def test_GO_annotate_genelist2_vs_enriched(self):
input_lines = open(pkg_file(__name__, 'data/genelist2.tsv'),
'r').read()
req = dict(job_req, **{'paste_data': input_lines})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
enrich_job = GOnetSubmission.objects.latest('submit_time')
df = enrich_job.enrich_res_df
enriched_terms = df[df['q'] < enrich_job.qvalue]['term']
custom_annotation = '\n'.join(enriched_terms)
req = dict(
job_req, **{
'paste_data': input_lines,
'analysis_type': 'annot',
'slim': 'custom',
'custom_terms': custom_annotation
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
annot_job = GOnetSubmission.objects.latest('submit_time')
G_enrich = cyjs.cyjs2nx(json.loads(enrich_job.network))
G_annot = cyjs.cyjs2nx(json.loads(annot_job.network))
self.assertSetEqual(set(G_enrich.nodes), set(G_annot.nodes))
self.assertSetEqual(set(G_enrich.edges), set(G_annot.edges))
def test_GO_enrichment_qval01_celcomp(self):
input_lines = open(pkg_file(__name__, 'data/genelist2.tsv'), 'r').read()
req = dict(job_req, **{'paste_data':input_lines, 'qvalue':0.01,
'namespace':'cellular_component'})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
net_dict = json.loads(sn.network)
G = cyjs.cyjs2nx(net_dict)
enriched = set(filter(lambda n: n.startswith('GO:'), G.nodes()))
for term in ['GO:0000786', 'GO:0044815', 'GO:0032993', 'GO:0000785']:
self.assertIn(term, enriched)
def test_resolution(self):
# CMKBR7 is synonym of CCR7
# O14804 is PNR
input_lines = '\n'.join([
"PNR", "KIAA0457", "ELDF10", "CCR7", "AIM", "O14804", "CMKBR7",
"Q9UNH8", "P28068", "FOO"
])
req = dict(job_req, **{'paste_data': input_lines})
URL = urls.reverse('GOnet-submit-form')
resp = c.post(URL, req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
res_d = sn.parsed_data['submit_name'].to_dict()
self.assertDictEqual(
res_d, {
'O75787': 'ELDF10',
'P32248': 'CCR7',
'Q9NRI5': 'KIAA0457',
'O14804': 'PNR',
'_00000': 'O14804',
'_00001': 'CMKBR7',
'Q9UNH8': 'Q9UNH8',
'P28068': 'P28068',
'P26358': 'AIM',
'_00002': 'FOO'
})
self.assertEqual(sn.parsed_data.loc['_00000', 'duplicate_of'],
'O14804')
#Test id mapping response
idmap_resp = c.get(
urls.reverse('GOnet-input-idmap', args=(str(sn.id), )))
b = io.StringIO()
b.write(idmap_resp.content.decode())
b.seek(0)
res = pd.read_csv(b, sep='\t', index_col=0)
self.assertEqual(res.loc['CMKBR7', 'Notes'], 'same as CCR7')
self.assertEqual(res.loc['O14804', 'Notes'], 'same as PNR')
self.assertEqual(res.loc['AIM', 'Description'],
'DNA (cytosine-5)-methyltransferase 1')
self.assertIn('ambiguous', res.loc['PNR', 'Notes'])
self.assertIn('not recognized', res.loc['FOO', 'Notes'])
#Test tricky Ensembl IDs
self.assertEqual(res.loc['P28068', 'Ensembl_ID'], 'ENSG00000242574')
# Check graph attributes
G = cyjs.cyjs2nx(json.loads(sn.network))
self.assertEqual(G.node['_00002']['data']['identified'], False)
self.assertEqual(G.node['P32248']['data']['identified'], True)
self.assertEqual(G.node['O14804']['data']['ambiguous'], True)
def test_GO_annotate_genelist1(self):
input_lines = open(pkg_file(__name__, 'data/genelist1.lst'),
'r').read()
req = dict(
job_req, **{
'paste_data': input_lines,
'analysis_type': 'annot',
'slim': 'goslim_immunol'
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
