def is_human_protein(bio_ontology, node):
if bio_ontology.get_ns(node) in {'HGNC', 'FPLX'}:
return True
elif bio_ontology.get_ns(node) == 'UP' and \
uniprot_client.is_human(bio_ontology.get_id(node)):
return True
return False
def is_human_protein(bio_ontology, node):
"""Return True if the given ontology node is a human protein."""
if bio_ontology.get_ns(node) == 'HGNC':
return True
elif bio_ontology.get_ns(node) == 'UP' and \
uniprot_client.is_human(bio_ontology.get_id(node)):
return True
return False
def add_ido_parents(bio_ontology: BioOntology):
ido_root = bio_ontology.label('IDO', '0')
bio_ontology.add_node(ido_root, name='infectious disease concept')
edges_to_add = []
for node in bio_ontology.nodes():
if bio_ontology.get_ns(node) == 'IDO' and not \
bio_ontology.get_parents(*bio_ontology.get_ns_id(node)):
edges_to_add.append((node, ido_root, {'type': 'isa'}))
bio_ontology.add_edges_from(edges_to_add)
def add_efo_parents(bio_ontology):
edges_to_add = []
efo_root = 'EFO:0000001'
for node in bio_ontology.nodes():
if bio_ontology.get_ns(node) == 'EFO' and \
not bio_ontology.get_parents(*bio_ontology.get_ns_id(node)):
edges_to_add.append((node, efo_root, {'type': 'isa'}))
print('Adding %d EFO isa edges.' % len(edges_to_add))
bio_ontology.add_edges_from(edges_to_add)
def add_drugbank_parents(bio_ontology):
edges_to_add = []
drugbank_root = 'DRUGBANK:DB00000'
bio_ontology.add_node(drugbank_root, name='Drugs')
for node in bio_ontology.nodes():
if bio_ontology.get_ns(node) == 'DRUGBANK' and \
not bio_ontology[node]:
edges_to_add.append((node, drugbank_root, {'type': 'isa'}))
print('Adding %d DRUGBANK isa edges.' % len(edges_to_add))
bio_ontology.add_edges_from(edges_to_add)
def is_protein_family(bio_ontology, node):
"""Return True if the given ontology node is a protein family."""
if bio_ontology.get_ns(node) == 'FPLX':
return True
return False
def is_non_human_protein(bio_ontology, node):
if bio_ontology.get_ns(node) == 'UP' and \
not uniprot_client.is_human(bio_ontology.get_id(node)):
return True
return False
networkx.draw_networkx(G, pos, node_color='pink')
edge_labels = networkx.get_edge_attributes(G, 'source')
networkx.draw_networkx_edge_labels(G, pos, edge_labels=edge_labels)
plt.show()
if __name__ == '__main__':
bio_ontology.initialize()
xrefs = [(e[0], e[1]) for e in bio_ontology.edges(data=True)
if e[2]['type'] == 'xref']
xrefg = bio_ontology.edge_subgraph(xrefs)
comps = networkx.algorithms.strongly_connected_components(xrefg)
problems = []
for comp in comps:
namespaces = [bio_ontology.get_ns(node) for node in comp]
cnt = Counter(namespaces)
if any(v > 1 for k, v in cnt.items()):
problems.append(comp)
print('Found %d problems in total' % len(problems))
problems_by_ns = defaultdict(list)
for problem in problems:
nscnt = Counter([bio_ontology.get_ns(n) for n in problem])
namespaces = [ns for ns, cnt in nscnt.items() if cnt > 1]
for ns in namespaces:
problems_by_ns[ns].append(problem)
for ns, problems_ns in problems_by_ns.items():
print(ns, len(problems_ns))