper().AssignTypes(mol)
# For each atom, get its local environment
return [get_atomic_environment(atom) for atom in OBMolAtomIter(mol)]
def all_to_all():
"""Function to compare all to all"""
# Set up the OpenBaebel conversion modules
sdconv = OBConversion()
ligref = OBMol()
# Define the residues and the proteisn to analyse
res_d, prot_list = get_dict("myFirstFile.txt")
# Now read in the ligand
sdconv.SetInFormat("sdf")
notatend = sdconv.ReadFile(ligref, "../mols.sdf")
out_d = {}
counter = 0
# Now read the ligand file
while notatend:
lig_name = ligref.GetTitle().strip(",")
prot_name = lig_name.split("_")[0]
if prot_name not in prot_list:
ligref = OBMol()
notatend = sdconv.Read(ligref)
continue
ligref.AddHydrogens()
counter += 1
print counter
for j, my_prot in enumerate(prot_list):
protref = read_prot(
r"C:\www\Protoype\media_coninchi\pdb" + "\\" + my_prot +
"al.pdb", res_d)
# Get the reference dictionary
refresdict = pp.getresiduedict(protref, res_d)
# Update this dict, to only residues in the binding site
new_d = get_fp(protref, ligref, res_d)
# Make sure it is a unique name for the output
while lig_name in out_d:
lig_name = lig_name + "Z"
# Add it to the dict
out_d[lig_name + my_prot] = {}
for res in new_d:
# Assign each residue the scores for each molecule
out_d[lig_name + my_prot][res] = new_d[res]
# Make the ligand
ligref = OBMol()
notatend = sdconv.Read(ligref)
# Now write the results out
write_res(out_d, res_d)
def ToOBMol(self):
atomCache={}
mol=OBMol()
mol.BeginModify()
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())
## mol.SetAromaticPerceived()
mol.AssignSpinMultiplicity()
## mol.UnsetAromaticPerceived()
mol.EndModify()
return mol
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
def one_to_many():
"""Function to take multiple confs of ONE ligand and generate their PLIFS against one template protein"""
# Set up the OpenBaebel conversion modules
sdconv = OBConversion()
ligref = OBMol()
# Define the residues and the proteisn to analyse
res_d, prot_list = get_dict("myFirstFile.txt")
# Now read in the ligand
sdconv.SetInFormat("sdf")
notatend = sdconv.ReadFile(ligref, "../out.sdf")
out_d = {}
counter = 0
my_prot = "1qmz"
protref = read_prot(
r"C:\www\Protoype\media_coninchi\pdb" + "\\" + my_prot + "al.pdb",
res_d)
# Now read the ligand file
while notatend:
lig_name = ligref.GetTitle().strip(",")
prot_name = lig_name.split("_")[0]
ligref.AddHydrogens()
counter += 1
print counter
# Get the reference dictionary
refresdict = pp.getresiduedict(protref, res_d)
# Update this dict, to only residues in the binding site
new_d = get_fp(protref, ligref, res_d)
# Add it to the dict
out_d[lig_name + str(counter)] = {}
for res in new_d:
# Assign each residue the scores for each molecule
out_d[lig_name + str(counter)][res] = new_d[res]
# Make the ligand
ligref = OBMol()
notatend = sdconv.Read(ligref)
# Now write the results out
write_res(out_d, res_d)
firstline = conflist.readline()
mollisttemp = [line for line in conflist]
mollist = []
scorelist = []
for mol in mollisttemp:
mollist.append(mol.split(',')[0])
scorelist.append(mol.split(',')[1])
os.chdir('..')
