mnx_inchikey_dict[inchikey] = mnx
inchikey = "-".join(inchikey.split('-')[0:2])
if (inchikey not in mnx_inchikey_dict):
mnx_inchikey_dict[inchikey] = mnx
inchikey = inchikey.split('-')[0]
if (inchikey not in mnx_inchikey_dict):
mnx_inchikey_dict[inchikey] = mnx
file_handle.close()
#Here we can cross-check the structures that are in ModelSEED to find ones where
#there is a match in eQuilibrator
compounds_helper = Compounds()
structures_dict = compounds_helper.loadStructures(["InChIKey"], ["ModelSEED"])
seed_mnx_structural_map = dict()
for cpd in structures_dict:
structure_type = 'InChIKey'
if (structure_type not in structures_dict[cpd]):
#The load structures function will return all compounds, so have to
#Check that the structure is there
continue
#As these are unique structures, i.e. 1-1 mapping with compound id,
#there's only ever one in each list for each compound
structure = list(structures_dict[cpd][structure_type].keys())[0]
#Here we check on three levels, we check the full string
#Then the deprotonated string, then the structure alone
#As per email from Elad and Moritz, we should not expect
#!/usr/bin/env python
import os, sys, re
temp = list()
header = 1
from BiochemPy import Compounds
import pybel
from rdkit.Chem import AllChem
from rdkit import RDLogger
lg = RDLogger.logger()
lg.setLevel(RDLogger.ERROR)
#Load Structures and Aliases
CompoundsHelper = Compounds()
Structures_Dict = CompoundsHelper.loadStructures(["SMILE", "InChI"],
["KEGG", "MetaCyc"])
Structures_Root = os.path.dirname(__file__) + "/../../Biochemistry/Structures/"
file_handle_dict = dict()
for source in "KEGG", "MetaCyc":
for struct_type in "InChI", "SMILE":
for struct_stage in "Charged", "Original":
file_string = "_".join((source, struct_type, struct_stage))
file_name = Structures_Root + source + "/" + struct_type + "_" + struct_stage + "_Formulas_Charges.txt"
file_handle_dict[file_string] = open(file_name, "w")
resolved_structures = open('Resolved_Structures.txt', 'w')
unresolved_structures = open('Unresolved_Structures.txt', 'w')
for struct_type in sorted(Structures_Dict.keys()):
# if(struct_type != 'InChI'):
#!/usr/bin/env python
import os, sys
temp = list()
header = 1
sys.path.append('../../Libs/Python')
from BiochemPy import Reactions, Compounds, InChIs
CompoundsHelper = Compounds()
Compounds_Dict = CompoundsHelper.loadCompounds()
Structures_Dict = CompoundsHelper.loadStructures(["InChI"], ["ModelSEED"])
diff_file = open("Compound_Charge_Differences.txt", 'w')
for cpd in sorted(Compounds_Dict.keys()):
if (cpd not in Structures_Dict):
#diff_file.write("Zero structures for "+cpd+"\n")
continue
if ('InChI' not in Structures_Dict[cpd]):
#diff_file.write("No InChI structure for "+cpd+"\n")
continue
current_charge = float(Compounds_Dict[cpd]['charge'])
#Parse out InChI formula and layers
inchi = list(Structures_Dict[cpd]['InChI'].keys())[0]
(inchi_formula, inchi_layers) = InChIs.parse(inchi)
inchi_charge = InChIs.charge(inchi_layers['q'], inchi_layers['p'])
if (inchi_charge != current_charge):
print("Warning: compound " + disambiguating_cpd +
" is obsolete, consider using the non-obsolete version")
Disambiguation_Object['from'] = {
'id': disambiguating_cpd,
'structures': {},
'aliases': {},
'names': {},
'formula': compounds_dict[disambiguating_cpd]['formula'],
'charge': compounds_dict[disambiguating_cpd]['charge'],
'mass': compounds_dict[disambiguating_cpd]['mass']
}
Aliases_Dict = compounds_helper.loadMSAliases()
Names_Dict = compounds_helper.loadNames()
Structures_Dict = compounds_helper.loadStructures(["InChI", "SMILE"],
["KEGG", "MetaCyc"])
#For reverse lookup
reverse_aliases_dict = dict()
for cpd in Aliases_Dict:
for source in Aliases_Dict[cpd]:
for alias in Aliases_Dict[cpd][source]:
if (alias not in reverse_aliases_dict):
reverse_aliases_dict[alias] = dict()
if (source not in reverse_aliases_dict[alias]):
reverse_aliases_dict[alias][source] = dict()
reverse_aliases_dict[alias][source][cpd] = 1
reverse_structures_dict = dict()
for type in Structures_Dict:
for alias in Structures_Dict[type]:
#!/usr/bin/env python
import os, sys
temp = list()
header = 1
sys.path.append('../../Libs/Python')
from BiochemPy import Compounds
CompoundsHelper = Compounds()
Structures_Dict = CompoundsHelper.loadStructures(["SMILE", "InChIKey"],
["ModelSEED"])
Compounds_Dict = CompoundsHelper.loadCompounds()
for cpd in sorted(Compounds_Dict.keys()):
if (cpd not in Structures_Dict):
Compounds_Dict[cpd]['inchikey'] = ""
Compounds_Dict[cpd]['smiles'] = ""
else:
Compounds_Dict[cpd]['inchikey'] = Structures_Dict[cpd].get(
'InChIKey', "")
Compounds_Dict[cpd]['smiles'] = Structures_Dict[cpd].get('SMILE', "")
print "Saving compounds"
CompoundsHelper.saveCompounds(Compounds_Dict)
cpd = array.pop(0)
if (len(header) == 0):
header = array
continue
if (cpd not in Overridden_Fields):
Overridden_Fields[cpd] = dict()
for i in range(len(array)):
if (array[i] == 'null' or array[i] == 10000000):
continue
Overridden_Fields[cpd][header[i]] = array[i]
compounds_helper = Compounds()
structures_dict = compounds_helper.loadStructures(
["SMILE", "InChI", "InChIKey"], ["ModelSEED"])
ignoring_structures_dict = compounds_helper.loadStructures(
["SMILE", "InChIKey"], ["KEGG", "MetaCyc"])
aliases_dict = compounds_helper.loadSourceAliases()
compounds_dict = compounds_helper.loadCompounds()
Structures_Root = os.path.dirname(__file__) + "/../../Biochemistry/Structures/"
inchikey_dict = dict()
smiles_dict = dict()
for cpd in structures_dict:
if ('InChIKey' in structures_dict[cpd]):
for struct in structures_dict[cpd]['InChIKey'].keys():
if (struct not in inchikey_dict):
inchikey_dict[struct] = list()
inchikey_dict[struct].append(cpd)
if ('SMILE' in structures_dict[cpd]):