def generate_data(name,inchi,rt='null'):
cmpd = pcp.get_compounds(inchi,'inchi')
cmpd = pcp.get_compounds(inchi,'inchi')
props = cmpd[0].to_dict(properties=['cactvs_fingerprint',
'isomeric_smiles', 'xlogp', 'rotatable_bond_count','charge','complexity',
'exact_mass','fingerprint'])
smiles=props['isomeric_smiles']
props['mol']=Chem.MolFromSmiles(smiles)
props['RT'] = rt
props['Name'] = name
# props['System'] = row['System']
desc = np.array(fps_plus_mw(props['mol']))
descdf = pd.DataFrame(desc)
descdf = descdf.T
# descdf.reindex([index])
newdf=pd.DataFrame(props,index=[0])
finaldf=pd.concat([descdf,newdf],axis=1)
return finaldf
def Join_Mummichog_Matches_Molecular_Features(result_df, file_dir):
# result_df = result_df.drop(['Unnamed: 0'], axis=1)
running_cmpd_list = []
for index, row in result_df.iterrows():
if index == 0:
name = row['Compound Name']
# print(name)
if name not in running_cmpd_list:
running_cmpd_list.append(name)
# inchi = row['InChI']
cmpd = pcp.get_compounds(name, 'name')
props = cmpd[0].to_dict(properties=[
'cactvs_fingerprint', 'isomeric_smiles', 'xlogp',
'rotatable_bond_count', 'charge', 'complexity',
'exact_mass', 'fingerprint'
])
smiles = props['isomeric_smiles']
props['mol'] = Chem.MolFromSmiles(smiles)
props['RT'] = row['rtmin']
props['Name'] = name
props['System'] = 'xcms'
desc = np.array(fps_plus_mw(props['mol']))
descdf = pd.DataFrame(desc)
descdf = descdf.T
descdf.reindex([index])
newdf = pd.DataFrame(props, index=[index])
finaldf = pd.concat([descdf, newdf], axis=1)
else:
print('cmpd already queried')
# print('test')
else:
name = row['Compound Name']
if name not in running_cmpd_list:
running_cmpd_list.append(name)
try:
cmpd = pcp.get_compounds(name, 'name')
except:
print('line bypassed')
pass
try:
props = cmpd[0].to_dict(properties=[
'cactvs_fingerprint', 'isomeric_smiles', 'xlogp',
'rotatable_bond_count', 'charge', 'complexity',
'exact_mass', 'fingerprint'
])
except:
print('line bypassed')
pass
# name = row['Name']
smiles = props['isomeric_smiles']
props['mol'] = Chem.MolFromSmiles(smiles)
props['RT'] = row['rtmin']
props['Name'] = name
props['System'] = 'xcms'
newdf = pd.DataFrame(props, index=[index])
desc = np.array(fps_plus_mw(props['mol']))
cols = range(len(desc))
descdf = pd.DataFrame(desc)
descdf = descdf.T
descdf.index = [index]
# descdf = descdf.T
# descdf = pd.DataFrame(descdf, index=[index])
interdf = pd.concat([descdf, newdf], axis=1)
finaldf = finaldf.append(interdf)
# else:
# print('cmpd already queried')
# print('on index ' + str(index+1) + ' of ' + str(len(result_df)))
finaldf.to_pickle(file_dir + 'RT_Folder\\mummichog_rt_features.p')
return finaldf
def SMRT_Database_Processor(smrt_file):
cid_dict = {}
with open(smrt_file, 'r') as f:
state = ''
num = 1
for line in f.readlines():
if state == 'Log CID':
key = line
state = 'Wait for RT Time'
if state == 'Log RT':
cid_dict[key] = float(line)
state = ''
print('On Compound {}'.format(str(num)))
num += 1
if line.startswith('> <PUBCHEM_COMPOUND_CID>'):
state = 'Log CID'
if line.startswith('> <RETENTION_TIME>'):
state = 'Log RT'
num = 1
saving_count = 1
list_of_df = []
list_of_unprocessed = []
list_of_processed = []
for cid in cid_dict.keys():
print('On compound {}'.format(num))
try:
cmpd = pcp.get_compounds(cid, 'cid')
list_of_processed.append(cid)
except:
list_of_unprocessed.append(cid)
time.sleep(10)
pd.DataFrame(list_of_unprocessed).to_pickle(
'compounds_skipped.pickle')
# props = cmpd[0].to_dict(properties=['cactvs_fingerprint',
# 'isomeric_smiles', 'xlogp', 'rotatable_bond_count','charge','complexity',
# 'exact_mass','fingerprint'])
name = cmpd[0].iupac_name
rt = cid_dict[cid]
props = cmpd[0].to_dict(properties=[
'cactvs_fingerprint', 'isomeric_smiles', 'xlogp',
'rotatable_bond_count', 'charge', 'complexity', 'exact_mass',
'fingerprint'
])
smiles = props['isomeric_smiles']
props['mol'] = Chem.MolFromSmiles(smiles)
props['Name'] = name
props['System'] = 'SMRT DATA'
props['RT'] = rt
desc = np.array(fps_plus_mw(props['mol']))
descdf = pd.DataFrame(desc)
descdf = descdf.T
descdf.reindex([num])
newdf = pd.DataFrame(props, index=[0])
finaldf = pd.concat([descdf, newdf], axis=1)
list_of_df.append(finaldf)
num += 1
saving_count += 1
if saving_count > 999:
final_df = pd.concat(list_of_df)
final_df.to_pickle('compoundsupto' + str(num) + '.pickle')
saving_count = 1
list_of_df = []
with open('processed.p', 'wb') as f:
pickle.dump(list_of_processed, f)
pd.DataFrame(list_of_unprocessed).to_pickle('compounds_skipped.pickle')
# props = cmpd[0].to_dict(properties=['cactvs_fingerprint',
# 'isomeric_smiles', 'xlogp', 'rotatable_bond_count','charge','complexity',
# 'exact_mass','fingerprint'])
name = cmpd[0].iupac_name
rt = cid_dict[cid]
props = cmpd[0].to_dict(properties=[
'cactvs_fingerprint', 'isomeric_smiles', 'xlogp',
'rotatable_bond_count', 'charge', 'complexity', 'exact_mass',
'fingerprint'
])
smiles = props['isomeric_smiles']
props['mol'] = Chem.MolFromSmiles(smiles)
props['Name'] = name
props['System'] = 'SMRT DATA'
props['RT'] = rt
desc = np.array(fps_plus_mw(props['mol']))
descdf = pd.DataFrame(desc)
descdf = descdf.T
descdf.reindex([num])
newdf = pd.DataFrame(props, index=[0])
finaldf = pd.concat([descdf, newdf], axis=1)
list_of_df.append(finaldf)
num += 1
saving_count += 1
if saving_count > 999:
final_df = pd.concat(list_of_df)
final_df.to_pickle('compoundsupto' + str(num) + '.pickle')
saving_count = 1
list_of_df = []
with open('processed.p', 'wb') as f:
pickle.dump(list_of_processed, f)