def test_indexing(self):
def subindex_length(x):
return (
len(x.keys()),
len(x._indicies.keys()),
len(x._names.keys())
)
self.assertEqual(
subindex_length(Gene),
(0, 0, 0)
)
_ = Gene(hgncid=1, symbol='EZH1')
self.assertEqual(
subindex_length(Gene),
(1, 1, 1)
)
_ = Gene(hgncid=2, symbol='EZH2')
self.assertEqual(
subindex_length(Gene),
(2, 2, 2)
)
Gene.clear()
self.assertEqual(
subindex_length(Gene),
(0, 0, 0)
)
def test_json(self):
g = Gene(hgncid=1, symbol='Foo')
self.assertEqual(
g.toJSON(),
{'id': 1, 'name': 'Foo', 'symbol': 'Foo'}
)
self.assertEqual(
g.toJSON(verbose=True),
{'id': 1, 'name': 'Foo', 'symbol': 'Foo', 'hpo': set()}
)
def make_ontology_with_annotation() -> pyhpo.ontology.OntologyClass:
"""
Generates the following Ontology
- HP:0001
- HP:0011
- HP:0021 <- Gene/Disease 0 / Negative Disease 2
- HP:0031(*)
- HP:0012 <- Gene/Disease 1
- HP:0031(*)
- HP:0041
- HP:0013
"""
Gene.clear()
Omim.clear()
Orpha.clear()
Decipher.clear()
items = make_terms()
terms = Ontology(from_obo_file=False)
for item in items:
terms._append(item)
terms._connect_all()
genes = make_genes(2)
genes[0].hpo.add(21)
genes[1].hpo.add(12)
omim = make_omim(3)
omim[0].hpo.add(21)
omim[1].hpo.add(12)
omim[2].negative_hpo.add(21)
decipher = make_decipher(3)
decipher[0].hpo.add(21)
decipher[1].hpo.add(12)
decipher[2].negative_hpo.add(21)
orpha = make_orpha(3)
orpha[0].hpo.add(21)
orpha[1].hpo.add(12)
orpha[2].negative_hpo.add(21)
ds._add_omim_to_ontology(terms)
ds._add_decipher_to_ontology(terms)
ds._add_orpha_to_ontology(terms)
_add_genes_to_ontology(terms)
return terms
def _parse_phenotype_to_gene_file(path: str) -> None:
filename = os.path.join(path, FILENAME)
with open(filename) as fh:
reader = csv.reader(remove_outcommented_rows(fh), delimiter='\t')
for cols in reader:
gene = Gene(hgncid=int(cols[HGNC_ID]), symbol=cols[GENE_SYMBOL])
gene.hpo.add(id_from_string(cols[HPO_ID]))
def test_set_with_one_annotation(self):
ci = HPOSet([self.ontology[11], self.ontology[13], self.ontology[31]])
assert len(ci.all_genes()) == 1
assert len(ci.omim_diseases()) == 1
assert len(ci.orpha_diseases()) == 1
assert len(ci.decipher_diseases()) == 1
assert ci.all_genes() == set([Gene.get(0)])
assert ci.omim_diseases() == set([Omim.get(0)])
assert ci.orpha_diseases() == set([Orpha.get(0)])
assert ci.decipher_diseases() == set([Decipher.get(0)])
def test_get_gene(self):
g1 = Gene(hgncid=1, symbol='EZH1')
g2 = Gene(hgncid=2, symbol='EZH2')
self.assertEqual(Gene.get(1), g1)
self.assertEqual(Gene.get(2), g2)
self.assertEqual(Gene.get('1'), g1)
self.assertEqual(Gene.get('EZH1'), g1)
self.assertEqual(Gene.get('EZH2'), g2)
self.assertRaises(
KeyError,
lambda: Gene.get('GBA')
)
self.assertRaises(
KeyError,
lambda: Gene.get(12)
)
def test_gene_building(self):
a = Gene(hgncid=1, symbol='EZH2')
self.assertEqual(
a.name,
'EZH2'
)
self.assertEqual(
a.id,
1
)
self.assertEqual(
a.symbol,
'EZH2'
)
self.assertEqual(
a.hpo,
set()
)
def test_singleton_handling(self):
a = Gene(hgncid=1, symbol='EZH2')
# When no name is given, ID is used for comparison
b = Gene(hgncid=1, symbol=None)
# EZH1 does not exist, but ID is used for comparison
c = Gene(hgncid=1, symbol='EZH1')
# EZH2 exists, is used for comparison
d = Gene(hgncid=2, symbol='EZH2')
# EZH2 exists, is used for comparison
e = Gene(hgncid=None, symbol='EZH2')
# EZH1 does not exist. ID does not exist => New Gene
f = Gene(hgncid=2, symbol='EZH1')
# ID is used for comparison
g = Gene(hgncid=2, symbol='EZH2')
# EZH1 is used for comparison
h = Gene(hgncid=1, symbol='EZH1')
# EZH1 is used for comparison
i = Gene(hgncid=None, symbol='EZH1')
self.assertIs(a, b)
self.assertIs(a, c)
self.assertIs(a, d)
self.assertIs(a, e)
self.assertIsNot(a, f)
self.assertIs(a, g)
self.assertIsNot(a, h)
self.assertIs(f, h)
self.assertIs(f, i)
self.assertEqual(a, b)
self.assertEqual(a, c)
self.assertEqual(a, d)
self.assertEqual(a, e)
self.assertEqual(a, g)
self.assertEqual(b, c)
self.assertEqual(b, d)
self.assertEqual(b, e)
self.assertEqual(b, g)
self.assertEqual(c, d)
self.assertEqual(c, e)
self.assertEqual(c, g)
self.assertEqual(d, e)
self.assertEqual(d, g)
self.assertEqual(e, g)
self.assertNotEqual(a, f)
self.assertNotEqual(a, h)
self.assertNotEqual(a, i)
self.assertNotEqual(b, f)
self.assertNotEqual(b, h)
self.assertNotEqual(b, i)
self.assertNotEqual(c, f)
self.assertNotEqual(c, h)
self.assertNotEqual(c, i)
self.assertNotEqual(d, f)
self.assertNotEqual(d, h)
self.assertNotEqual(d, i)
self.assertNotEqual(e, f)
self.assertNotEqual(e, h)
self.assertNotEqual(e, i)
self.assertEqual(f, h)
self.assertEqual(f, i)
self.assertEqual(h, i)
self.assertEqual(
len(Gene.keys()),
2
)
self.assertEqual(
len(Gene.values()),
2
)
self.assertEqual(
len(set(Gene.values())),
2
)
def tearDown(self):
Gene.clear()
def test_string_representation(self):
g = Gene(hgncid=1, symbol='Foo')
self.assertEqual(str(g), 'Foo')
def test_equality(self):
g = Gene(hgncid=1, symbol='Foo')
self.assertEqual(g, 1)
self.assertEqual(g, 'Foo')
def setUp(self):
Gene.clear()
def make_genes(n) -> List[pyhpo.annotations.GeneSingleton]:
# Ensure to remove all items from Gene object
Gene.clear()
return [Gene(hgncid=i, symbol='Gene{}'.format(i)) for i in range(n)]
def tearDown():
Gene.clear()
Omim.clear()
Orpha.clear()
Decipher.clear()
def all_genes() -> Set['GeneSingleton']:
return set(Gene.keys())