1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
|
# ----------------------------------------------------------------------------
# Copyright (c) 2016-2022, QIIME 2 development team.
#
# Distributed under the terms of the Modified BSD License.
#
# The full license is in the file LICENSE, distributed with this software.
# ----------------------------------------------------------------------------
import unittest
import biom
import qiime2
import pandas as pd
import numpy as np
from q2_feature_table import group
class TestGroup(unittest.TestCase):
def test_identity_groups(self):
# These map to the same values as before
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['a', 'b', 'c'], name='foo',
index=pd.Index(['a', 'b', 'c'], name='sampleid')))
feature_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['x', 'y'], name='foo',
index=pd.Index(['x', 'y'], name='featureid')))
table = biom.Table(np.array([[1, 2, 3], [30, 20, 10]]),
sample_ids=sample_mc.to_series().index,
observation_ids=feature_mc.to_series().index)
# Sample x Sum
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(table, result)
# Sample x Mean
result = group(table, axis='sample', metadata=sample_mc,
mode='mean-ceiling')
self.assertEqual(table, result)
# Sample x Median
result = group(table, axis='sample', metadata=sample_mc,
mode='median-ceiling')
self.assertEqual(table, result)
# Feature x Sum
result = group(table, axis='feature', metadata=feature_mc, mode='sum')
self.assertEqual(table, result)
# Feature x Mean
result = group(table, axis='feature', metadata=feature_mc,
mode='mean-ceiling')
self.assertEqual(table, result)
# Feature x Median
result = group(table, axis='feature', metadata=feature_mc,
mode='median-ceiling')
self.assertEqual(table, result)
def test_one_to_one_rename(self):
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['a_new', 'b_new', 'c_new'], name='foo',
index=pd.Index(['a', 'b', 'c'], name='sampleid')))
original_sample_ids = sample_mc.to_series().index
new_sample_ids = list(sample_mc.to_series())
feature_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['x_new', 'y_new'], name='foo',
index=pd.Index(['x', 'y'], name='featureid')))
original_feature_ids = feature_mc.to_series().index
new_feature_ids = list(feature_mc.to_series())
data = np.array([[1, 2, 3], [30, 20, 10]])
table = biom.Table(data, sample_ids=original_sample_ids,
observation_ids=original_feature_ids)
# Sample renames
expected = biom.Table(data, sample_ids=new_sample_ids,
observation_ids=original_feature_ids)
# Sample x Sum
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(expected, result)
# Sample X Mean
result = group(table, axis='sample', metadata=sample_mc,
mode='mean-ceiling')
self.assertEqual(expected, result)
# Sample X Mean
result = group(table, axis='sample', metadata=sample_mc,
mode='median-ceiling')
self.assertEqual(expected, result)
# Feature renames
expected = biom.Table(data, sample_ids=original_sample_ids,
observation_ids=new_feature_ids)
# Feature X Sum
result = group(table, axis='feature', metadata=feature_mc, mode='sum')
self.assertEqual(expected, result)
# Feature X Mean
result = group(table, axis='feature', metadata=feature_mc,
mode='mean-ceiling')
self.assertEqual(expected, result)
# Feature X Median
result = group(table, axis='feature', metadata=feature_mc,
mode='median-ceiling')
self.assertEqual(expected, result)
def test_superset_feature_group(self):
feature_mc = qiime2.CategoricalMetadataColumn(
pd.Series(
['g0', 'g0', 'g1', 'g2', 'g1', 'g2', 'extra'], name='foo',
index=pd.Index(['a', 'b', 'c', 'd', 'e', 'f', 'g'],
name='featureid')))
data = np.array([
[1, 0, 0],
[1, 10, 10],
[0, 0, 100],
[5, 5, 5],
[0, 1, 100],
[7, 8, 9]])
# g is missing on purpose
table = biom.Table(data, sample_ids=['s1', 's2', 's3'],
observation_ids=['a', 'b', 'c', 'd', 'e', 'f'])
expected = biom.Table(
np.array([[2, 10, 10], [0, 1, 200], [12, 13, 14]]),
sample_ids=['s1', 's2', 's3'],
observation_ids=['g0', 'g1', 'g2'])
result = group(table, axis='feature', metadata=feature_mc, mode='sum')
self.assertEqual(expected, result)
