def generate_annotated_hpo_network(obo_file,
disease_to_phenotype_file,
annotations_file=None,
ages_distribution_file=None):
hpo_network = load_network(obo_file)
alt2prim = generate_alternate_ids(hpo_network)
# load phenotypes to diseases associations
(
disease_records,
phenotype_to_diseases,
) = load_d2p(disease_to_phenotype_file, hpo_network, alt2prim)
# load hpo network
hpo_network = annotate(
hpo_network,
phenotype_to_diseases,
len(disease_records),
alt2prim,
annotations_file=annotations_file,
ages_distribution_file=ages_distribution_file,
)
return hpo_network, alt2prim, disease_records
def test_ic_custom(self):
"""Calculate the information content of a phenotype when multiple annotations are present"""
custom_annotation_file = os.path.join(self.parent_dir, 'data/test.score-long.txt')
hpo_network = load_network(self.obo_file)
hpo_network = annotate(hpo_network, self.phenotype_to_diseases, self.num_diseases_annotated, self.alt2prim,
annotations_file=custom_annotation_file)
self.assertAlmostEqual(hpo_network.nodes[self.hpo_id]['ic'], 6.38, 1)
def test_annotate_network(self):
hpo_network = load_network(self.obo_file)
alt2prim = generate_alternate_ids(hpo_network)
# load phenotypes to diseases associations
disease_to_phenotype_file = os.path.join(self.parent_dir,
'data/phenotype.hpoa')
disease_records, phenotype_to_diseases = load_d2p(
disease_to_phenotype_file, hpo_network, alt2prim)
num_diseases_annotated = len(disease_records)
hpo_network = annotate(hpo_network, phenotype_to_diseases,
num_diseases_annotated, alt2prim)
self.assertAlmostEqual(hpo_network.nodes['HP:0010863']['ic'], 5.69, 2)
def setUpClass(cls):
# parent dir
cls.parent_dir = os.path.dirname(os.path.realpath(__file__))
# load and process the network
cls.obo_file = os.path.join(cls.parent_dir, 'data/hp.obo')
cls.hpo_network = load_network(cls.obo_file)
cls.alt2prim = generate_alternate_ids(cls.hpo_network)
# load phenotypes to genes associations
cls.disease_to_phenotype_file = os.path.join(cls.parent_dir, 'data/phenotype.hpoa')
cls.disease_records, cls.phenotype_to_diseases = load_d2p(cls.disease_to_phenotype_file, cls.hpo_network, cls.alt2prim)
cls.num_diseases_annotated = len(cls.disease_records)
cls.hpo_network = annotate(cls.hpo_network, cls.phenotype_to_diseases, cls.num_diseases_annotated, cls.alt2prim)
cls.hpo_id = 'HP:0010863'
cls.disease_to_phenotype_output_file = os.path.join(cls.parent_dir, 'data/phenotype.noparents.hpoa')
def setUp(cls):
# parent dir
cls.parent_dir = os.path.dirname(os.path.realpath(__file__))
# load and process the network
cls.obo_file = os.path.join(cls.parent_dir, 'data/hp.obo')
cls.hpo_network = load_network(cls.obo_file)
cls.alt2prim = generate_alternate_ids(cls.hpo_network)
cls.ages_distribution_file = os.path.join(cls.parent_dir,
'data/phenotype_age.tsv')
# load phenotypes to genes associations
cls.disease_to_phenotype_file = os.path.join(cls.parent_dir,
'data/phenotype.hpoa')
cls.disease_records, cls.phenotype_to_diseases = load_d2p(
cls.disease_to_phenotype_file, cls.hpo_network, cls.alt2prim)
cls.num_diseases_annotated = len(cls.disease_records)
cls.hpo_network = annotate(cls.hpo_network, cls.phenotype_to_diseases,
cls.num_diseases_annotated, cls.alt2prim)
# create instance the scorer class
cls.scorer = Scorer(cls.hpo_network, min_score_mask=None)
def test_load_network(self):
hpo_network = load_network(self.obo_file)
self.assertEqual(len(hpo_network), 28)
