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
|
# ----------------------------------------------------------------------------
# Copyright (c) 2013--, scikit-bio development team.
#
# Distributed under the terms of the Modified BSD License.
#
# The full license is in the file LICENSE.txt, distributed with this software.
# ----------------------------------------------------------------------------
from unittest import TestCase, main
from pathlib import Path
import tempfile
import os
import re
import h5py
import numpy as np
import skbio
from skbio.io import write
from skbio import Protein
from skbio.util import get_data_path
from skbio.embedding._protein import ProteinEmbedding
from skbio.embedding._protein import ProteinVector
from skbio.io.format.embed import (
_embed_sniffer, _embed_to_generator,
_generator_to_embed, _embed_to_protein,
_protein_to_embed, _protein_to_vector,
_vector_to_protein
)
class TestWriteError(TestCase):
def setUp(self):
self.tempdir = tempfile.TemporaryDirectory()
def test_write_function(self):
# note: tiny_embedding_file stores embeddings for our 20 seqs.
with open(get_data_path('tiny_embedding_file.npz'), 'rb') as fh:
embeddings = np.load(fh)
emb_list = [embeddings[arrs] for arrs in embeddings.files]
# note: pdb_hits.txt stores sequence strings for our 20 seqs.
seqs = np.loadtxt(get_data_path('pdb_hits.txt'), dtype=str)
seq_list = [" ".join(list(re.sub(r"[UZOB]", "X", str(seq))))
for seq in seqs]
embed_list = []
for emb, seq in zip(emb_list, seq_list):
embed_list.append(ProteinEmbedding(embedding=emb, sequence=seq))
# create generator object for write testing
embed_list = (emb for emb in embed_list)
file_path = os.path.join(self.tempdir.name, 'test_pdb_hits.h5')
write(embed_list, 'embed', into=get_data_path(file_path))
self.assertTrue(os.path.exists(file_path))
class EmbedTests(TestCase):
def setUp(self):
# single sequence
rk = 5 # latent dimension of residues
self.sequences = (
[
(
np.load(get_data_path('embed1.txt.npy')),
Protein(('IGKEEIQQRLAQFVDHWKELKQLAAARGQRL'
'EESLEYQQFVANVEEEEAWINEKMTLVASED'),
metadata={"id": "seq1"})
),
(
np.load(get_data_path('embed2.txt.npy')),
Protein(('QQNKELNFKLREKQNEIFELKKIAETLRSKL'
'EKYVDITKKLEDQNLNLQIKISDLEKKLSDA'),
metadata={"id": "seq2"})
)
]
)
self.tempdir = tempfile.TemporaryDirectory()
tempdir = Path(self.tempdir.name)
self.writable_emb_path = str(tempdir / Path('test.emb'))
self.writable_emb_path2 = str(tempdir / Path('test2.emb'))
self.valid_embed_path = get_data_path('prot.emb')
self.invalid_embed_path = str(tempdir / Path('invalid'))
self.nonembed_hdf5_path = str(tempdir / Path('other.hdf5'))
with open(self.invalid_embed_path, 'wb') as fp:
fp.write(b'this is not a embed file')
with h5py.File(self.nonembed_hdf5_path, 'w') as fp:
fp['stuff'] = [1, 2, 3]
def test_sniffer(self):
self.assertEqual(_embed_sniffer(self.valid_embed_path), (True, {}))
self.assertEqual(_embed_sniffer(self.invalid_embed_path), (False, {}))
self.assertEqual(_embed_sniffer(self.nonembed_hdf5_path), (False, {}))
def test_read_write_single(self):
for emb, seq in self.sequences:
fh = self.writable_emb_path
obj = ProteinEmbedding(emb, seq)
_protein_to_embed(obj, fh)
emb2 = _embed_to_protein(fh)
np.testing.assert_array_equal(emb, emb2.embedding)
self.assertEqual(str(seq), str(emb2))
def test_read_write_generator(self):
writable_emb_path2 = 'test2.emb'
objs1 = [ProteinEmbedding(emb, seq) for emb, seq in self.sequences]
_generator_to_embed(objs1, self.writable_emb_path2)
objs2 = _embed_to_generator(self.writable_emb_path2)