def test_superset_sample_group(self):
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['g0', 'g1', 'g2', 'g0', 'g1', 'g2'], name='foo',
index=pd.Index(['s1', 's2', 's3', 's4', 's5', 's6'],
name='sampleid')))
data = np.array([
[0, 1, 2, 3],
[10, 11, 12, 13],
[100, 110, 120, 130]])
table = biom.Table(data, sample_ids=['s1', 's2', 's4', 's5'],
observation_ids=['x', 'y', 'z'])
expected = biom.Table(
np.array([[2, 4], [22, 24], [220, 240]]),
sample_ids=['g0', 'g1'],
observation_ids=['x', 'y', 'z'])
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(expected, result)
def test_missing_feature_ids(self):
feature_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['g0', 'g1', 'g2', 'g1', 'g2', 'extra'], name='foo',
index=pd.Index(['a', 'c', 'd', 'e', 'f', 'g'],
name='featureid')))
data = np.array([
[1, 0, 0],
[1, 10, 10],
[0, 0, 100],
[5, 5, 5],
[0, 1, 100],
[7, 8, 9]])
# g is missing on purpose
table = biom.Table(data, sample_ids=['s1', 's2', 's3'],
observation_ids=['a', 'b', 'c', 'd', 'e', 'f'])
with self.assertRaisesRegex(ValueError, "not present.*'b'"):
group(table, axis='feature', metadata=feature_mc, mode='sum')
def test_missing_sample_ids(self):
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['g0', 'g2', 'g0', 'g2'], name='foo',
index=pd.Index(['s1', 's3', 's4', 's6'],
name='sampleid')))
data = np.array([
[0, 1, 2, 3],
[10, 11, 12, 13],
[100, 110, 120, 130]])
table = biom.Table(data, sample_ids=['s1', 's2', 's4', 's5'],
observation_ids=['x', 'y', 'z'])
with self.assertRaisesRegex(ValueError, 'not present.*s2.*s5') as e:
group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertIn('s2', str(e.exception))
self.assertIn('s5', str(e.exception))
def test_reorder(self):
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['c', 'b', 'a'], name='foo',
index=pd.Index(['c', 'b', 'a'], name='sampleid')))
data = np.array([[1, 2, 3], [30, 20, 10]])
table = biom.Table(data, sample_ids=['a', 'b', 'c'],
observation_ids=['x', 'y'])
expected = biom.Table(np.array([[3, 2, 1], [10, 20, 30]]),
sample_ids=['c', 'b', 'a'],
observation_ids=['x', 'y'])
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(expected, result)
def test_group_to_single_id(self):
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['all_samples', 'all_samples', 'all_samples'],
name='foo', index=pd.Index(['a', 'b', 'c'],
name='sampleid')))
data = np.array([[1, 2, 3], [30, 20, 10]])
table = biom.Table(data, sample_ids=['a', 'b', 'c'],
observation_ids=['x', 'y'])
expected = biom.Table(np.array([[6], [60]]),
sample_ids=['all_samples'],
observation_ids=['x', 'y'])
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(expected, result)
def test_empty_metadata_values(self):
# Trusting that the code is sane enough to not invent a distinction
# between feature and sample metadata where there is none
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['a_new', 'a_new', None], name='foo',
index=pd.Index(['a', 'b', 'c'], name='sampleid')))
sample_ids = sample_mc.to_series().index
data = np.array([[1, 2, 3], [30, 20, 10]])
table = biom.Table(data, sample_ids=sample_ids,
observation_ids=['x', 'y'])
with self.assertRaisesRegex(ValueError, "missing.*value.*'c'"):
group(table, axis='sample', metadata=sample_mc, mode='sum')
nan_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['a_new', float('nan'), 'a_new'], name='foo',
index=pd.Index(['a', 'b', 'c'], name='id')))
with self.assertRaisesRegex(ValueError, "missing.*value.*'b'"):
group(table, axis='sample', metadata=nan_mc, mode='sum')
def test_empty_only_in_superset(self):
# Trusting that the code is sane enough to not invent a distinction
# between feature and sample metadata where there is none
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['a_new', 'a_new', 'b_new', None], name='foo',
index=pd.Index(['a', 'b', 'c', 'd'], name='sampleid')))
data = np.array([[1, 2, 3], [30, 20, 10]])
table = biom.Table(data, sample_ids=['a', 'b', 'c'],
observation_ids=['x', 'y'])
expected = biom.Table(np.array([[2, 3], [25, 10]]),
sample_ids=['a_new', 'b_new'],
observation_ids=['x', 'y'])
