sys.exit()
id_name = None
if args.idfile is not None:
id_name = IDMap(args.idfile)
gene_ontology = OBO(args.obo)
logger.info('Populating gene associations')
if args.ass:
gene_ontology.populate_annotations(args.ass,
gene_col=args.gcol,
term_col=args.term_col)
elif args.gmt:
gmt = GMT(args.gmt)
gene_ontology.populate_annotations_from_gmt(gmt)
else:
sys.stderr.write(
"--Provide gene annotations from an association file or a GMT file")
exit()
if args.pub_filter:
pub_counts = defaultdict(set)
for (term_id, term) in gene_ontology.go_terms.iteritems():
if term.namespace != args.nspace:
continue
for a in term.annotations:
pub_counts[a.ref].add((term, a))
for (ref, annots) in pub_counts.iteritems():
if len(annots) > 50:
logger.info('Removing %i annotations from: %s', ref, len(annots))
if args.ubiq_genes:
ubiq = set()
with open(args.ubiq_genes) as f:
for l in f.readlines():
ubiq.add(l.strip())
logger.info('Total ubiquitous genes: %i', len(ubiq))
onto = None
if args.tissue_onto:
onto = OBO(args.tissue_onto)
tissue_genes = GMT(args.tissue_genes)
if onto:
onto.populate_annotations_from_gmt(tissue_genes)
onto.propagate()
tissue_genes = onto.as_gmt()
with open(args.pos) as f:
edge_lines = f.readlines()
tissue_std_edges = defaultdict(dict)
for line in edge_lines:
g1, g2, std = line.strip().split()[:3]
edge = frozenset([g1, g2])
# Skip edges where both genes are ubiquitous
if ubiq and len(edge & ubiq) == 2:
continue