def InchiToInchiKey(inchi):
"""Return the InChI key for the given InChI string. Return None on error"""
ret = rdinchi.InchiToInchiKey(inchi)
if ret:
return ret
else:
return None
def smiles_to_inchi_inchikey(smiles,verbose=0):
error_counter=0
warning_counter=0
inchis=[]
inchis2=[]
inchiKeys=[]
for elem in smiles:
mol=Chem.MolFromSmiles(elem)
if(mol!=None):
inchi, retcode, message, logs, aux =rdinchi.MolToInchi(mol)
inchiKey=rdinchi.InchiToInchiKey(inchi)
inchis.append(inchi)
inchiKeys.append(inchiKey)
try:
mol, retcode, message, logs=rdinchi.InchiToMol(inchi)
if(mol!=None):
inchi2, retcode, message, logs, aux =rdinchi.MolToInchi(mol)
inchis2.append(inchi2)
else:
inchis2.append("XXX")
if(verbose!=0):
print(elem+ ": InChI cannot converted to mol object, added XXX instead!")
except:
if(verbose!=0):
print(retcode)
print(message)
print("Smiles:"+elem)
inchis2.append("XXX")
else:
inchis.append("XXX")
inchiKeys.append("XXX")
inchis2.append("XXX")
if(verbose!=0):
print(elem+ ": can not converted added XXX instead! ")
error_counter=error_counter+1
for i1,i2 in zip(inchis,inchis2):
if(i1!=i2):
if(verbose!=0):
print("Warning:"+i1+" - "+i2)
warning_counter=warning_counter+1
print("\nGeneration of InChI and InChIKey from SMILES is completed.")
print("Total errors:"+str(error_counter))
print("Total warnings:"+str(warning_counter)+"\n")
return inchis,inchiKeys
def parse_file(input_file, db_name):
""" takes all text from nanpdb database file and returns a list of lists
with NPs which is easy to use.
input_file: nanpdb database txt file
db_name: database name
"""
all_lines = input_file.split('\n')
all_lines = all_lines[:-1]
all_info_list = []
for line in all_lines:
line = line.split('\t')
info_per_row_list = []
for value in line:
my_string = ""
if len(value) == 0:
value = "NA"
my_string += value
info_per_row_list += [my_string]
info_per_row_list += [db_name]
all_info_list += [info_per_row_list]
attribute_names = ['MonoisotopicMass', 'InChI', 'SMILES', 'Identifier',\
'InChIKey2', 'InChIKey1', 'MolecularFormula', 'kingdom_name',\
'superclass_name', 'class_name', 'subclass_name']
mol_mass_list = []
inchi_list = []
SMILES_list = []
identifier_list = []
inchi_key1_list = []
inchi_key2_list = []
mol_formula_list = []
NA_list = []
for line in all_info_list:
# generate molecules
m = Chem.MolFromSmiles(line[0])
# MonoisotopicMass
mol_mass = str(Descriptors.ExactMolWt(m))[:-5]
mol_mass_list += [mol_mass]
# InChI
inchi = rdinchi.MolToInchi(m)
inchi_list += [inchi[0]]
# SMILES
SMILES_list += [line[0]]
# Identifier
identifier_list += [line[1]]
# MolecularFormula
mol_formula = rdMolDescriptors.CalcMolFormula(m)
mol_formula_list += [mol_formula]
# NA list
nr_of_structures = len(all_info_list)
NA_list += ['NA'] * nr_of_structures
# InChIKey
inchi_key_list = []
inchi_key_list2 = []
for inchi in inchi_list:
inchi_key = rdinchi.InchiToInchiKey(inchi)
inchi_key_list2 += [inchi_key]
inchi_key_list += inchi_key_list2
# InChiKey1 and InChiKey2
for inchikey in inchi_key_list:
inchikey = inchikey.split('-')
inchikey1 = inchikey[0]
inchikey2 = inchikey[1]
inchi_key1_list += [inchikey1]
inchi_key2_list += [inchikey2]
overall_list = [mol_mass_list]+[inchi_list]+[SMILES_list]+\
[identifier_list]+[inchi_key2_list]+[inchi_key1_list]+[mol_formula_list]+\
[NA_list]+[NA_list]+[NA_list]+[NA_list]
return attribute_names, overall_list
def parse_data(input_file):
""" takes all text from norine database file and returns a list of lists
with all CLASS data and an attribute list.
