class TestLabels(unittest.TestCase):
def setUp(self):
self.go = OBO('files/test_data/go.obo')
self.go.populate_annotations(GO_URL, remote_location=True)
self.dsrepair_term = self.go.get_term(DSREPAIR_ID)
lines = open('files/go_neg_slim.txt').readlines()
self.slim_terms = set([l.strip() for l in lines])
def testOntoLabelsNeg(self):
ol = OntoLabels(obo=self.go, slim_terms=self.slim_terms)
(pos, neg) = ol.get_labels(DSREPAIR_ID)
self.assertTrue(len(pos) > 0)
self.assertEqual(pos,set(self.dsrepair_term.get_annotated_genes()))
similar_term = self.go.get_term(DNA_REP_ID).get_annotated_genes()
slim_overlap = self.go.get_ancestors(DNA_REP_ID) & \
self.go.get_ancestors(DSREPAIR_ID) & \
self.slim_terms
self.assertTrue(DNA_METAB in slim_overlap)
self.assertTrue(len(neg & set(similar_term)) == 0)
for dterm in self.go.get_descendents(DNA_METAB):
dgenes = set(self.go.get_term(dterm).get_annotated_genes())
self.assertTrue(len(neg & dgenes) == 0)
class TestOBO(unittest.TestCase):
def setUp(self):
self.go = OBO('files/test_data/go.obo')
self.dsrepair_term = self.go.get_term(DSREPAIR_ID)
def test_load_remote(self):
"""Test loading an obo file from a URL"""
self.go = OBO()
self.go.load_obo(GO_URL, remote_location=True)
self.test_load()
def test_load(self):
"""Test an obo is loaded"""
# Check that root biologcal process term is loaded
term = self.go.get_term('GO:0008150')
self.assertTrue(term is not None and term.name == 'biological_process')
self.assertTrue(len(term.parent_of) > 0)
def test_add_annotation(self):
"""Test that adding a gene annotations is correct"""
self.assertTrue(self.dsrepair_term is not None)
self.dsrepair_term.add_annotation('672')
genes = self.dsrepair_term.get_annotated_genes()
self.assertEqual(genes[0], '672')
def test_propagation(self):
"""Test that gene propagation adds annotations to the correct terms"""
self.assertTrue(self.dsrepair_term is not None)
self.dsrepair_term.add_annotation('672')
self.go.propagate()
term_count = 0
for term in self.go.get_termobject_list():
term_count += len(term.get_annotated_genes())
self.assertEqual(term_count, 26)
def test_direct_annotation(self):
"""Test that direct annotation status is preserved"""
self.assertTrue(self.dsrepair_term is not None)
self.go.add_annotation(go_id=DSREPAIR_ID,
gid='672',
ref=None,
direct=True)
self.go.propagate()
direct_count, total = 0, 0
for term in self.go.get_termobject_list():
for a in term.annotations:
if a.direct:
direct_count += 1
self.assertEqual(term, self.dsrepair_term)
total += 1
self.assertEqual(direct_count, 1)
self.assertEqual(total, 26)
def test_ancestors(self):
"""Test that ancestor terms are correct"""
self.assertTrue(self.dsrepair_term is not None)
self.assertEqual(len(self.go.get_ancestors(self.dsrepair_term.go_id)),
25)
def test_parents(self):
"""Test that parent terms are stored correctly"""
parents = self.dsrepair_term.child_of
self.assertEqual(len(parents), 2)
self.assertTrue(GOTerm('GO:0006302') in parents)
self.assertTrue(GOTerm('GO:0000725') in parents)
def test_heads(self):
"""Test that there are only three head nodes, corresponding to:
biological process, molecular function, and cellular component
"""
heads = set(
[term for term in self.go.get_termobject_list() if term.head])
self.assertEqual(len(heads), 3)
self.assertTrue(GOTerm('GO:0008150') in heads)
self.assertTrue(GOTerm('GO:0003674') in heads)
self.assertTrue(GOTerm('GO:0005575') in heads)
def test_obsolete(self):
"""Test that none of the stored terms are obsolete"""
for term in self.go.get_termobject_list():
self.assertTrue(not term.obsolete)
for term in self.go.get_obsolete_terms():
self.assertTrue(term.obsolete)
def test_term_equals(self):
"""Test GOTerm equals method"""
self.assertEqual(self.dsrepair_term, GOTerm(DSREPAIR_ID))
def tearDown(self):
self.go = None
self.dsrepair_term = None
for (tissue, stds) in tissue_std_edges.iteritems():
c3 = tissue_std_edges[tissue]['0']
for c3_edge in list(c3):
# Limit negative edges to positive genes
if len(c3_edge & tissue_all_pos) < 2:
c3.remove(c3_edge)
for (tissue, stds) in tissue_std_edges.iteritems():
c1 = tissue_std_edges[tissue]['1']
c1_g = reduce(frozenset.union, c1)
c1_tspc = c1_g - ubiq
c3 = tissue_std_edges[tissue]['0']
c3_g = reduce(frozenset.union, c3)
c3_tspc = c3_g - ubiq
c2 = within_edges_pos - c1
c4 = within_edges_neg - c3
if onto:
for dtissue in onto.get_descendents(tissue):
if dtissue in tissue_std_edges:
c2 -= tissue_std_edges[dtissue]['1']
c4 -= tissue_std_edges[dtissue]['0']
for atissue in onto.get_ancestors(tissue):
if atissue in tissue_std_edges:
c2 -= tissue_std_edges[atissue]['1']
c4 -= tissue_std_edges[atissue]['0']
print tissue, len(c1_tspc), len(c1), len(c2), len(c3), len(c4)