def test_preprocess_main_simple(self):
exp, conv_table = preprocess.main('testdata/exp_data-001.csv',
'testdata/conv_table-001.tsv')
self.assertEquals((10, 3), exp.shape)
for i in range(3):
for j in range(3):
self.assertAlmostEquals(exp.values[i, j], -0.8164965809277261)
def test_preprocess_main_simple2(self):
ref_exp = pd.read_csv('testdata/expected_pp_exp_data-002.csv',
index_col=0,
header=0)
exp, conv_table = preprocess.main('testdata/exp_data-002.csv',
'testdata/conv_table-002.tsv')
#exp.to_csv('testdata/expected_pp_exp_data-002.csv')
self.assertTrue(np.isclose(ref_exp, exp).all())
def test_cluster(self):
exp, conv_table = preprocess.main('testdata/exp_data-002.csv',
'testdata/conv_table-002.tsv')
expected_clusters = []
with open('testdata/expected_clusters-002.csv', 'r') as infile:
for line in infile:
expected_clusters.append(line.strip().split(','))
clusters = coexpression.cluster(exp)
self.assertEquals(7, len(clusters))
self.assertEquals(expected_clusters, clusters)
def test_membership_to_incidence(self):
exp_data, conv_table = preprocess.main(
'testdata/ref_exp-000.csv', 'testdata/identifier_mappings.txt')
with open('testdata/ref_ovx_membs-001.json') as infile:
ovx_membs = json.load(infile)
ref_ovx_matrix = pd.read_csv('testdata/ref_ovx_matrix-001.csv',
index_col=0,
header=0)
ovx_matrix = biclusters.membership_to_incidence(ovx_membs, exp_data)
#ovx_matrix.to_csv('testdata/ref_ovx_matrix-001.csv')
self.assertTrue(np.isclose(ref_ovx_matrix, ovx_matrix).all())
def test_get_principal_df(self):
exp_data, conv_table = preprocess.main('testdata/ref_exp-000.csv',
'testdata/identifier_mappings.txt')
with open('testdata/coexpressionDictionary-001.json') as infile:
revised_clusters = json.load(infile)
ref_principal_df = pd.read_csv('testdata/ref_principal_df-001.csv', index_col=0, header=0)
axes = mechinf.get_principal_df(revised_clusters, exp_data,
subkey=None, min_number_genes=1)
#axes.to_csv('testdata/ref_principal_df-001.csv', header=True, index=True)
self.assertTrue(np.isclose(ref_principal_df, axes).all())
def test_enrichment(self):
exp_data, conv_table = preprocess.main('testdata/ref_exp-000.csv',
'testdata/identifier_mappings.txt')
with open('testdata/coexpressionDictionary-001.json') as infile:
revised_clusters = json.load(infile)
axes = pd.read_csv('testdata/ref_principal_df-001.csv', index_col=0, header=0)
database_path = os.path.join('miner2/data', "tfbsdb_tf_to_genes.pkl")
with open('testdata/ref_mechout-001.json') as infile:
ref_mechout = json.load(infile)
mechout = mechinf.enrichment(axes, revised_clusters, exp_data,
correlation_threshold=0.2,
num_cores=5,
database_path=database_path)
#with open('testdata/ref_mechout-001.json', 'w') as outfile:
# json.dump(mechout, outfile)
self.compare_dicts(ref_mechout, mechout)
def test_cluster(self):
ref_exp = pd.read_csv('testdata/expected_pp_exp_data-002.csv',
index_col=0,
header=0)
exp, conv_table = preprocess.main('testdata/exp_data-002.csv',
'testdata/conv_table-002.tsv')
self.assertTrue(np.isclose(ref_exp, exp).all()) # just to make sure
expected_clusters = []
with open('testdata/expected_clusters-002.csv', 'r') as infile:
for line in infile:
expected_clusters.append(line.strip().split(','))
clusters = coexpression.cluster(exp)
#with open('testdata/expected_clusters-002.csv', 'w') as outfile:
# for cluster in clusters:
# outfile.write('%s\n' % (','.join(cluster)))
self.assertEquals(len(expected_clusters), len(clusters))
self.assertEquals(expected_clusters, clusters)