result = group(table, axis='sample', metadata=sample_mc,
mode='mean-ceiling')
self.assertEqual(expected, result)
def test_numeric_strings(self):
data = np.array([[1, 2, 3], [30, 20, 10]])
table = biom.Table(data, sample_ids=['a', 'b', 'c'],
observation_ids=['x', 'y'])
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['-4.2', '-4.2', '-4.2'], name='foo',
index=pd.Index(['a', 'b', 'c'], name='sampleid')))
expected = biom.Table(np.array([[6], [60]]),
sample_ids=['-4.2'],
observation_ids=['x', 'y'])
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(expected, result)
def test_empty_table(self):
mc = qiime2.CategoricalMetadataColumn(
pd.Series(['a_new', 'b_new'], name='foo',
index=pd.Index(['a', 'b'], name='id')))
table = biom.Table(np.array([[]]), sample_ids=[], observation_ids=[])
with self.assertRaisesRegex(ValueError, 'empty table'):
group(table, axis='sample', metadata=mc, mode='sum')
with self.assertRaisesRegex(ValueError, 'empty table'):
group(table, axis='feature', metadata=mc, mode='sum')
def _shared_setup(self):
sample_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['treatment', 'treatment', 'control', 'other',
'control', 'other', 'other'], name='foo',
index=pd.Index(['a', 'b', 'c', 'd', 'e', 'f', 'g'],
name='sampleid')))
feature_mc = qiime2.CategoricalMetadataColumn(
pd.Series(['g0', 'g1', 'g1', 'g1', 'g0'], name='foo',
index=pd.Index(['v', 'w', 'x', 'y', 'z'],
name='featureid')))
data = np.array([
# t t c o c o o
# a b c d e f g
[0, 0, 0, 0, 1, 0, 2], # v g0
[10, 10, 10, 10, 10, 100, 1], # w g1
[12, 3, 14, 0, 0, 3, 34], # x g1
[1, 1, 1, 1, 1, 1, 1], # y g1
[0, 1, 11, 111, 1111, 20, 20]]) # z g0
table = biom.Table(data, sample_ids=sample_mc.to_series().index,
observation_ids=feature_mc.to_series().index)
return sample_mc, feature_mc, table
def test_feature_sum(self):
sample_mc, feature_mc, table = self._shared_setup()
expected = biom.Table(
np.array([[0, 1, 11, 111, 1112, 20, 22],
[23, 14, 25, 11, 11, 104, 36]]),
sample_ids=sample_mc.to_series().index,
observation_ids=['g0', 'g1'])
result = group(table, axis='feature', metadata=feature_mc, mode='sum')
self.assertEqual(expected, result)
def test_feature_mean_ceiling(self):
sample_mc, feature_mc, table = self._shared_setup()
expected = biom.Table(
np.array([[0, 1, 6, 56, 556, 10, 11],
[8, 5, 9, 4, 4, 35, 12]]),
sample_ids=sample_mc.to_series().index,
observation_ids=['g0', 'g1'])
result = group(table, axis='feature', metadata=feature_mc,
mode='mean-ceiling')
self.assertEqual(expected, result)
def test_feature_median_ceiling(self):
sample_mc, feature_mc, table = self._shared_setup()
expected = biom.Table(
np.array([[0, 1, 6, 56, 556, 10, 11],
[10, 3, 10, 1, 1, 3, 1]]),
sample_ids=sample_mc.to_series().index,
observation_ids=['g0', 'g1'])
result = group(table, axis='feature', metadata=feature_mc,
mode='median-ceiling')
self.assertEqual(expected, result)
def test_sample_sum(self):
sample_mc, feature_mc, table = self._shared_setup()
expected = biom.Table(
np.array([[0, 1, 2],
[20, 20, 111],
[15, 14, 37],
[2, 2, 3],
[1, 1122, 151]]),
sample_ids=['treatment', 'control', 'other'],
observation_ids=feature_mc.to_series().index)
result = group(table, axis='sample', metadata=sample_mc, mode='sum')
self.assertEqual(expected, result)
def test_sample_mean_ceiling(self):
sample_mc, feature_mc, table = self._shared_setup()
expected = biom.Table(
np.array([[0, 1, 1],
[10, 10, 37],
[8, 7, 13],
[1, 1, 1],
[1, 561, 51]]),
sample_ids=['treatment', 'control', 'other'],
observation_ids=feature_mc.to_series().index)
result = group(table, axis='sample', metadata=sample_mc,
mode='mean-ceiling')
self.assertEqual(expected, result)
def test_sample_median_ceiling(self):
sample_mc, feature_mc, table = self._shared_setup()
expected = biom.Table(
np.array([[0, 1, 0],
[10, 10, 10],
[8, 7, 3],
[1, 1, 1],
[1, 561, 20]]),
sample_ids=['treatment', 'control', 'other'],
observation_ids=feature_mc.to_series().index)
result = group(table, axis='sample', metadata=sample_mc,
mode='median-ceiling')
self.assertEqual(expected, result)
if __name__ == '__main__':
unittest.main()
|