net_dict = json.loads(sn.network)
G = cyjs.cyjs2nx(net_dict)
self.assertTrue(G.has_edge('GO:0042254', 'Q9HC36'))
def test_GO_annotate_genelist2(self):
input_lines = open(pkg_file(__name__, 'data/genelist2.tsv'),
'r').read()
custom_annotation = open(
pkg_file(__name__, 'data/custom_annotation.txt'), 'r').read()
req = dict(
job_req, **{
'paste_data': input_lines,
'analysis_type': 'annot',
'slim': 'custom',
'custom_terms': custom_annotation
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
self.assertEqual(resp.status_code, 200)
sn = GOnetSubmission.objects.latest('submit_time')
net_dict = json.loads(sn.network)
G = cyjs.cyjs2nx(net_dict)
self.assertListEqual(list(G.predecessors('P29376')), ['GO:0071300'])
self.assertListEqual(list(G.predecessors('Q5TBA9')), ['GO:0016043'])
self.assertListEqual(list(G.predecessors('P16403')), ['GO:0065003'])
# Test node GO:0071300 (cellular response to retinoic acid)
n = list(
filter(lambda n: n['data']['id'] == 'GO:0071300',
net_dict['elements']['nodes']))[0]
self.assertEqual(n['data']['tot_gn'],
len(O.get_attr('GO:0071300', 'human')))
# Test CSV response
csv_resp = c.get(urls.reverse('GOnet-csv-res', args=(str(sn.id), )))
b = io.StringIO()
b.write(csv_resp.content.decode())
b.seek(0)
res = pd.read_csv(b, sep=',', index_col=1)
self.assertIn('LTK', res.loc['GO:0032526', 'Genes'])
# Test TXT response
txt_resp = c.get(urls.reverse('GOnet-txt-res', args=(str(sn.id), )))
b = io.StringIO()
b.write(txt_resp.content.decode())
b.seek(0)
line_found = False
for line in b:
if line.strip().startswith('GO:0032526'):
self.assertIn('LTK', line)
line_found = True
break
self.assertTrue(line_found)
def test_GO_annotate_genelist2(self):
input_lines = open(pkg_file(__name__, 'data/genelist2.tsv'),
'r').read()
input_data_df = pd.read_csv(pkg_file(__name__, 'data/genelist2.tsv'),
sep='\t',
header=None)
req = dict(
job_req, **{
'paste_data': input_lines,
'analysis_type': 'annot',
'slim': 'goslim_immunol'
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
self.assertEqual(resp.status_code, 200)
sn = GOnetSubmission.objects.latest('submit_time')
net = json.loads(sn.network)
G = cyjs.cyjs2nx(net)
self.assertTrue(G.has_edge('GO:0007165', 'P29376'))
# Test recognition of user-supplied contrast values
gene_nodes = filter(lambda n: not n['data']['name'].startswith('GO:'),
net['elements']['nodes'])
gene_nodes = list(gene_nodes)
self.assertEqual(len(list(filter(lambda node: float(node['data']['expr:user_supplied'])>0, gene_nodes))), \
np.sum(input_data_df[1]>0) - 1 ) #-1 for HIST1H2AM
self.assertEqual(len(list(filter(lambda node: float(node['data']['expr:user_supplied'])<0, gene_nodes))), \
np.sum(input_data_df[1]<0))
#Test CSV response
csv_resp = c.get(urls.reverse('GOnet-csv-res', args=(str(sn.id), )))
res = io.StringIO()
res.write(csv_resp.content.decode())
res.seek(0)
res_df = pd.read_csv(res, sep=',', index_col=0)
self.assertIn('GO:0007165', set(res_df['GO_term_ID']))
self.assertEqual(res_df.index[0], 1)
#Test TXT response
txt_resp = c.get(urls.reverse('GOnet-txt-res', args=(str(sn.id), )))
res = io.StringIO()
res.write(txt_resp.content.decode())
res.seek(0)
goterms = set()
for line in res:
goterms.add(line.split()[0])