input_file: norine database txt file
"""
attribute_names = ['MonoisotopicMass', 'InChI', 'SMILES', 'Identifier',\
'InChIKey2', 'InChIKey1', 'MolecularFormula', 'kingdom_name',\
'superclass_name', 'class_name', 'subclass_name']
mol_mass_list = []
inchi_list = []
SMILES_list = []
identifier_list = []
inchi_key1_list = []
inchi_key2_list = []
mol_formula_list = []
NA_list = []
pre_SMILES_list = []
identifier_list = []
all_lines = input_file.split('\n')
all_lines = all_lines[2:]
for line in all_lines:
line = line.split('\t')
#Convert to mol and remove invalid structures
smile_string = ''
id_string = ''
m = line[2]
id_name = line[0]
mol = Chem.MolFromSmiles(m)
if mol != None:
smile_string += m
id_string += id_name
pre_SMILES_list += [smile_string]
#Source identifiers
identifier_list += [id_string]
pre_inchi_list = []
for smile in pre_SMILES_list:
#Generate mol
m = Chem.MolFromSmiles(smile)
#SMILES
sm = Chem.MolToSmiles(m)
SMILES_list += [sm]
#Monoisotopic mass
mol_weigth = Descriptors.ExactMolWt(m)
mol_mass_list += [mol_weigth]
#Mol Forumula
mol_formula = rdMolDescriptors.CalcMolFormula(m)
mol_formula_list += [mol_formula]
# InChI
inchi = rdinchi.MolToInchi(m)
pre_inchi_list += [inchi[0]]
# InChIKey1 and InChIKey2
for inchi in pre_inchi_list:
if not str(inchi).startswith('InCh'):
inchi = 'NA'
inchi_list += [inchi]
pre_inchi_key_list = []
for inchi2 in inchi_list:
if inchi2 == 'NA':
inchi_key = "NA-NA"
pre_inchi_key_list += [inchi_key]
if inchi2 != 'NA':
inchi_key = rdinchi.InchiToInchiKey(inchi2)
pre_inchi_key_list += [inchi_key]
for inchi_key in pre_inchi_key_list:
inchi_key = inchi_key.split('-')
inchi_key2 = inchi_key[1]
inchi_key2_list += [inchi_key2]
inchi_key1 = inchi_key[0]
inchi_key1_list += [inchi_key1]
# NA list
nr_of_structures = len(SMILES_list)
NA_list += ['NA'] * nr_of_structures
overall_list = [mol_mass_list]+[inchi_list]+[SMILES_list]+\
[identifier_list]+[inchi_key2_list]+[inchi_key1_list]+[mol_formula_list]+\
[NA_list]+[NA_list]+[NA_list]+[NA_list]
return attribute_names, overall_list
def create_CLASS_data(data_dict):
""" Generates CLASS data for the strepto data present in the strep_dict.
