DNA序列局部比对(Smith–Waterman algorithm)

生物信息原理作业第三弹:DNA序列局部比对,利用Smith–Waterman算法,python3.6代码实现。

实例以及原理均来自https://en.wikipedia.org/wiki/Smith%E2%80%93Waterman_algorithm

DNA序列局部比对

转载请保留出处!

 1 import numpy as np
 2 import pandas as pd
 3 sequence1 = 'TGTTACGG'
 4 sequence2 = 'GGTTGACTA'
 5 s1 = ''
 6 s2 = ''
 7 gap = -2
 8 score_matrix = pd.read_excel('score.xlsx')      #匹配得分
 9 print(score_matrix)
10 best_matrix = np.empty(shape= (len(sequence2)+1,len(sequence1)+1),dtype = int)
11 def get_match_score(s1,s2):
12     score = score_matrix[s1][s2]
13     return score
14 
15 def get_matrix_max(matrix):                    #得到最大分数下标
16     Max = matrix.max()
17     for i in range(len(sequence2)+1):
18         for j in range(len(sequence1)+1):
19             if matrix[i][j] == Max:
20                 return (i,j)
21             
22 for i in range(len(sequence2)+1):
23     for j in range(len(sequence1)+1):
24         if i == 0 or j == 0:
25             best_matrix[i][j] = 0
26         else:
27             match = get_match_score(sequence2[i-1],sequence1[j-1])
28             gap1_score = best_matrix[i-1][j] + gap
29             gap2_score = best_matrix[i][j-1] + gap
30             match_score = best_matrix[i-1][j-1]+match
31             score = max(gap1_score,gap2_score,match_score)
32             if score>0:
33                 best_matrix[i][j] = score
34             else:
35                 best_matrix[i][j] = 0
36 print(best_matrix)
37 
38 #traceback
39 i,j = get_matrix_max(best_matrix)
40 while(best_matrix[i][j]!= 0):
41     match = get_match_score(sequence2[i-1],sequence1[j-1])
42     if i>0 and j>0 and best_matrix[i][j] == best_matrix[i-1][j-1]+match:
43         s1 += sequence1[j-1]
44         s2 += sequence2[i-1]
45         i-=1;j-=1
46     elif i>0 and best_matrix[i,j] == best_matrix[i-1,j]+gap:
47         s1+='-'
48         s2+=sequence2[i-1]
49         i-=1
50     else:
51         s1+=sequence1[j-1]
52         s2+='-'
53         j-=1
54 print(s1[::-1]+'\n'+s2[::-1])

感觉我的得分矩阵写成Excel不必要,等我熟悉一下Numpy和Python命令行之后会修改的。

posted @ 2017-11-30 18:15  orange1002  阅读(5276)  评论(2编辑  收藏  举报