-
Notifications
You must be signed in to change notification settings - Fork 0
/
align.py
196 lines (174 loc) · 7.01 KB
/
align.py
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
from Bio.SubsMat import MatrixInfo
from itertools import product
from itertools import combinations
from operator import sub
import numpy as np
def read_scoring_matrix(ime):
"""Loads scoring matrix from given name from Biopython MatrixInfo or from given path to file.
Args:
ime (str): Name of BioPythons MatrixInfo variable or name to a file with scoring matrix (eg. *.txt).
Returns:
dict: Dictionary scoring matrix.
"""
scoring_matrix = {}
try:
scoring_matrix_temp = getattr(MatrixInfo, ime) # use built in matrix if exists
for key1, key2 in scoring_matrix_temp: # biopython built in matrices are not symmetric -.-
scoring_matrix[(key1, key2)] = scoring_matrix_temp[(key1, key2)]
scoring_matrix[(key2, key1)] = scoring_matrix_temp[(key1, key2)]
except AttributeError: # else load from file
with open(ime) as f:
stolpci = f.readline().strip().split()
for line in f:
prvi, *vrstica = line.strip().split()
for i, vrednost in enumerate(vrstica):
scoring_matrix[(prvi, stolpci[i])] = int(vrednost)
return scoring_matrix
class Align:
def __init__(self, scoring_matrix, linear_gap_penalty):
self.sm = read_scoring_matrix(scoring_matrix)
self.lgp = linear_gap_penalty
def dinamicna_tabela_global(self, s, t):
m = len(t)
n = len(s)
M = {}
M[(-1, -1)] = 0 # zgornji levi kot tabele
P = {}
# inicializacija
for j in range(n):
M[(-1, j)], P[(-1, j)] = M[(-1, j - 1)] - self.lgp, (-1, j - 1)
for i in range(m):
M[(i, -1)], P[(i, -1)] = M[(i - 1, -1)] - self.lgp, (i - 1, -1)
# izracun vrednosti v tabeli
for i in range(m):
for j in range(n):
M[(i, j)], P[(i, j)] = max(
(M[(i - 1, j)] - self.lgp, (i - 1, j)),
(M[(i, j - 1)] - self.lgp, (i, j - 1)),
(M[(i - 1, j - 1)] + self.sm[(t[i], s[j])], (i - 1, j - 1))
)
return M, P
def dinamicna_tabela_local(self, s, t):
m = len(t)
n = len(s)
M = {}
M[(-1, -1)] = 0 # zgornji levi kot tabele
P = {}
# inicializacija
for j in range(n):
M[(-1, j)], P[(-1, j)] = 0, (-2, -2)
for i in range(m):
M[(i, -1)], P[(i, -1)] = 0, (-2, -2)
# izracun vrednosti v tabeli
for i in range(m):
for j in range(n):
M[(i, j)], P[(i, j)] = max(
(M[(i - 1, j)] - self.lgp, (i - 1, j)),
(M[(i, j - 1)] - self.lgp, (i, j - 1)),
(M[(i - 1, j - 1)] + self.sm[(t[i], s[j])], (i - 1, j - 1)),
(0, (-2, -2))
)
return M, P
def trace_back_local(self, s, t):
M, P = self.dinamicna_tabela_local(s, t)
razlika, (i2, j2) = max((vrednost, indeksa) for indeksa, vrednost in M.items())
i1, j1 = i2, j2
s1 = ""
t1 = ""
while M[(i1, j1)] != 0:
x,y = P[(i1, j1)]
if y == j1:
s1 = "-" + s1
t1 = t[i1] + t1
elif x == i1:
t1 = "-" + t1
s1 = s[j1] + s1
else:
s1 = s[j1] + s1
t1 = t[i1] + t1
i1, j1 = P[(i1, j1)]
return razlika, s1, t1
def trace_back_global(self, s, t):
m = len(t)
n = len(s)
M, P = self.dinamicna_tabela_global(s,t)
s = iter(s)
t = iter(t)
el = (m - 1, n - 1)
pot = []
while True:
pot.append(el)
if el == (-1, -1):
break
el = P[el]
pot = list(reversed(pot))
s1 = []
t1 = []
for i in range(len(pot) - 1):
prvi = pot[i]
drugi = pot[i + 1]
if prvi[0] == drugi[0]:
t1.append("-")
else:
t1.append(next(t))
if prvi[1] == drugi[1]:
s1.append("-")
else:
s1.append(next(s))
return M[(m - 1, n - 1)], "".join(s1), "".join(t1)
def kazen(self, sez, indeksi, komb):
"""Punishment - auxiliary function for multiple alignment"""
return sum(1 if not ((i * j == 1 and niz1[ind1] == niz2[ind2]) or (i == 0 and j == 0)) else 0 for
(niz1, ind1, i), (niz2, ind2, j) in combinations(zip(sez, indeksi, komb), 2))
def dinamicna_tabela_multiple(self, sez):
M = {}
M[tuple(np.repeat(-1, len(sez)))] = 0 # zgornji levi kot tabele
P = {}
indexes = [tuple(np.subtract(dxs, np.ones(len(sez), dtype="int"))) for dxs in
list(np.ndindex(tuple(len(i) + 1 for i in sez)))]
for index in indexes:
if index != tuple(np.repeat(-1, len(sez))):
kombinacije = product([0, 1], repeat=len(sez))
next(kombinacije) # same nicle izpustimo
# ce kljuca ni, vzamemo zelo majhno vrednost (-100000), ki nima vpliva, saj iscemo max
M[index], P[index] = max((M.get(tuple(map(sub, index, komb)), -100000) - self.kazen(sez, index, komb),
tuple(map(sub, index, komb))) for komb in kombinacije)
return M, P
def trace_back_multiple(self, sez):
M, P = self.dinamicna_tabela_multiple(sez)
el = tuple(len(s) - 1 for s in sez)
iters = [iter(s) for s in sez]
pot = []
while True:
pot.append(el)
if el == tuple(np.repeat(-1, len(sez))):
break
el = P[el]
pot = list(reversed(pot))
words = [[] for i in range(len(sez))]
for i in range(len(pot) - 1):
prvi = pot[i]
drugi = pot[i + 1]
for i in range(len(sez)):
if prvi[i] == drugi[i]:
words[i].append("-")
else:
words[i].append(next(iters[i]))
return M[tuple(len(s)-1 for s in sez)], ["".join(s) for s in words]
def global_alignment(self, s, t):
"""global alignment of strings s and t,
returns score and aligned strings s1 and t1"""
d, s1, t1 = self.trace_back_global(s, t)
return d, s1, t1
def local_alignment(self, s, t):
"""local alignment of strings s and t,
returns score and aligned substrings s1 and t1"""
d, s1, t1 = self.trace_back_local(s, t)
return d, s1, t1
def multiple_alignment(self, sez):
"""multiple alignment of all strings in array sez
returns score and array of aligned substrings
NOTE: IT IS NOT EFFICIENT and can take a lot of time
"""
d, words = self.trace_back_multiple(sez)
return d, words