def test_per_sample_sequences_complex(self):
maximum = 2
def bin_f(x):
return x['SequenceID'].rsplit('_', 1)[0]
exp = sorted([('a', {
'SequenceID': 'a_2',
'Sequence': 'AATTGGCC-a2'
}), ('a', {
'SequenceID': 'a_3',
'Sequence': 'AATTGGCC-a3'
}), ('b', {
'SequenceID': 'b_2',
'Sequence': 'AATTGGCC-b2'
}), ('b', {
'SequenceID': 'b_1',
'Sequence': 'AATTGGCC-b1'
}), ('c', {
'SequenceID': 'c_1',
'Sequence': 'AATTGGCC-c1'
}), ('c', {
'SequenceID': 'c_2',
'Sequence': 'AATTGGCC-c2'
})],
key=lambda x: x[0])
obs = isubsample(self.mock_sequence_iter(self.sequences),
maximum,
bin_f=bin_f,
buf_size=1)
self.assertEqual(sorted(obs, key=lambda x: x[0]), exp)
def subsample_dm(distmat, mapping_file, max, category, output):
"""Subsample the distmat to max samples per category value"""
mf = pd.read_csv(mapping_file, '\t', converters=defaultdict(str),
index_col='#SampleID')
id_to_cat = dict(mf[category])
def bin_f(x):
return id_to_cat[x]
dm = read(distmat, into=DistanceMatrix)
dm = dm.filter([id for _, id in isubsample(dm.ids, max, bin_f=bin_f)])
dm.to_file(output)
def test_per_sample_sequences_complex(self):
maximum = 2
bin_f = lambda x: x['SequenceID'].rsplit('_', 1)[0]
exp = sorted([('a', {'SequenceID': 'a_2', 'Sequence': 'AATTGGCC-a2'}),
('a', {'SequenceID': 'a_3', 'Sequence': 'AATTGGCC-a3'}),
('b', {'SequenceID': 'b_2', 'Sequence': 'AATTGGCC-b2'}),
('b', {'SequenceID': 'b_1', 'Sequence': 'AATTGGCC-b1'}),
('c', {'SequenceID': 'c_1', 'Sequence': 'AATTGGCC-c1'}),
('c', {'SequenceID': 'c_2', 'Sequence': 'AATTGGCC-c2'})],
key=lambda x: x[0])
obs = isubsample(self.mock_sequence_iter(self.sequences), maximum,
bin_f=bin_f, buf_size=1)
self.assertEqual(sorted(obs, key=lambda x: x[0]), exp)
def subsample_dm(distmat, mapping_file, max, category, output):
"""Subsample the distmat to max samples per category value"""
mf = pd.read_csv(mapping_file,
'\t',
converters=defaultdict(str),
dtype=str)
mf.set_index('#SampleID', inplace=True)
id_to_cat = dict(mf[category])
def bin_f(x):
return id_to_cat.get(x)
dm = read(distmat, into=DistanceMatrix)
dm = dm.filter([id for _, id in isubsample(dm.ids, max, bin_f=bin_f)])
dm.to_file(output)
def test_isubsample_simple(self):
maximum = 10
def bin_f(x):
return x['SequenceID'].rsplit('_', 1)[0]
# note, the result here is sorted by sequence_id but is in heap order
# by the random values associated to each sequence
exp = sorted([('a', {
'SequenceID': 'a_5',
'Sequence': 'AATTGGCC-a5'
}), ('a', {
'SequenceID': 'a_1',
'Sequence': 'AATTGGCC-a1'
}), ('a', {
'SequenceID': 'a_4',
'Sequence': 'AATTGGCC-a4'
}), ('a', {
'SequenceID': 'a_3',
'Sequence': 'AATTGGCC-a3'
}), ('a', {
'SequenceID': 'a_2',
'Sequence': 'AATTGGCC-a2'
}), ('b', {
'SequenceID': 'b_2',
'Sequence': 'AATTGGCC-b2'
}), ('b', {
'SequenceID': 'b_1',
'Sequence': 'AATTGGCC-b1'
}), ('c', {
'SequenceID': 'c_3',
'Sequence': 'AATTGGCC-c3'
}), ('c', {
'SequenceID': 'c_2',
'Sequence': 'AATTGGCC-c2'
}), ('c', {
'SequenceID': 'c_1',
'Sequence': 'AATTGGCC-c1'
})],
key=lambda x: x[0])
obs = isubsample(self.mock_sequence_iter(self.sequences),
maximum,
bin_f=bin_f)
self.assertEqual(sorted(obs, key=lambda x: x[0]), exp)
def test_per_sample_sequences_min_seqs(self):
maximum = 10
minimum = 3
bin_f = lambda x: x['SequenceID'].rsplit('_', 1)[0]
# note, the result here is sorted by sequence_id but is in heap order
# by the random values associated to each sequence
exp = sorted([('a', {'SequenceID': 'a_5', 'Sequence': 'AATTGGCC-a5'}),
('a', {'SequenceID': 'a_1', 'Sequence': 'AATTGGCC-a1'}),
('a', {'SequenceID': 'a_4', 'Sequence': 'AATTGGCC-a4'}),
('a', {'SequenceID': 'a_3', 'Sequence': 'AATTGGCC-a3'}),
('a', {'SequenceID': 'a_2', 'Sequence': 'AATTGGCC-a2'}),
('c', {'SequenceID': 'c_3', 'Sequence': 'AATTGGCC-c3'}),
('c', {'SequenceID': 'c_2', 'Sequence': 'AATTGGCC-c2'}),
('c', {'SequenceID': 'c_1', 'Sequence': 'AATTGGCC-c1'})],
key=lambda x: x[0])
obs = isubsample(self.mock_sequence_iter(self.sequences), maximum,
minimum, bin_f=bin_f)
self.assertEqual(sorted(obs, key=lambda x: x[0]), exp)
def test_binf_is_none(self):
maximum = 2
items = [1, 2]
exp = [(True, 1), (True, 2)]
obs = isubsample(items, maximum)
self.assertEqual(list(obs), exp)
def test_max_lt_zero(self):
gen = isubsample([1, 2, 3], maximum=-10)
with self.assertRaises(ValueError):
next(gen)
def test_min_gt_max(self):
gen = isubsample([1, 2, 3], maximum=2, minimum=10)
with self.assertRaises(ValueError):
next(gen)
def test_binf_is_none(self):
maximum = 2
items = [1, 2]
exp = [(True, 1), (True, 2)]
obs = isubsample(items, maximum)
self.assertEqual(list(obs), exp)
def test_max_lt_zero(self):
gen = isubsample([1, 2, 3], maximum=-10)
with self.assertRaises(ValueError):
next(gen)
def test_min_gt_max(self):
gen = isubsample([1, 2, 3], maximum=2, minimum=10)
with self.assertRaises(ValueError):
next(gen)
