def _add_node_metadata(self, node_id, agent):
agent_type = _get_agent_type(agent)
node_attribute = {'po': node_id, 'n': 'type', 'v': agent_type}
self.cx['nodeAttributes'].append(node_attribute)
for db_name, db_ids in agent.db_refs.items():
if not db_ids:
logger.warning('Missing db_id for %s' % agent)
continue
elif isinstance(db_ids, int):
db_id = str(db_ids)
elif isinstance(db_ids, list):
db_id = db_ids[0][0]
else:
db_id = db_ids
url = get_identifiers_url(db_name, db_id)
if not url:
continue
db_name_map = {
'UP': 'UniProt',
'PUBCHEM': 'PubChem',
'IP': 'InterPro',
'NXPFA': 'NextProtFamily',
'PF': 'Pfam',
'CHEBI': 'ChEBI'
}
name = db_name_map.get(db_name)
if not name:
name = db_name
node_attribute = {'po': node_id, 'n': name, 'v': url}
self.cx['nodeAttributes'].append(node_attribute)
def id_url(ag):
# Return identifier URLs in a prioritized order
# TODO: we should add handling for UPPRO here, however, that would require
# access to UniProt client resources in the context of the DB REST API
# which could be problematic
for db_name in link_namespace_order:
if db_name in ag.db_refs:
# Handle a special case where a list of IDs is given
if isinstance(ag.db_refs[db_name], list):
db_id = ag.db_refs[db_name][0]
if db_name == 'WM':
db_id = db_id[0]
else:
db_id = ag.db_refs[db_name]
# We can add more name spaces here if there are issues
if db_name in {'CHEBI'}:
db_id = ensure_prefix('CHEBI', db_id)
# Here we validate IDs to make sure we don't surface invalid
# links.
if not validate_id(db_name, db_id):
logger.debug('Invalid grounding encountered: %s:%s' %
(db_name, db_id))
continue
# Finally, we return a valid identifiers.org URL
return get_identifiers_url(db_name, db_id)
def get_annotation_text(stmt, annotate_agents=True):
ea = EnglishAssembler(stmts=[stmt])
annotation_text = ea.make_model()
if annotate_agents:
inserts = []
for agent_wc in ea.stmt_agents[0]:
for insert_begin, insert_len in inserts:
if insert_begin < agent_wc.coords[0]:
agent_wc.update_coords(insert_len)
db_ns, db_id = get_grounding(agent_wc.db_refs, grounding_ns)
if not db_ns:
continue
identifiers_url = \
identifiers.get_identifiers_url(db_ns, db_id)
grounding_text = '[%s](%s)' % (agent_wc.name, identifiers_url)
insert_len = len(grounding_text) - agent_wc.coords[1] + \
agent_wc.coords[0]
inserts.append((agent_wc.coords[0], insert_len))
before_part = annotation_text[:agent_wc.coords[0]]
after_part = annotation_text[agent_wc.coords[1]:]
annotation_text = ''.join(
[before_part, grounding_text, after_part])
return annotation_text
def test_get_identifiers_url():
# Get latest standard URL for a given namespace and ID
for ns_tuple, urls in ns_mapping.items():
url = get_identifiers_url(*ns_tuple)
assert url == urls[0], (url, urls[0], ns_tuple)
def test_signor():
sid = 'SIGNOR-PF15'
assert get_identifiers_url('SIGNOR', sid) == \
'https://signor.uniroma2.it/relation_result.php?id=%s' % sid
def test_chembl():
cid = '1229517'
assert get_identifiers_url('CHEMBL', cid) == \
'https://identifiers.org/chembl.compound:CHEMBL%s' % cid
assert get_identifiers_url('CHEMBL', 'CHEMBL%s' % cid) == \
'https://identifiers.org/chembl.compound:CHEMBL%s' % cid
def test_signor():
sid = 'SIGNOR-PF15'
assert get_identifiers_url('SIGNOR', sid) == \
'https://identifiers.org/signor:%s' % sid