self.assertIn('GO:0007165', goterms)
def test_GO_annotate_genelist8_large(self):
input_lines = open(pkg_file(__name__, 'data/genelist8.tsv'),
'r').read().split('\n')
input_str = '\n'.join(input_lines[:2500])
req = dict(
job_req, **{
'paste_data': input_str,
'analysis_type': 'annot',
'slim': 'goslim_generic'
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
G = cyjs.cyjs2nx(json.loads(sn.network))
self.assertEqual(G.node['Q8TCT6']['data']['nodesymbol'], 'SPPL3')
#Check expression values (DICE-DB)
resp = c.get(
urls.reverse('GOnet-get-expression',
kwargs={
'jobid': str(sn.id),
'celltype': 'DICE-CD4 T cell (stim)'
}))
expr_vals = json.loads(resp.content.decode())
self.assertAlmostEqual(expr_vals['A1XBS5'], 0.09010716525639434)
# Test TXT response
txt_resp = c.get(urls.reverse('GOnet-txt-res', args=(str(sn.id), )))
b = io.StringIO()
b.write(txt_resp.content.decode())
b.seek(0)
# Test CSV response
csv_resp = c.get(urls.reverse('GOnet-csv-res', args=(str(sn.id), )))
b = io.StringIO()
b.write(csv_resp.content.decode())
b.seek(0)
res = pd.read_csv(b, sep=',', index_col=0)
def test_GO_annot_mouse_genes(self):
input_lines = open(pkg_file(__name__, 'data/genelist7.txt'),
'r').read()
req = dict(
job_req, **{
'paste_data': input_lines,
'analysis_type': 'annot',
'slim': 'goslim_generic',
'organism': 'mouse'
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
net_dict = json.loads(sn.network)
G = cyjs.cyjs2nx(net_dict)
self.assertIn('GO:0006950', G.predecessors('MGI:1351618'))
self.assertEqual(len(G.node['GO:0048856']['data']['xgenes']), 35)
self.assertEqual(G.node['MGI:1351618']['data']['ensembl_id'],
'ENSMUSG00000014905')
self.assertEqual(G.node['MGI:1351618']['data']['uniprot_id'], 'Q9QYI6')
self.assertEqual(G.node['MGI:1351618']['data']['mgi_id'],
'MGI:1351618')
self.assertEqual(G.node['MGI:1351618']['data']['desc'],
'DnaJ heat shock protein family (Hsp40) member B9')
def test_GO_annot_goslim_generic(self):
input_lines = open(pkg_file(__name__, 'data/genelist2.tsv'),
'r').read()
req = dict(
job_req, **{
'paste_data': input_lines,
'analysis_type': 'annot',
'slim': 'goslim_generic',
'namespace': 'cellular_component'
})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
G = cyjs.cyjs2nx(json.loads(sn.network))
self.assertTrue(G.has_edge('GO:0005886',
'P32248')) # CCR7 in plasma membrane
self.assertEqual(len(G.node['O60282']['data']['slimterms']), 7)
ids = list(
filter(lambda n: G.node[n]['data']['nodesymbol'] == 'ZNF761',
G.nodes()))
self.assertEqual(len(ids), 1)
znf_node = G.node[ids[0]]
self.assertEqual(len(znf_node['data']['slimterms']), 4)
def test_GO_enrichment_default(self):
input_lines = open(pkg_file(__name__, 'data/genelist2.tsv'), 'r').read()
req = dict(job_req, **{'paste_data':input_lines})
resp = c.post(urls.reverse('GOnet-submit-form'), req, follow=True)
sn = GOnetSubmission.objects.latest('submit_time')
self.assertIn('GO:0006334', set(sn.enrich_res_df.query('q<0.05')['term']))
net_dict = json.loads(sn.network)
G_case = cyjs.cyjs2nx(net_dict)
# Test CSV response
csv_resp = c.get(urls.reverse('GOnet-csv-res', args=(str(sn.id),)))
b = io.StringIO(); b.write(csv_resp.content.decode()); b.seek(0)
res = pd.read_csv(b, sep=',', index_col=0)
self.assertEqual(res.index[0], 1)