input_file: streptodb dictionary
"""
attribute_names = ['MonoisotopicMass', 'InChI', 'SMILES', 'Identifier',\
'InChIKey2', 'InChIKey1', 'MolecularFormula', 'kingdom_name',\
'superclass_name', 'class_name', 'subclass_name']
mol_mass_list = []
inchi_list = []
SMILES_list = []
identifier_list = []
inchi_key1_list = []
inchi_key2_list = []
mol_formula_list = []
NA_list = []
# Identifier
identifier_list = data_dict['compound_id']
# SMILES
SMILES_list = data_dict['canonical_smiles']
for SMILE in SMILES_list:
# generate molecules
m = Chem.MolFromSmiles(SMILE)
# MonoisotopicMass
mol_mass = str(Descriptors.ExactMolWt(m))[:-3]
mol_mass_list += [mol_mass]
# InChI
inchi = rdinchi.MolToInchi(m)
inchi_list += [inchi[0]]
# MolecularFormula
mol_formula = rdMolDescriptors.CalcMolFormula(m)
mol_formula_list += [mol_formula]
# NA list
nr_of_structures = len(data_dict['canonical_smiles'])
NA_list += ['NA'] * nr_of_structures
# InChIKey
inchi_key_list = []
inchi_key_list2 = []
for inchi in inchi_list:
inchi_key = rdinchi.InchiToInchiKey(inchi)
inchi_key_list2 += [inchi_key]
inchi_key_list += inchi_key_list2
# InChiKey1 and InChiKey2
for inchikey in inchi_key_list:
inchikey = inchikey.split('-')
inchikey1 = inchikey[0]
inchikey2 = inchikey[1]
inchi_key1_list += [inchikey1]
inchi_key2_list += [inchikey2]
overall_list = [mol_mass_list]+[inchi_list]+[SMILES_list]+\
[identifier_list]+[inchi_key2_list]+[inchi_key1_list]+[mol_formula_list]+\
[NA_list]+[NA_list]+[NA_list]+[NA_list]
return attribute_names, overall_list
def generate_data(input_file):
""" takes all text from the input structure data file and returns a list of
lists with all generated data needed for the sqlite database.
input_file: input structure txt file
"""
mol_mass_list = []
inchi_list = []
SMILES_list = []
identifier_list = []
inchi_key1_list = []
inchi_key2_list = []
mol_formula_list = []
NA_list = []
pre_SMILES_list = []
identifier_list = []
all_lines = input_file.split('\n')
if all_lines[-1] == '':
all_lines = all_lines[:-1]
for line in all_lines:
line = line.split('\t')
#Convert to mol and remove invalid structures
smile_string = ''
id_string = ''
m = line[0]
id_name = line[1]
mol = Chem.MolFromSmiles(m)
if mol != None:
smile_string += m
id_string += id_name
pre_SMILES_list += [smile_string]
#Source identifiers
identifier_list += [id_string]
pre_inchi_list = []
for smile in pre_SMILES_list:
#Generate mol
m = Chem.MolFromSmiles(smile)
#SMILES, canonical
sm = Chem.MolToSmiles(m)
SMILES_list += [sm]
#Monoisotopic mass
mol_weigth = Descriptors.ExactMolWt(m)
mol_mass_list += [mol_weigth]
#Mol Forumula
mol_formula = rdMolDescriptors.CalcMolFormula(m)
mol_formula_list += [mol_formula]
# InChI
inchi = rdinchi.MolToInchi(m)
pre_inchi_list += [inchi[0]]
# InChIKey1 and InChIKey2
for inchi in pre_inchi_list:
if not str(inchi).startswith('InCh'):
inchi = 'NA'
inchi_list += [inchi]
pre_inchi_key_list = []
for inchi2 in inchi_list:
if inchi2 == 'NA':
inchi_key = "NA-NA"
pre_inchi_key_list += [inchi_key]
if inchi2 != 'NA':
inchi_key = rdinchi.InchiToInchiKey(inchi2)
pre_inchi_key_list += [inchi_key]
for inchi_key in pre_inchi_key_list:
inchi_key = inchi_key.split('-')
inchi_key2 = inchi_key[1]
inchi_key2_list += [inchi_key2]
inchi_key1 = inchi_key[0]
inchi_key1_list += [inchi_key1]
# NA list
nr_of_structures = len(SMILES_list)
NA_list += ['NA'] * nr_of_structures
overall_list = [mol_mass_list]+[inchi_list]+[SMILES_list]+\
[identifier_list]+[inchi_key2_list]+[inchi_key1_list]+[mol_formula_list]+\
[NA_list]+[NA_list]+[NA_list]+[NA_list]
return overall_list
