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
|
# ----------------------------------------------------------------------------
# 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
import numpy as np
import numpy.testing as npt
from skbio.util import get_data_path
from skbio import Protein
from skbio.embedding._embedding import SequenceVector, embed_vec_to_numpy
from skbio.embedding._protein import (
ProteinEmbedding,
ProteinVector
)
class ProteinEmbeddingTests(TestCase):
def setUp(self):
self.emb = np.load(get_data_path("embed1.txt.npy"))
self.seq = ("IGKEEIQQRLAQFVDHWKELKQLAAARGQRLEESLEYQ"
"QFVANVEEEEAWINEKMTLVASED")
self.invalid_seq = (
"$GKEEIQQRLAQFVDHWKELKQLAAARGQRLEESLEYQ"
"QFVANVEEEEAWINEKMTLVASED")
def test_clipping(self):
emb, s = self.emb, self.seq
nemb = np.zeros((emb.shape[0] + 2, emb.shape[1]))
nemb[1:-1] = emb
p2_emb = ProteinEmbedding(nemb, s, clip_head=True, clip_tail=True)
npt.assert_array_equal(p2_emb.embedding, emb)
self.assertEqual(p2_emb.sequence, s)
def test_str(self):
emb, s = self.emb, self.seq
p_emb = ProteinEmbedding(emb, s)
self.assertEqual(str(p_emb), s)
self.assertEqual(p_emb.sequence, s)
byte_s = np.array([b"I", b"G", b"K", b"E", b"E", b"I", b"Q",
b"Q", b"R", b"L", b"A", b"Q", b"F", b"V",
b"D", b"H", b"W", b"K", b"E", b"L", b"K",
b"Q", b"L", b"A", b"A", b"A", b"R", b"G",
b"Q", b"R", b"L", b"E", b"E", b"S", b"L",
b"E", b"Y", b"Q", b"Q", b"F", b"V", b"A",
b"N", b"V", b"E", b"E", b"E", b"E", b"A",
b"W", b"I", b"N", b"E", b"K", b"M", b"T",
b"L", b"V", b"A", b"S", b"E", b"D"], dtype="|S1")
npt.assert_array_equal(p_emb.residues, byte_s)
self.assertEqual(str(p_emb.ids.tobytes().decode("ascii")), s)
def test_skbio_protein(self):
emb, s = self.emb, self.seq
p_emb = ProteinEmbedding(emb, Protein(s))
self.assertEqual(str(p_emb), s)
self.assertEqual(p_emb.sequence, s)
def test_str_spaces(self):
seq = ("I G K E E I Q Q R L A Q F V D H W K E L K Q L A "
"A A R G Q R L E E S L E Y Q Q F V A N V E E E E "
"A W I N E K M T L V A S E D")
p_emb = ProteinEmbedding(self.emb, seq)
self.assertEqual(str(p_emb), self.seq)
self.assertEqual(p_emb.sequence, self.seq)
def test_embedding(self):
emb, s = self.emb, self.seq
p_emb = ProteinEmbedding(emb, s)
self.assertTupleEqual(p_emb.embedding.shape, (62, 1024))
def test_assert_length(self):
with self.assertRaises(ValueError):
ProteinEmbedding(self.emb, self.seq + "A")
def test_invalid_sequence(self):
emb, s = self.emb, self.invalid_seq
with self.assertRaises(ValueError):
ProteinEmbedding(emb, s)
def test_repr(self):
emb, s = self.emb, self.seq
p_emb = ProteinEmbedding(emb, s)
self.assertIn("ProteinEmbedding", repr(p_emb))
class ProteinVectorTests(TestCase):
def setUp(self):
rk = 10
self.emb = np.random.randn(rk)
self.seq = Protein(
"IGKEEIQQRLAQFVDHWKELKQLAAARGQRLEESLEYQQFVANVEEEEAWINEKMTLVASED",
metadata={"id": "seq1"}
)
self.vector1 = np.array([1, 2, 3])
self.vector2 = np.array([4, 5, 6])
self.vector3 = np.array([7, 8, 9])
self.bad_vector = np.array([7, 8])
self.bad_vector2 = np.array([[7, 8], [7, 9]])
self.protein_vectors = [ProteinVector(self.vector1, "IGKE"),
ProteinVector(self.vector2, "EIQQ"),
ProteinVector(self.vector3, "RLAQ")]
def test_valid_protein_vector(self):
ProteinVector(self.emb, self.seq)
ProteinVector(self.emb, str(self.seq))
ProteinVector(self.emb, str(self.seq).encode("ascii"))
def test_invalid_protein_vector(self):
seq = "$GKEEIQQRLAQFVDHWKELKQLAAARGQRLEESLEYQQFVANVEEEEAWINEKMTLVASED^^"
with self.assertRaises(ValueError):
ProteinVector(self.emb, seq)
with self.assertRaises(ValueError):
ProteinVector(self.bad_vector2, seq)
def test_invalid_vector_shape(self):
msg = "Only one vector per sequence is allowed."
with self.assertRaisesRegex(ValueError, msg):
ProteinVector(np.vstack([self.emb, self.emb]), self.seq)
def test_repr(self):
pv = ProteinVector(self.emb, self.seq)
self.assertIn("ProteinVector", repr(pv))
self.assertIn("vector dimension", repr(pv))
def test_to_numpy(self):
# confirm that Protein objects can be casted to numpy
exp = np.array([self.vector1, self.vector2, self.vector3])
obs = embed_vec_to_numpy(self.protein_vectors)
npt.assert_array_equal(obs, exp)
def test_to_numpy_raises(self):
# assert that all types are the same
lst = [ProteinVector(self.vector1, "IGKE"),
SequenceVector(self.vector2, "EIQQ"),
SequenceVector(self.bad_vector, "RLAQ")]
with self.assertRaises(ValueError):
embed_vec_to_numpy(lst)
# assert that all objects subclass EmbeddingVector
lst = [Protein("IGKE"),
Protein("EIQQ"),
Protein("RLAQ")]
with self.assertRaises(ValueError):
embed_vec_to_numpy(lst)
if __name__ == "__main__":
main()
|
