def run_n2s():
n2s = {}
if os.path.isfile('n2s.pickle'):
with open('n2s.pickle') as fin:
n2s = pickle.load(fin)
print('Starting with %s names' % len(n2s))
result_dir = '../examples/mp/results'
for filename in os.listdir(result_dir):
if not filename.endswith('.json'):
continue
print(filename)
with open('%s/%s' % (result_dir, filename)) as fin:
results = json.loads(fin.read().decode('utf8'))
for compound in results:
if 'names' not in compound or 'melting_points' not in compound:
continue
for name in compound['names']:
if name not in n2s:
results = cirpy.query(name.encode('utf8'), 'smiles',
['name_by_opsin', 'name_by_cir'])
print(name)
print([(r.value, r.resolver) for r in results])
n2s[name] = [(r.value, r.resolver) for r in results]
with open('n2s.pickle', 'w') as fout:
pickle.dump(n2s, fout)
def process_cas(request):
cas_no = request.GET.get('cas_number')
try:
obj = Odorant.objects.get(cas_number__exact=cas_no)
data = {
'object_exists': obj.get_absolute_url(),
'object_exists_name': str(obj),
}
return JsonResponse(data)
except ObjectDoesNotExist:
pass
try:
smiles = cirpy.query(cas_no, 'smiles')[0].value
pcp_query = pcp.get_compounds(smiles, 'smiles')[0]
cid_no = pcp_query.cid
except IndexError:
return JsonResponse({
'error': 'No compound found for this CAS number'
})
if smiles and cid_no:
data = {
'chemical_name': Odorant.scrape_compound_name(cid_no),
'iupac_name': pcp_query.iupac_name,
'structure_url': 'https://pubchem.ncbi.nlm.nih.gov/image/imgsrv.fcgi?cid={}&t=l'.format(cid_no),
'hidden_cid': cid_no,
'smiles': smiles,
}
return JsonResponse(data)
def test_query_dict(self):
"""Test dict-style access to result attributes."""
results = query('Morphine', 'inchi')
self.assertEqual(len(results), 2)
self.assertEqual(results[1]['value'], 'InChI=1/C17H19NO3/c1-18-7-6-17-10-3-5-13(20)16(17)21-15-12(19)4-2-9(14(15)17)8-11(10)18/h2-5,10-11,13,16,19-20H,6-8H2,1H3/t10-,11+,13?,16-,17-/m0/s1')
self.assertEqual(results[1]['notation'], 'Morphine')
self.assertEqual(results[1]['resolver'], 'name_by_cir')
def test_query_dict(self):
"""Test dict-style access to result attributes."""
results = query('Morphine', 'inchi')
self.assertEqual(len(results), 2)
self.assertEqual(
results[1]['value'],
'InChI=1/C17H19NO3/c1-18-7-6-17-10-3-5-13(20)16(17)21-15-12(19)4-2-9(14(15)17)8-11(10)18/h2-5,10-11,13,16,19-20H,6-8H2,1H3/t10-,11+,13?,16-,17-/m0/s1'
)
self.assertEqual(results[1]['notation'], 'Morphine')
self.assertEqual(results[1]['resolver'], 'name_by_cir')
def test_morphine_inchi(self):
"""Test morphine query for inchi returns expected result."""
results = query('morphine', 'inchi')
self.assertEqual(len(results), 2)
self.assertEqual(results[1].input, 'morphine')
self.assertEqual(results[1].representation, 'inchi')
self.assertEqual(results[1].resolver, 'name_by_cir')
self.assertEqual(results[1].input_format, 'chemical name (CIR)')
self.assertEqual(results[1].notation, 'Morphine')
self.assertEqual(results[1].value, 'InChI=1/C17H19NO3/c1-18-7-6-17-10-3-5-13(20)16(17)21-15-12(19)4-2-9(14(15)17)8-11(10)18/h2-5,10-11,13,16,19-20H,6-8H2,1H3/t10-,11+,13?,16-,17-/m0/s1')
def test_morphine_inchi(self):
"""Test morphine query for inchi returns expected result."""
