def test_list_techniques(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init(epidb)
res = epidb.add_technique("illumina 450k", "illumina system",
{"price": "low"}, self.admin_key)
self.assertSuccess(res)
res = epidb.add_technique(
"RRBS", "Reduced representation bisulfite sequencing", {},
self.admin_key)
self.assertSuccess(res)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertSuccess(res)
res, techniques = epidb.list_techniques(self.admin_key)
self.assertSuccess(res, techniques)
self.assertEqual(len(techniques), 3)
technique_names = [x[1] for x in techniques]
self.assertTrue("illumina 450k" in technique_names)
self.assertTrue("RRBS" in technique_names)
self.assertTrue("ChIP-seq" in technique_names)
def test_duplicate(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init(epidb)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertSuccess(res)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertFailure(res)
def test_remove_technique(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init_base(epidb)
sample_id = self.sample_ids[0]
regions_data = helpers.load_bed("hg19_chr1_1")
format = data_info.EXPERIMENTS["hg19_chr1_1"]["format"]
(res, add_technique_id) = epidb.add_technique(
"Hypster Technique", "I knew this technique before everybody else",
{}, self.admin_key)
self.assertSuccess(res, add_technique_id)
# adding two experiments with the same data should work
res, eid = epidb.add_experiment("test_exp1", "hg19", "Methylation",
sample_id, "Hypster Technique",
"ENCODE", "desc1", regions_data,
format, None, self.admin_key)
self.assertSuccess(res)
res = epidb.remove(add_technique_id, self.admin_key)
self.assertFailure(res)
res = epidb.remove(eid, self.admin_key)
self.assertSuccess(res)
res = epidb.remove(add_technique_id, self.admin_key)
self.assertSuccess(res)
def test_add_techniques(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init(epidb)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertSuccess(res)
res, info = epidb.info(res[1], self.admin_key)
self.assertSuccess(res, info)
self.assertEquals(info[0]["name"], 'ChIP-seq')
self.assertEquals(info[0]["description"], "ChIP-sequencing")
def test_list_similar_techniques(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init(epidb)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertSuccess(res)
res, similar_techniques = epidb.list_similar_techniques(
"chipseq", self.admin_key)
self.assertSuccess(res, similar_techniques)
self.assertEquals(similar_techniques[0][1], 'ChIP-seq')
def test_empty_result(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init(epidb)
s, g1 = epidb.add_genome("hg19", "", "chr1 10000\nchr2 30000",
self.admin_key)
self.assertSuccess(s, g1)
s, t1 = epidb.add_technique("chip-seq", "", None, self.admin_key)
self.assertSuccess(s, t1)
(s, q) = epidb.select_regions(None, "hg19", None, None, "chip-seq",
None, None, 1000, 2000, self.admin_key)
self.assertSuccess(s, q)
(s, req) = epidb.get_regions(q, "CHROMOSOME,START,END", self.admin_key)
self.assertSuccess(s, q)
rs = self.get_regions_request(req)
self.assertEquals(rs, "")
def test_enriochment_overlap(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init_base(epidb)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertSuccess(res)
sample_id = self.sample_ids[0]
broad_peak_format = ",".join([
"CHROMOSOME",
"START",
"END",
"NAME",
"SCORE",
"STRAND",
"SIGNAL_VALUE",
"P_VALUE",
"Q_VALUE",
])
with open("data/wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed", 'r') as f:
file_data = f.read()
(res, exp) = epidb.add_experiment(
"wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed", "hg19",
"H3k4me3", "s1", "ChIPseq", "ENCODE",
"wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed from ENCODE",
file_data, broad_peak_format, None, self.admin_key)
self.assertSuccess(res, exp)
res, q_exp = epidb.select_experiments(
"wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed", "chr1", None,
None, self.admin_key)
self.assertSuccess(res, q_exp)
## Testting with all overlaps
res, q_tiling = epidb.tiling_regions(10000, "hg19", None,
self.admin_key)
res, r_id = epidb.enrich_regions_overlap(
q_exp, q_tiling,
{"H3k4me3": ["wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed"]},
"hg19", self.admin_key)
result = self.get_regions_request(r_id)
self.assertEquals(result['enrichment']['results'][0]['p_value_log'],
float('Inf'))
self.assertEquals(
result, {
'enrichment': {
'count_query_regions':
3074,
'count_universe_regions':
313669,
'results': [{
'c': 0,
'b': 30861,
'description': '',
'p_value_log': float('Inf'),
'experiment_size': 33270,
'database_name': 'H3k4me3',
'max_rank': 1,
'support_rank': 1,
'dataset':
'wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed',
'biosource': 'K562',
'odd_rank': 1,
'odds_ratio': float('Inf'),
'epigenetic_mark': 'H3k4me3',
'mean_rank': 1.0,
'log_rank': 1,
'support': 3066,
'd': 279742,
'msg': '',
'error': False
}]
}
})
## Testting without any overlap
_, q_input = epidb.input_regions("hg19", "chr1\t1\t2", self.admin_key)
res, r_id = epidb.enrich_regions_overlap(
q_input, q_tiling,
{"H3k4me3": ["wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed"]},
"hg19", self.admin_key)
result = self.get_regions_request(r_id)
self.assertEquals(result['enrichment']['results'][0]['p_value_log'],
0.0)
self.assertEquals(
{
'enrichment': {
'count_query_regions':
1,
'count_universe_regions':
313669,
'results': [{
'c': 1,
'b': 33927,
'description': '',
'p_value_log': 0.0,
'experiment_size': 33270,
'database_name': 'H3k4me3',
'max_rank': 1,
'support_rank': 1,
'dataset':
'wgEncodeBroadHistoneH1hescH3k4me3StdPk.bed',
'biosource': 'K562',
'odd_rank': 1,
'odds_ratio': 0.0,
'epigenetic_mark': 'H3k4me3',
'mean_rank': 1.0,
'log_rank': 1,
'support': 0,
'd': 279741,
'error': False,
'msg': ''
}]
}
}, result)
def test_aggregation_bug_on_normalfields_better_data(self):
epidb = DeepBlueClient(address="localhost", port=31415)
self.init_base(epidb)
res = epidb.add_technique("ChIP-seq", "ChIP-sequencing", {},
self.admin_key)
self.assertSuccess(res)
sample_id = self.sample_ids[0]
broad_peak_format = ",".join([
"CHROMOSOME",
"START",
"END",
"NAME",
"SCORE",
"STRAND",
"SIGNAL_VALUE",
"P_VALUE",
"Q_VALUE",
])
with open("data/wgEncodeBroadHistoneH1hescH3k27me3StdPk.bed",
'r') as f:
file_data = f.read()
(res, a_1) = epidb.add_experiment(
"wgEncodeBroadHistoneH1hescH3k27me3StdPk.bed", "hg19",
"H3k4me3", "s1", "ChIPseq", "ENCODE",
"wgEncodeBroadHistoneH1hescH3k27me3StdPk.bed from ENCODE",
file_data, broad_peak_format, None, self.admin_key)
self.assertSuccess(res, a_1)
res, q_exp = epidb.select_experiments(
"wgEncodeBroadHistoneH1hescH3k27me3StdPk.bed", "chr1", None,
None, self.admin_key)
self.assertSuccess(res, q_exp)
# Insert annotation
cpg_island = ",".join([
"CHROMOSOME", "START", "END", "NAME", "LENGTH", "NUM_CPG",
"NUM_GC", "PER_CPG", "PER_CG", "OBS_EXP"
])
with open("data/cpgIslandExtFull.txt", 'r') as f:
file_data = f.read()
(res, a_1) = epidb.add_annotation("Cpg Islands", "hg19",
"Complete CpG islands",
file_data, cpg_island, None,
self.admin_key)
self.assertSuccess(res, a_1)
res, q_cgi = epidb.select_annotations("Cpg Islands", "hg19", None,
None, None, self.admin_key)
self.assertSuccess(res, q_cgi)
res, qid_3 = epidb.aggregate(q_exp, q_cgi, "SCORE", self.admin_key)
self.assertSuccess(res, qid_3)
res, chr_sizes = epidb.select_annotations("Chromosomes size for hg19",
"hg19", None, None, None,
self.admin_key)
res, qid_5 = epidb.aggregate(qid_3, chr_sizes, "@AGG.COUNT",
self.admin_key)
status, req = epidb.get_regions(
qid_5,
"CHROMOSOME,@AGG.MIN,@AGG.MAX,@AGG.MEDIAN,@AGG.MEAN,@AGG.VAR,@AGG.SD,@AGG.COUNT",
self.admin_key)
rs = self.get_regions_request(req)
expected = 'chr1\t0.0000\t8.0000\t0.0000\t0.5142\t0.5301\t0.7281\t2462'
self.assertEquals(rs, expected)