self.assertListEqual(list(res.columns), ['GO_term_ID', 'GO_term_def', 'P',
'P_FDR_adj', 'NofGenes', 'Genes'])
res.set_index('GO_term_ID', inplace=True)
self.assertEqual(res.index.name, 'GO_term_ID')
self.assertIn('GO:0006334', set(res.index))
self.assertEqual(res.loc['GO:2000520', 'Genes'], "CCR7|HAVCR2")
# Test TXT response
txt_resp = c.get(urls.reverse('GOnet-txt-res', args=(str(sn.id),)))
b = io.StringIO()
b.write(txt_resp.content.decode())
b.seek(0)
enrichterms = []
for line in b:
enrichterms.append(line.split()[0])
self.assertIn('GO:0006334', set(enrichterms))
# Test specific terms
t_edge_name = 'GO:0006334 (interacts with) HIST1H2BC'
e = list(filter(lambda e: e['data']['name']==t_edge_name,
net_dict['elements']['edges']))[0]
self.assertDictEqual(e['data']['specific_terms'],
{'GO:0006334': {'refs': ['PMID:21873635',
'PMID:422550', 'PMID:9119399'],
'specific_term_name': 'nucleosome assembly'}})
# Test relation
# "histone H3-K27 trimethylation" is_a "histone H3-K27 methylation"
t_edge_name = 'GO:0070734 (interacts with) GO:0098532'
e = list(filter(lambda e: e['data']['name']==t_edge_name,
net_dict['elements']['edges']))[0]
self.assertEqual(e['data']['relation'], 'is_a')
# Test node ABCB1
n = list(filter(lambda n: n['data']['nodesymbol']=='ABCB1', net_dict['elements']['nodes']))[0]
#self.assertAlmostEqual(n['data']['val'], 0.60217044443096)
self.assertAlmostEqual(n['data']['expr:user_supplied'], 0.6021704444)
self.assertEqual(n['data']['ensembl_id'], 'ENSG00000085563')
# Test node GO:0098532 (histone H3-K27 trimethylation)
n = list(filter(lambda n: n['data']['id']=='GO:0098532', net_dict['elements']['nodes']))[0]
self.assertLess(n['data']['P'], 8.0e-07)
# Should be rather significant
self.assertLess(n['data']['Padj'], 0.01)
self.assertEqual(n['data']['tot_gn'], 6)
# Test resolved attribute
n = list(filter(lambda e: e['data']['nodesymbol']=='LPPR2', net_dict['elements']['nodes']))[0]
self.assertEqual(n['data']['uniprot_id'], 'Q96GM1')
self.assertEqual(n['data']['ensembl_id'], 'ENSG00000105520')
self.assertEqual(n['data']['desc'], 'Phospholipid phosphatase-related protein type 2')
self.assertEqual(n['data']['primname'], 'PLPPR2')
n = list(filter(lambda e: e['data']['nodesymbol']=='LTC4S', net_dict['elements']['nodes']))[0]
self.assertEqual(n['data']['uniprot_id'], 'Q16873')
self.assertEqual(n['data']['ensembl_id'], 'ENSG00000213316')
#Check expression values (protein atlas)
resp = c.get(urls.reverse('GOnet-get-expression',
kwargs={'jobid':str(sn.id), 'celltype':'HPA-adipose tissue'}))
expr_vals = json.loads(resp.content.decode())
self.assertEqual(expr_vals['O14503'], 170.0)
#Check expression values (DICE-DB)
resp = c.get(urls.reverse('GOnet-get-expression',
kwargs={'jobid':str(sn.id), 'celltype':'DICE-Th1Th17'}))
expr_vals = json.loads(resp.content.decode())
self.assertAlmostEqual(expr_vals['P10721'], 9.893615)
#Test id mapping response
idmap_resp = c.get(urls.reverse('GOnet-input-idmap', args=(str(sn.id),)))
b = io.StringIO(); b.write(idmap_resp.content.decode()); b.seek(0)
res = pd.read_csv(b, sep='\t', index_col=0)
self.assertEqual(res.loc['HIST1H2AM', 'Notes'], 'same as HIST1H2AG')
self.assertEqual(res.loc['LPPR2', 'Preferred_name'], 'PLPPR2')