def test_get_pubmed_id_for_pubchem_id(self):
pubmed_ids = QueryPubChem.get_pubmed_id_for_pubchem_id('3500')
self.assertIsNotNone(pubmed_ids)
self.assertEqual(pubmed_ids, ['10860942[uid]', '11961255[uid]'])
# wrong id
pubmed_ids = QueryPubChem.get_pubmed_id_for_pubchem_id('35000')
self.assertIsNone(pubmed_ids)
def test_get_pubchem_id_for_chembl_id(self):
pubchem_id = QueryPubChem.get_pubchem_id_for_chembl_id('CHEMBL521')
self.assertIsNotNone(pubchem_id)
self.assertEqual(pubchem_id, '3672')
# empty result
pubchem_id = QueryPubChem.get_pubchem_id_for_chembl_id('chembl521')
self.assertIsNone(pubchem_id)
# wrong id
pubchem_id = QueryPubChem.get_pubchem_id_for_chembl_id('3400')
self.assertIsNone(pubchem_id)
def test_update_metabolite_desc(self):
conn = Neo4jConnection(self.rtxConfig.neo4j_bolt,
self.rtxConfig.neo4j_username,
self.rtxConfig.neo4j_password)
nodes = conn.get_metabolite_nodes()
# generate random number array
random_indexes = random_int_list(0, len(nodes) - 1, 100)
for i in random_indexes:
# retrieve data from BioLink API
node_id = nodes[i]
pubchem_id = QueryKEGG.map_kegg_compound_to_pub_chem_id(node_id)
hmdb_url = QueryPubChem.get_description_url(pubchem_id)
desc = QueryHMDB.get_compound_desc(hmdb_url)
# retrieve data from Neo4j
node = conn.get_node(node_id)
self.assertIsNotNone(node)
self.assertIsNotNone(node['n']['id'])
self.assertIsNotNone(node['n']['description'])
self.assertEqual(node_id, node['n']['id'])
if node['n']['description'] != "None":
self.assertEqual(desc, node['n']['description'])
conn.close()
def test_get_description_url(self):
url = QueryPubChem.get_description_url("3324")
self.assertIsNotNone(url)
self.assertEqual(url, "http://www.hmdb.ca/metabolites/HMDB0000243")
# empty result
url = QueryPubChem.get_description_url("3500")
self.assertIsNone(url)
# wrong arg format
url = QueryPubChem.get_description_url("GO:2342343")
self.assertIsNone(url)
# wrong arg type
url = QueryPubChem.get_description_url(3500)
self.assertIsNone(url)
def read_interactions():
int_data = pandas.read_csv(QueryDGIdb.INTERACTIONS_TSV_URL, sep='\t')
int_data.fillna('', inplace=True)
res_list = []
for index, row in int_data.iterrows():
pmids = row['PMIDs']
gene_name = row['gene_name']
gene_claim_name = row['gene_claim_name']
if gene_name != '':
gene_symbol = gene_name
else:
if gene_claim_name != '':
gene_symbol = gene_claim_name
else:
continue
assert ',' not in gene_symbol
uniprot_ids_set = QueryDGIdb.mygene.convert_gene_symbol_to_uniprot_id(
gene_symbol)
if len(uniprot_ids_set) == 0:
continue
drug_chembl_id = row['drug_chembl_id']
drug_name = row['drug_name']
if drug_chembl_id != '':
if type(drug_chembl_id) == float:
print(row)
assert ',' not in drug_chembl_id
drug_chembl_id_set = {drug_chembl_id}
if drug_name == '':
print("warning; ChEMBL compound has no drug name",
file=sys.stderr)
else:
if drug_name != '':
assert ',' not in drug_name
drug_chembl_id_set = QueryPubChem.get_chembl_ids_for_drug(
drug_name)
if len(drug_chembl_id_set) == 0:
drug_chembl_id_set = QueryChEMBL.get_chembl_ids_for_drug(
drug_name)
if len(drug_chembl_id_set) == 0:
continue
else:
continue
interaction_claim_source_field = row['interaction_claim_source']
interaction_types_field = row['interaction_types']
if interaction_types_field != '':
assert type(interaction_types_field) == str
predicate_list = interaction_types_field.split(',')
else:
predicate_list = ['affects']
for uniprot_id in uniprot_ids_set:
for predicate_str in predicate_list:
res_list.append({
'drug_chembl_id':
drug_chembl_id,
'drug_name':
drug_name,
'predicate':
QueryDGIdb.predicate_map[predicate_str],
'predicate_extended':
predicate_str,
'protein_uniprot_id':
uniprot_id,
'protein_gene_symbol':
gene_symbol,
'sourcedb':
interaction_claim_source_field,
'pmids':
pmids
})
# print(drug_chembl_id + '\t' + predicate_str + '\t' + uniprot_id + '\t' + interaction_claim_source_field + '\t' + ','.join(pmids_list))
return res_list
def test_get_chembl_ids_for_drug(self):
sets = QueryPubChem.get_chembl_ids_for_drug('gne-493')
self.assertIsNotNone(sets)
self.assertEqual(sets, {'CHEMBL1084926'})