def go_gene_sets(tax_id: str) -> None:
domain = 'go'
ontology = go.Ontology(filename=f'{data_path}/{domain}/gene_ontology.obo')
annotations = go.Annotations(tax_id,
filename=f'{data_path}/{domain}/{tax_id}.tab',
ontology=ontology)
def to_gene_set(term: go.Term) -> Optional[GeneSet]:
genes = annotations.get_genes_by_go_term(term.id)
if len(genes) > 0:
return GeneSet(
gs_id=term.id,
name=term.name,
genes=set(genes),
hierarchy=('GO', term.namespace),
organism=tax_id,
link=f'http://amigo.geneontology.org/amigo/term/{term.id}')
gene_sets = GeneSets([
gs for gs in [to_gene_set(term) for term in ontology.terms.values()]
if gs is not None
])
for gs_group in gene_sets.split_by_hierarchy():
hierarchy = gs_group.common_hierarchy()
gs_group.to_gmt_file_format(
f'{data_path}/gene_sets/{filename(hierarchy, tax_id)}')
def go_gene_sets(org):
""" Returns gene sets from GO.
"""
ontology = go.Ontology()
annotations = go.Annotations(org, ontology=ontology)
gene_sets = []
for termn, term in ontology.terms.items():
genes = annotations.get_genes_by_go_term(termn)
hier = ('GO', term.namespace)
if len(genes) > 0:
gs = GeneSet(gs_id=termn, name=term.name, genes=genes, hierarchy=hier,
organism=org, link=GO_TERM_LINK.format(termn))
gene_sets.append(gs)
return GeneSets(gene_sets)
def Load(self):
a = self.available_annotations[self.annotation_index]
if self.ontology is None:
self.ontology = go.Ontology()
if a.taxid != self.loaded_annotation_code:
self.annotations = None
gc.collect() # Force run garbage collection
self.annotations = go.Annotations(a.taxid)
self.loaded_annotation_code = a.taxid
count = defaultdict(int)
geneSets = defaultdict(set)
for anno in self.annotations.annotations:
count[anno.evidence] += 1
geneSets[anno.evidence].add(anno.gene_id)
for etype in go.evidenceTypesOrdered:
ecb = self.evidenceCheckBoxDict[etype]
ecb.setEnabled(bool(count[etype]))
ecb.setText(etype + ": %i annots(%i genes)" % (count[etype], len(geneSets[etype])))
from orangecontrib.bioinformatics import go
ontology = go.Ontology()
annotations = go.Annotations("4932", ontology=ontology)
# keys are symbol names, values are Entrez IDs
genes_ids = {'Yta7p': '853186', 'RPN2': '854735', 'RPT2': '851557'}
res = annotations.get_enriched_terms(genes_ids.values())
print(res)
print("Enriched terms:")
for go_id, (genes, p_value, ref) in res.items():
if p_value < 0.05:
print(ontology[go_id].name + " with p-value: %.4f " % p_value +
", ".join(genes))
# And again for slims
annotations.ontology.set_slims_subset('goslim_yeast')
res = annotations.get_enriched_terms(genes_ids.values(), slims_only=True)
print("\n\nEnriched slim terms:")
for go_id, (genes, p_value, _) in res.items():
if p_value < 0.2:
print(ontology[go_id].name + " with p-value: %.4f " % p_value +
", ".join(genes))