def test_classify(self):
# simplest gotu workflow
input_fp = join(self.datdir, 'align', 'bowtie2')
samples, files, demux = parse_samples(input_fp)
mapper, chunk = build_mapper()
ranks = ['none']
obs = classify(mapper, files, samples=samples, demux=demux,
ranks=ranks, chunk=chunk)['none']
self.assertEqual(obs['S01']['G000011545'], 48)
self.assertNotIn('G000007145', obs['S02'])
self.assertEqual(obs['S03']['G000009345'], 640)
self.assertEqual(obs['S04']['G000240185'], 4)
self.assertEqual(obs['S05']['G000191145'], 10)
# complex genus/process stratification workflow
input_fp = join(self.datdir, 'align', 'burst')
coords_fp = join(self.datdir, 'function', 'coords.txt.xz')
map_fps = [join(self.datdir, 'function', 'uniref.map.xz'),
join(self.datdir, 'function', 'go', 'process.tsv.xz')]
strata_dir = join(self.datdir, 'output', 'burst.genus.map')
samples, files, demux = parse_samples(input_fp)
tree, rankdic, namedic, root = build_hierarchy(
map_fps=map_fps, map_as_rank=True)
mapper, chunk = build_mapper(coords_fp=coords_fp, overlap=80)
stratmap = parse_strata(strata_dir, samples)
obs = classify(
mapper, files, samples=samples, demux=demux, tree=tree,
rankdic=rankdic, namedic=namedic, root=root, stratmap=stratmap,
chunk=chunk, ranks=['process'])['process']
self.assertEqual(obs['S01'][('Thermus', 'GO:0005978')], 2)
self.assertEqual(obs['S02'][('Bacteroides', 'GO:0006814')], 1)
self.assertEqual(obs['S03'][('Escherichia', 'GO:0006813')], 2)
self.assertEqual(len(obs['S04']), 39)
def test_classify(self):
# simplest gotu workflow
input_path = join(self.datdir, 'align', 'bowtie2')
samples, files, demux = parse_samples(input_path)
mapper = build_mapper()
ranks = ['none']
obs = classify(mapper,
files,
samples=samples,
demux=demux,
ranks=ranks)['none']
self.assertEqual(obs['S01']['G000011545'], 48)
self.assertNotIn('G000007145', obs['S02'])
self.assertEqual(obs['S03']['G000009345'], 640)
self.assertEqual(obs['S04']['G000240185'], 4)
self.assertEqual(obs['S05']['G000191145'], 10)
def test_parse_samples(self):
# file (assuming demultiplexed)
fp = join(self.tmpdir, 'input.fq')
open(fp, 'a').close()
obs = parse_samples(fp)
self.assertIsNone(obs[0])
self.assertListEqual(obs[1], [fp])
self.assertTrue(obs[2])
# confirm demux
obs = parse_samples(fp, demux=True)
self.assertIsNone(obs[0])
self.assertListEqual(obs[1], [fp])
self.assertTrue(obs[2])
# force non-demux
obs = parse_samples(fp, demux=False)
self.assertListEqual(obs[0], ['input'])
self.assertDictEqual(obs[1], {fp: 'input'})
self.assertFalse(obs[2])
# provide correct extension
obs = parse_samples(fp, ext='.fq', demux=False)
self.assertListEqual(obs[0], ['input'])
self.assertDictEqual(obs[1], {fp: 'input'})
# provide wrong extension
with self.assertRaises(ValueError) as ctx:
parse_samples(fp, ext='.fastq', demux=False)
self.assertEqual(str(ctx.exception),
('Filepath and filename extension do not match.'))
# provide Id list (no effect since demux)
obs = parse_samples(fp, samples='input')
self.assertListEqual(obs[0], ['input'])
self.assertListEqual(obs[1], [fp])
# provide correct Id list
obs = parse_samples(fp, samples='input', demux=False)
self.assertListEqual(obs[0], ['input'])
self.assertDictEqual(obs[1], {fp: 'input'})
# provide wrong Id list
with self.assertRaises(ValueError) as ctx:
parse_samples(fp, samples='hello', demux=False)
self.assertEqual(
str(ctx.exception),
('Provided sample IDs and actual files are inconsistent.'))
# directory (assuming per-sample)
obs = parse_samples(self.tmpdir)
self.assertListEqual(obs[0], ['input'])
self.assertDictEqual(obs[1], {fp: 'input'})
self.assertFalse(obs[2])
# directory empty
remove(fp)
with self.assertRaises(ValueError) as ctx:
parse_samples(self.tmpdir)
self.assertEqual(str(ctx.exception),
('No valid file found in directory.'))
# multiple files
for i in range(1, 4):
open(join(self.tmpdir, f'S{i}.sam'), 'a').close()
obs = parse_samples(self.tmpdir)[1]
exp = {join(self.tmpdir, f'S{i}.sam'): f'S{i}' for i in range(1, 4)}
self.assertDictEqual(obs, exp)
# add an irrelevant file
fp = join(self.tmpdir, 'readme.txt')
open(fp, 'a').close()
obs = parse_samples(self.tmpdir)[1]
exp[fp] = 'readme'
self.assertDictEqual(obs, exp)
# specify extension to target alignment files only
del exp[fp]
obs = parse_samples(self.tmpdir, ext='.sam')[1]
self.assertDictEqual(obs, exp)
# specify sample Ids
obs = parse_samples(self.tmpdir, samples='S1,S2,S3')[1]
self.assertDictEqual(obs, exp)
# some samples are not found
remove(fp)
with self.assertRaises(ValueError) as ctx:
parse_samples(self.tmpdir, samples='S1,S2,S4')
self.assertEqual(
str(ctx.exception),
('Provided sample IDs and actual files are inconsistent.'))
# force demux
obs = parse_samples(self.tmpdir, demux=True)
self.assertIsNone(obs[0])
self.assertTrue(obs[2])
exp = [join(self.tmpdir, f'S{i}.sam') for i in range(1, 4)]
self.assertListEqual(obs[1], exp)
# sample Ids are ignored when demux
obs = parse_samples(self.tmpdir, samples='S1,S2,S4', demux=True)
self.assertListEqual(obs[0], ['S1', 'S2', 'S4'])
exp = [join(self.tmpdir, f'S{i}.sam') for i in range(1, 4)]
self.assertListEqual(obs[1], exp)
# sample-to-file map
fp = join(self.tmpdir, 'sample.list')
with open(fp, 'w') as f:
for i in (range(1, 4)):
print(f'S{i}\tS{i}.sam', file=f)
obs = parse_samples(fp)
self.assertListEqual(obs[0], ['S1', 'S2', 'S3'])
exp = {join(self.tmpdir, f'S{i}.sam'): f'S{i}' for i in range(1, 4)}
self.assertDictEqual(obs[1], exp)
# some samples only
obs = parse_samples(fp, samples='S1,S2')
self.assertListEqual(obs[0], ['S1', 'S2'])
# some samples are not found
with self.assertRaises(ValueError) as ctx:
parse_samples(fp, samples='S1,S2,S4')
self.assertEqual(
str(ctx.exception),
('Provided sample IDs and actual files are inconsistent.'))
remove(fp)
# not a valid path
with self.assertRaises(ValueError) as ctx:
parse_samples('im/not/path')
self.assertEqual(str(ctx.exception),
('"im/not/path" is not a valid file or directory.'))