def atom_count(str_data, in_format):
mol = OBMol()
conv = OBConversion()
conv.SetInFormat(in_format)
conv.ReadString(mol, str_data)
return mol.NumAtoms()
def sucess_analyses(output, leader_list, n, ref):
print "Calculating RMSD with the reference"
ext = ref.split(".")[1]
conv = OBConversion()
conv.SetInFormat(ext)
mol = OBMol()
conv.ReadFile(mol, ref)
str_info = "\t%s\t %20s\t %20s\n" % ("File", "Model", "RMSD")
for i in range(0, n):
rmsd = getRMSD(leader_list[i], mol)
str_info += "%s\t%20s \t%20.3f\n" % (leader_list[i].getFileBelow(),
leader_list[i].getID(), rmsd)
out_log = file(output + "_rmsd_" + ".info", "w")
out_log.write(str_info)
out_log.close()
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 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 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 loadReferenceMolecule(file_name):
ext = file_name.split(".")[1]
mol = Molecula()
conv = OBConversion()
conv.SetInFormat(ext)
conv.ReadFile(mol, file_name)
return mol
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 read_prot(prot_file, res_d):
"""Function to read in a protein to an OBMol"""
conv = OBConversion()
protref = OBMol()
conv.SetInFormat("pdb")
conv.ReadFile(protref, prot_file)
# Now assign the residue names
i = 0
my_res = []
for residue in OBResidueIter(protref):
i += 1
residue.SetName(residue.GetName() + str(residue.GetNum()))
my_res.append(residue.GetName())
# Now check that all the residues exist and print out if not
fail_counter = 0
fail_list = []
# Loop through the res and check they are in the list
for res_me in res_d:
if res_me not in my_res:
fail_counter += 1
fail_list.append(res_me)
# If it's out of register by one do again
if fail_counter > 0:
i = 0
my_res = []
for residue in OBResidueIter(protref):
i += 1
residue.SetName(residue.GetName() + str(residue.GetNum()))
my_res.append(residue.GetName())
# Now check that all the residues exist and print out if not
fail_counter = 0
fail_list = []
# Loop through the res and check they are in the list
for res_me in res_d:
if res_me not in my_res:
fail_counter += 1
fail_list.append(res_me)
out_err.write(prot_file + ",")
out_err.write(str(fail_counter) + "\n")
out_err.write(str(fail_list))
out_err.write(str(my_res))
out_err.write(str(res_d))
protref.AddHydrogens()
return protref
def load_conformation(mol_list):
for mol in mol_list:
file_name = mol.getFileBelow().split('.')[0] + ".pdb"
#########################################################
#Substiuir esses passos para remocao do openbabel ######
conv = OBConversion()
conv.SetInFormat("pdb")
end = conv.ReadFile(mol, file_name)
for i in range(1, int(mol.getID())):
mol = Molecula()
end = conv.Read(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 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
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)
print 'The protein ligand folder can not be found'
sys.exit(1)
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)
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)
def getLigandListLeaders(list_files):
obmol_list = []
for f in list_files:
mol = Molecula()
#########################################################
#Substiuir esses passos para remocao do openbabel ######
pdb_file_name = f + ".pdb"
conv = OBConversion()
conv.SetInFormat("pdb")
end = conv.ReadFile(mol, pdb_file_name)
obmol_list.append(mol)
###########################################################
##### Etapa de busca da informacao da energia #############
j = 0
for f in list_files:
log_file_name = f + ".log"
file_log = open(log_file_name)
for lines in file_log:
if re.search("^\$Leader_Info", lines) is not None:
aux = int(re.search("\d+", lines).group(0))
obmol_list[j].setIDLog(aux)
elif re.search("Total_Energy", lines) is not None:
obmol_list[j].setTotalEnergy(
float(re.search(".\d+.\d+", lines).group(0)))
elif re.search("vdW", lines) is not None:
obmol_list[j].setVdw(
float(re.search(".\d+.\d+", lines).group(0)))
elif re.search("Coulomb", lines) is not None:
obmol_list[j].setCoulomb(
float(re.search(".\d+.\d+", lines).group(0)))
elif re.search("^}", lines) is not None:
obmol_list[j].setFileBelow(log_file_name)
obmol_list[j].setInteractionEnergy()
break
j += 1
file_log.close()
return obmol_list
def getLigandListRMSD(list_files, ref):
obmol_list = []
rmsd_list = []
j = 0
for f in list_files:
mol = Molecula()
#########################################################
#Substiuir esses passos para remocao do openbabel ######
pdb_file_name = f + ".pdb"
conv = OBConversion()
conv.SetInFormat("pdb")
end = conv.ReadFile(mol, pdb_file_name)
obmol_list.append(mol)
while end:
mol = Molecula()
end = conv.Read(mol)
obmol_list.append(mol)
obmol_list.pop() #retira a ultima molecula
###########################################################
##### Etapa de busca da informacao da energia #############
log_file_name = f + ".log"
file_log = open(log_file_name)
for lines in file_log:
if re.search("^\$Leader_Info", lines) is not None:
obmol_list[j].setIDLog(int(re.search("\d+", lines).group(0)))
elif re.search("Total_Energy", lines) is not None:
obmol_list[j].setTotalEnergy(
float(re.search(".\d+.\d+", lines).group(0)))
elif re.search("vdW", lines) is not None:
obmol_list[j].setVdw(
float(re.search(".\d+.\d+", lines).group(0)))
elif re.search("Coulomb", lines) is not None:
obmol_list[j].setCoulomb(
float(re.search(".\d+.\d+", lines).group(0)))
elif re.search("^}", lines) is not None:
obmol_list[j].setFileBelow(log_file_name)
obmol_list[j].setInteractionEnergy()
j += 1
qsort_RMSD(obmol_list, 0, len(obmol_list) - 1, ref)
rmsd_list.append(obmol_list[0])
obmol_list = []
file_log.close()
j = 0
qsort_RMSD(rmsd_list, 0, len(rmsd_list) - 1, ref)
return rmsd_list