for obj1, obj2 in zip(objs1, objs2):
np.testing.assert_array_equal(obj1.embedding, obj2.embedding)
self.assertEqual(str(obj1), str(obj2))
def test_write_generator(self):
sequences = [
(
np.load(get_data_path('embed1.txt.npy')),
Protein(('IGKEEIQQRLAQFVDHWKELKQLAAARGQRL'
'EESLEYQQFVANVEEEEAWINEKMTLVASED'),
metadata={"id": "seq1"})
),
(
np.load(get_data_path('embed2.txt.npy')),
Protein(('QQNKELNFKLREKQNEIFELKKIAETLRSKL'
'EKYVDITKKLEDQNLNLQIKISDLEKKLSDA'),
metadata={"id": "seq2"})
)
]
f = lambda x: ProteinEmbedding(*x)
objs1 = (x for x in map(f, sequences))
tempdir = Path(tempfile.mkdtemp())
writable_emb_path = str(tempdir / Path('test.emb'))
skbio.io.write(objs1, format='embed', into=writable_emb_path)
objs2 = iter(skbio.io.read(writable_emb_path, format='embed',
constructor=ProteinEmbedding))
for obj1, obj2 in zip(objs1, objs2):
np.testing.assert_array_equal(obj1.embedding, obj2.embedding)
self.assertEqual(str(obj1), str(obj2))
class VectorTests(TestCase):
def setUp(self):
# single sequence
rk = 10 # latent dimension of residues
self.sequences = (
[
(
np.random.randn(rk),
Protein(('IGKEEIQQRLAQFVDHWKELKQLAAARGQRL'
'EESLEYQQFVANVEEEEAWINEKMTLVASED'),
metadata={"id": "seq1"})
),
(
np.random.randn(rk),
Protein(('QQNKELNFKLREKQNEIFELKKIAETLRSKL'
'EKYVDITKKLEDQNLNLQIKISDLEKKLSDA'),
metadata={"id": "seq2"})
)
]
)
self.tempdir = tempfile.TemporaryDirectory()
tempdir = Path(self.tempdir.name)
self.writable_emb_path = str(tempdir / Path('test.emb'))
self.writable_emb_path2 = str(tempdir / Path('test2.emb'))
self.valid_embed_path = get_data_path('prot_vec.emb')
self.invalid_embed_path = str(tempdir / Path('invalid'))
self.nonembed_hdf5_path = str(tempdir / Path('other.hdf5'))
with open(self.invalid_embed_path, 'wb') as fp:
fp.write(b'this is not a embed file')
with h5py.File(self.nonembed_hdf5_path, 'w') as fp:
fp['stuff'] = [1, 2, 3]
def test_sniffer(self):
# make sure that the sniffer throws errors as expected
self.assertEqual(_embed_sniffer(self.valid_embed_path), (True, {}))
self.assertEqual(_embed_sniffer(self.invalid_embed_path), (False, {}))
self.assertEqual(_embed_sniffer(self.nonembed_hdf5_path), (False, {}))
emb, seq = self.sequences[0]
obj = ProteinVector(emb, seq)
_protein_to_vector(obj, str(Path(self.tempdir.name) / Path("prot_vec.emb")))
def test_read_write_single(self):
for emb, seq in self.sequences:
fh = self.writable_emb_path
obj = ProteinVector(emb, seq)
_protein_to_vector(obj, fh)
emb2 = _vector_to_protein(fh)
np.testing.assert_array_equal(
emb, emb2.embedding.ravel())
self.assertEqual(str(seq), str(emb2))
def test_read_write_generator(self):
writable_emb_path2 = 'test2.emb'
objs1 = [ProteinVector(emb, seq) for emb, seq in self.sequences]
_generator_to_embed(objs1, self.writable_emb_path2)
objs2 = _embed_to_generator(self.writable_emb_path2,
constructor=ProteinVector)
for obj1, obj2 in zip(objs1, objs2):
np.testing.assert_array_equal(obj1.embedding, obj2.embedding)
self.assertEqual(str(obj1), str(obj2))
def test_write_generator(self):
sequences = self.sequences
f = lambda x: ProteinVector(*x)
objs1 = (x for x in map(f, sequences))
tempdir = Path(tempfile.mkdtemp())
writable_emb_path = str(tempdir / Path('test.emb'))
skbio.io.write(objs1, format='embed', into=writable_emb_path)
objs2 = iter(skbio.io.read(writable_emb_path, format='embed',
constructor=ProteinVector))
for obj1, obj2 in zip(objs1, objs2):
np.testing.assert_array_equal(obj1.embedding, obj2.embedding)
self.assertEqual(str(obj1), str(obj2))
if __name__ == '__main__':
main()
|