# opening the molecule files
pbf = protein_ligand_folder + '/protein_bindingsite_fixed.mol2'
conv = OBConversion()
conv.SetInFormat("mol2")
protfix = OBMol()
protref = OBMol()
ligref = OBMol()
docklig = OBMol()
dockprot = OBMol()
conv.ReadFile(protfix, pbf)
conv.ReadFile(protref, protein_reference)
conv.ReadFile(ligref, ligand_reference)
refresdict = getresiduedict(protref, residue_of_choice)
refringdict = getringdict(protref)
fixringdict = getringdict(protfix)
ringinteraction(refresdict, refringdict, residue_of_choice, protref,
ligref)
(*i).c->GetType(), (*i).d->GetType(), (*i).V, (*i).n, (*i).cosNPhi0,
(*i).tor * RAD_TO_DEG, (*i).energy);
```
"""
import sys
sys.path.insert(0, "/usr/local/lib/python3.6/site-packages")
import os
from openbabel import OBMol, OBConversion, OBMolAtomIter, OBForceField
mof = "csdac-linkers-cml/uio66-HNC3-alkane.cml"
obconversion = OBConversion()
obconversion.SetInAndOutFormats("cml", "cml")
obmol = OBMol()
obconversion.ReadFile(obmol, mof)
ff = OBForceField.FindForceField("UFF")
ff.SetLogToStdOut()
ff.SetLogLevel(3)
if not ff.Setup(obmol):
print("Error: could not setup force field")
ff.GetAtomTypes(obmol)
for atom_idx, obatom in enumerate(OBMolAtomIter(obmol)):
ff_atom_type = obatom.GetData("FFAtomType").GetValue()
print(ff_atom_type)
print(ff.Energy(True))
print(ff.E_Torsion())
def django_run(target, opt="XTAL"):
"""Function to take multiple confs of ONE ligand and generate their PLIFS against one template protein"""
# Set up the OpenBaebel conversion modules
sdconv = OBConversion()
ligref = OBMol()
# Define the residues and the proteisn to analyse
if os.path.isfile(
os.path.join(os.path.split(sys.argv[0])[0], 'data/res_def.py')):
res_d = [
trans_res(x) for x in ast.literal_eval(
open(
os.path.join(
os.path.split(sys.argv[0])[0],
'data/res_def.py')).read())[target.title].split()
]
print res_d
# Molecules
# Now read in the ligand
plif_method = PlifMethod()
plif_method.text = "PYPLIF"
feature_list = [
"POLAR", "FACE", "EDGE", "ACCEPTOR", "DONOR", "NEGATIVE", "POSITIVE"
]
try:
plif_method.validate_unique()
plif_method.save()
except ValidationError:
plif_method = PlifMethod.objects.get(text="PYPLIF")
out_d = {}
counter = 0
# Create a file for the protein
t = tempfile.NamedTemporaryFile(suffix=".pdb", delete=False)
my_prot = Protein.objects.get(code=target.title + "TEMP")
t.write(my_prot.pdb_info.name)
t.close()
protref = read_prot(t.name, res_d)
t = tempfile.NamedTemporaryFile(suffix=".sdf", delete=False)
t.close()
sdconv.SetInFormat("sdf")
if opt == "XTAL":
mols = Molecule.objects.exclude(
prot_id__code__contains=target.title).filter(
prot_id__target_id=target)
elif opt == "LLOOMMPPAA":
mols = []
sps = SynthPoint.objects.filter(target_id=target)
for s in sps:
mols.extend([m for m in s.mol_id.all()])
else:
print "UNKNOWN OPTION"
return
for dj_mol in mols:
out_sd = Chem.SDWriter(t.name)
out_sd.write(Chem.MolFromMolBlock(str(dj_mol.sdf_info)))
out_sd.close()
sdconv.ReadFile(ligref, t.name)
# Now make the new plif
new_plif = Plif()
new_plif.mol_id = dj_mol
new_plif.prot_id = my_prot
new_plif.method_id = plif_method
try:
new_plif.validate_unique()
new_plif.save()
except ValidationError:
new_plif = Plif.objects.get(mol_id=dj_mol,
prot_id=my_prot,
method_id=plif_method)
lig_name = ligref.GetTitle().strip(",")
prot_name = lig_name.split("_")[0]
ligref.AddHydrogens()
counter += 1
refresdict = pp.getresiduedict(protref, res_d)
new_d = get_fp(protref, ligref, res_d)
for res in new_d:
new_res = PlifRes()
new_res.res_name = res[:3]
new_res.res_num = int(res[3:])
new_res.prot_id = my_prot
try:
new_res.validate_unique()
new_res.save()
except ValidationError:
new_res = PlifRes.objects.get(res_name=res[:3],
res_num=int(res[3:]),
prot_id=my_prot)
new_plif.res_id.add(new_res)
for bit_num, bit in enumerate(new_d[res]):
new_bit = PlifBit()
new_bit.feature = feature_list[bit_num]
new_bit.method_id = plif_method
new_bit.res_id = new_res
try:
new_bit.validate_unique()
new_bit.save()
my_fun(dj_mol, new_bit, new_plif, bit)
except ValidationError:
new_bit = PlifBit.objects.get(
feature=feature_list[bit_num],
method_id=plif_method,
res_id=new_res)
new_bit.save()
new_plif.bit_id.add(new_bit)
my_fun(dj_mol, new_bit, new_plif, bit)
ligref = OBMol()
notatend = sdconv.Read(ligref)
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()
