1
|
|
|
# Copyright 2014-2020 by Christopher C. Little. |
2
|
|
|
# This file is part of Abydos. |
3
|
|
|
# |
4
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
5
|
|
|
# it under the terms of the GNU General Public License as published by |
6
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
7
|
|
|
# (at your option) any later version. |
8
|
|
|
# |
9
|
|
|
# Abydos is distributed in the hope that it will be useful, |
10
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
11
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
12
|
|
|
# GNU General Public License for more details. |
13
|
|
|
# |
14
|
|
|
# You should have received a copy of the GNU General Public License |
15
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
16
|
|
|
|
17
|
|
|
"""abydos.distance._gotoh. |
18
|
|
|
|
19
|
1 |
|
Gotoh score |
20
|
|
|
""" |
21
|
|
|
from typing import Any, Callable, Optional, cast |
22
|
|
|
|
23
|
|
|
from numpy import float_ as np_float |
24
|
1 |
|
from numpy import zeros as np_zeros |
25
|
|
|
|
26
|
|
|
from ._needleman_wunsch import NeedlemanWunsch |
27
|
|
|
|
28
|
|
|
__all__ = ['Gotoh'] |
29
|
|
|
|
30
|
|
|
|
31
|
1 |
|
class Gotoh(NeedlemanWunsch): |
32
|
|
|
"""Gotoh score. |
33
|
1 |
|
|
34
|
1 |
|
The Gotoh score :cite:`Gotoh:1982` is essentially Needleman-Wunsch with |
35
|
|
|
affine gap penalties. |
36
|
1 |
|
|
37
|
|
|
.. versionadded:: 0.3.6 |
38
|
1 |
|
""" |
39
|
1 |
|
|
40
|
1 |
|
def __init__( |
41
|
|
|
self, |
42
|
1 |
|
gap_open: float = 1, |
43
|
|
|
gap_ext: float = 0.4, |
44
|
|
|
sim_func: Optional[Callable[[str, str], float]] = None, |
45
|
1 |
|
**kwargs: Any |
46
|
|
|
) -> None: |
47
|
|
|
"""Initialize Gotoh instance. |
48
|
|
|
|
49
|
|
|
Parameters |
50
|
|
|
---------- |
51
|
|
|
gap_open : float |
52
|
|
|
The cost of an open alignment gap (1 by default) |
53
|
|
|
gap_ext : float |
54
|
1 |
|
The cost of an alignment gap extension (0.4 by default) |
55
|
|
|
sim_func : function |
56
|
|
|
A function that returns the similarity of two characters (identity |
57
|
|
|
similarity by default) |
58
|
|
|
**kwargs |
59
|
|
|
Arbitrary keyword arguments |
60
|
|
|
|
61
|
|
|
|
62
|
|
|
.. versionadded:: 0.4.0 |
63
|
|
|
|
64
|
|
|
""" |
65
|
|
|
super(Gotoh, self).__init__(**kwargs) |
66
|
|
|
self._gap_open = gap_open |
67
|
|
|
self._gap_ext = gap_ext |
68
|
|
|
self._sim_func = cast( |
69
|
|
|
Callable[[str, str], float], |
70
|
|
|
NeedlemanWunsch.sim_matrix if sim_func is None else sim_func, |
71
|
|
|
) # type: Callable[[str, str], float] |
72
|
|
|
|
73
|
1 |
|
def sim_score(self, src: str, tar: str) -> float: |
74
|
1 |
|
"""Return the Gotoh score of two strings. |
75
|
1 |
|
|
76
|
1 |
|
Parameters |
77
|
1 |
|
---------- |
78
|
1 |
|
src : str |
79
|
|
|
Source string for comparison |
80
|
1 |
|
tar : str |
81
|
|
|
Target string for comparison |
82
|
|
|
|
83
|
|
|
Returns |
84
|
|
|
------- |
85
|
|
|
float |
86
|
|
|
Gotoh score |
87
|
|
|
|
88
|
|
|
Examples |
89
|
|
|
-------- |
90
|
|
|
>>> cmp = Gotoh() |
91
|
|
|
>>> cmp.sim_score('cat', 'hat') |
92
|
|
|
2.0 |
93
|
|
|
>>> cmp.sim_score('Niall', 'Neil') |
94
|
|
|
1.0 |
95
|
|
|
>>> round(cmp.sim_score('aluminum', 'Catalan'), 12) |