results = query('morphine', 'inchi')
self.assertEqual(len(results), 2)
self.assertEqual(results[1].input, 'morphine')
self.assertEqual(results[1].representation, 'inchi')
self.assertEqual(results[1].resolver, 'name_by_cir')
self.assertEqual(results[1].input_format, 'chemical name (CIR)')
self.assertEqual(results[1].notation, 'Morphine')
self.assertEqual(
results[1].value,
'InChI=1/C17H19NO3/c1-18-7-6-17-10-3-5-13(20)16(17)21-15-12(19)4-2-9(14(15)17)8-11(10)18/h2-5,10-11,13,16,19-20H,6-8H2,1H3/t10-,11+,13?,16-,17-/m0/s1'
)
def process_bioactive_identifier(request):
cas_no = request.GET.get('cas_number')
inchikey = request.GET.get('inchikey', '').strip()
obj = None
if cas_no:
obj = Bioactive.objects.filter(
chemical_properties__synonyms__icontains=cas_no).first()
elif inchikey:
obj = Bioactive.objects.filter(inchikey__exact=inchikey).first()
if obj:
data = {
'object_exists': obj.get_absolute_url(),
'object_exists_name': str(obj),
}
return JsonResponse(data)
try:
iupac_name = None
if cas_no:
smiles = cirpy.query(cas_no, 'smiles')[0].value
if '.' in smiles:
smiles = [i for i in smiles.split('.') if len(i) > 5][0]
pcp_query = pcp.get_compounds(smiles, 'smiles')[0]
if not pcp_query.iupac_name:
iupac_name = cirpy.resolve(smiles, 'iupac_name', ['smiles'])
else:
pcp_query = pcp.get_compounds(inchikey, 'inchikey')[0]
if not pcp_query.iupac_name:
iupac_name = cirpy.resolve(inchikey, 'iupac_name',
['stdinchikey'])
if not pcp_query.cid:
raise IndexError
except (IndexError, pcp.BadRequestError):
return JsonResponse({'error': 'No compound found for this CAS number'})
data = {
'chemical_name':
Bioactive.scrape_compound_name(pcp_query.cid),
'iupac_name':
pcp_query.iupac_name or iupac_name or 'n/a',
'inchikey':
pcp_query.inchikey,
'structure_url':
'https://pubchem.ncbi.nlm.nih.gov/image/imgsrv.fcgi?cid={}&t=l'.
format(pcp_query.cid),
'hidden_cid':
pcp_query.cid,
'smiles':
pcp_query.isomeric_smiles or pcp_query.canonical_smiles or '',
}
return JsonResponse(data)
def call_compound_data(self):
for cas_no in list(self.get_unique_cas_numbers())[:4]:
cirpy_query = cirpy.query(cas_no, 'smiles')
if cirpy_query:
smiles = cirpy_query[0].value
name = cirpy.Molecule(smiles).iupac_name
# usually a string but sometimes a list of strings
name = name[0] if isinstance(name, list) else name
if name:
pcp_data = pcp.get_compounds(smiles, 'smiles')
cid_no = pcp_data[0].cid
chem_properties = {
a: getattr(pcp_data[0], b)
for a, b in (('xlogp', 'xlogp'), ('hac',
'heavy_atom_count'),
('rbc', 'rotatable_bond_count'))
}
chem_properties.update({
'mw':
int(pcp_data[0].molecular_weight),
'synonyms':
' '.join(pcp_data[0].synonyms[:5]),
'hetac':
len(''.join(
[i for i in smiles
if i in [
'O',
'N',
'S',
]]))
})
self.compound_data.append({
'cas_number':
cas_no,
'smiles':
smiles,
'chemical_name':
CompoundMixin.scrape_compound_name(cid_no),
'iupac_name':
name.lower(),
'cid_number':
cid_no,
'chemical_properties':
chem_properties,
})
# may help avoid timeout with pcp.get_compounds
sleep(5)
def test_custom_resolvers(self):
"""Test expected results are returned when using custom name resolvers."""
results = query('2,4,6-trinitrotoluene', 'smiles')
self.assertEqual(len(results), 2)
self.assertEqual(
results[0],
Result(input='2,4,6-trinitrotoluene',
representation='smiles',
resolver='name_by_opsin',
input_format='IUPAC name (OPSIN)',
notation='2,4,6-trinitrotoluene',
value='Cc1c(cc(cc1[N+]([O-])=O)[N+]([O-])=O)[N+]([O-])=O'))
self.assertEqual(
results[1],
Result(input='2,4,6-trinitrotoluene',
representation='smiles',
resolver='name_by_cir',
input_format='chemical name (CIR)',
notation='2,4,6-Trinitrotoluene',
value='Cc1c(cc(cc1[N+]([O-])=O)[N+]([O-])=O)[N+]([O-])=O'))
def test_invalid_representation_query(self):
"""Test that HTTPError is raised when an invalid representation is specified."""
with self.assertRaises(HTTPError):
query('Morphine', 'ogiuewrgpw')
def test_no_result_query(self):
"""Test that an empty list is returned when there are no results."""
self.assertEqual(query('sjkvhaldfu', 'smiles'), [])
def test_custom_resolvers(self):
"""Test expected results are returned when using custom name resolvers."""
results = query('2,4,6-trinitrotoluene', 'smiles')
self.assertEqual(len(results), 2)
self.assertEqual(results[0], Result(input='2,4,6-trinitrotoluene', representation='smiles', resolver='name_by_opsin', input_format='IUPAC name (OPSIN)', notation='2,4,6-trinitrotoluene', value='Cc1c(cc(cc1[N+]([O-])=O)[N+]([O-])=O)[N+]([O-])=O'))
self.assertEqual(results[1], Result(input='2,4,6-trinitrotoluene', representation='smiles', resolver='name_by_cir', input_format='chemical name (CIR)', notation='2,4,6-Trinitrotoluene', value='Cc1c(cc(cc1[N+]([O-])=O)[N+]([O-])=O)[N+]([O-])=O'))
def test_invalid_representation_query(self):
"""Test that HTTPError is raised when an invalid representation is specified."""
with self.assertRaises(HTTPError):
query('Morphine', 'ogiuewrgpw')
def test_no_result_query(self):
"""Test that an empty list is returned when there are no results."""
self.assertEqual(query('sjkvhaldfu', 'smiles'), [])