96
|
|
|
-0.4 |
97
|
|
|
>>> cmp.sim_score('cat', 'hat') |
98
|
|
|
2.0 |
99
|
|
|
|
100
|
|
|
|
101
|
|
|
.. versionadded:: 0.1.0 |
102
|
|
|
.. versionchanged:: 0.3.6 |
103
|
|
|
Encapsulated in class |
104
|
|
|
|
105
|
|
|
""" |
106
|
|
|
d_mat = np_zeros((len(src) + 1, len(tar) + 1), dtype=np_float) |
107
|
|
|
p_mat = np_zeros((len(src) + 1, len(tar) + 1), dtype=np_float) |
108
|
|
|
q_mat = np_zeros((len(src) + 1, len(tar) + 1), dtype=np_float) |
109
|
|
|
|
110
|
|
|
d_mat[0, 0] = 0 |
111
|
|
|
p_mat[0, 0] = float('-inf') |
112
|
|
|
q_mat[0, 0] = float('-inf') |
113
|
1 |
|
for i in range(1, len(src) + 1): |
114
|
1 |
|
d_mat[i, 0] = float('-inf') |
115
|
1 |
|
p_mat[i, 0] = -self._gap_open - self._gap_ext * (i - 1) |
116
|
|
|
q_mat[i, 0] = float('-inf') |
117
|
1 |
|
if len(tar) > 1: |
118
|
1 |
|
q_mat[i, 1] = -self._gap_open |
119
|
1 |
|
for j in range(1, len(tar) + 1): |
120
|
1 |
|
d_mat[0, j] = float('-inf') |
121
|
1 |
|
p_mat[0, j] = float('-inf') |
122
|
1 |
|
if len(src) > 1: |
123
|
1 |
|
p_mat[1, j] = -self._gap_open |
124
|
1 |
|
q_mat[0, j] = -self._gap_open - self._gap_ext * (j - 1) |
125
|
1 |
|
|
126
|
1 |
|
for i in range(1, len(src) + 1): |
127
|
1 |
|
for j in range(1, len(tar) + 1): |
128
|
1 |
|
sim_val = self._sim_func(src[i - 1], tar[j - 1]) |
129
|
1 |
|
d_mat[i, j] = max( |
130
|
1 |
|
d_mat[i - 1, j - 1] + sim_val, |
131
|
1 |
|
p_mat[i - 1, j - 1] + sim_val, |
132
|
|
|
q_mat[i - 1, j - 1] + sim_val, |
133
|
1 |
|
) |
134
|
1 |
|
|
135
|
1 |
|
p_mat[i, j] = max( |
136
|
1 |
|
d_mat[i - 1, j] - self._gap_open, |
137
|
|
|
p_mat[i - 1, j] - self._gap_ext, |
138
|
|
|
) |
139
|
|
|
|
140
|
|
|
q_mat[i, j] = max( |
141
|
|
|
d_mat[i, j - 1] - self._gap_open, |
142
|
1 |
|
q_mat[i, j - 1] - self._gap_ext, |
143
|
|
|
) |
144
|
|
|
|
145
|
|
|
i, j = (n - 1 for n in d_mat.shape) |
|
|
|
|
146
|
|
|
return cast(float, max(d_mat[i, j], p_mat[i, j], q_mat[i, j])) |
147
|
1 |
|
|
148
|
|
|
def sim(self, src: str, tar: str) -> float: |
149
|
|
|
"""Return the normalized Gotoh score of two strings. |
150
|
|
|
|
151
|
|
|
Parameters |
152
|
1 |
|
---------- |
153
|
1 |
|
src : str |
154
|
|
|
Source string for comparison |
155
|
1 |
|
tar : str |
156
|
|
|
Target string for comparison |
157
|
|
|
|
158
|
|
|
Returns |
159
|
|
|
------- |
160
|
|
|
float |
161
|
|
|
Normalized Gotoh score |
162
|
|
|
|
163
|
|
|
Examples |
164
|
|
|
-------- |
165
|
|
|
>>> cmp = Gotoh() |
166
|
|
|
>>> cmp.sim('cat', 'hat') |
167
|
|
|
0.6666666666666667 |
168
|
|
|
>>> cmp.sim('Niall', 'Neil') |
169
|
|
|
0.22360679774997896 |
170
|
|
|
>>> round(cmp.sim('aluminum', 'Catalan'), 12) |
171
|
|
|
0.0 |
172
|
|
|
>>> cmp.sim('cat', 'hat') |
173
|
|
|
0.6666666666666667 |
174
|
|
|
|
175
|
|
|
|
176
|
|
|
.. versionadded:: 0.4.1 |
177
|
|
|
|
178
|
|
|
""" |
179
|
|
|
if src == tar: |
180
|
|
|
return 1.0 |
181
|
|
|
return max(0.0, self.sim_score(src, tar)) / ( |
182
|
|
|
self.sim_score(src, src) ** 0.5 * self.sim_score(tar, tar) ** 0.5 |
183
|
|
|
) |
184
|
|
|
|
185
|
|
|
|
186
|
1 |
|
if __name__ == '__main__': |
187
|
1 |
|
import doctest |
188
|
1 |
|
|
189
|
|
|
doctest.testmod() |
190
|
|
|
|