Skip to content

Instantly share code, notes, and snippets.

@afrendeiro
Created June 4, 2014 22:28
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save afrendeiro/ddb7d81104cf1ffdbc32 to your computer and use it in GitHub Desktop.
Save afrendeiro/ddb7d81104cf1ffdbc32 to your computer and use it in GitHub Desktop.
Exact pairwise alignment by dynamic programming. Forward recursion with pruning
#! /usr/bin/python
# Forward-recursion with pruning
# An algorithm for doing exact alignment of two sequences
# Forward recursion is used, with pruning of cells
# two sequences to align
S1 = 'APPLR'
S2 = 'APPLS'
# the DP matrix as a dict, prepopulated with None
H = {}
for l in range(0, len(S1) + 1):
for i in range(0, len(S2) + 1):
H[(l,i)] = None
# the end cell
Hn = (len(S1),len(S2))
# K, lower bound score
K = 737
# Queue
Q = []
def upperBound(v):
"""
finds an upper bound of the score of
the alignment from a cell v to the end-cell hN
"""
return 737 - H[v]
#raise NotImplementedError("Upper score bound not yet implemented")
def getForwardNeighbours(v):
"""
finds all forward neighbour cells to current cell.
Requires a tuple of coordinates (v).
Returns list of tuples of neighbours in right order.
"""
if v[0] >= 1 and v[1] >= 1 and v[0] < len(S1) and v[1] < len(S2):
return [(v[0], v[1] + 1), (v[0] + 1, v[1]), (v[0] + 1, v[1] + 1)]
else:
return []
def getScore(v, w):
"""
Calculates the score of going from one cell (v) to another (w).
Requires a tuple of coordinates (v).
Returns list of tuples of neighbours in right order.
"""
if v[0] == w[0] or v[1] == w[1]:
return H[v] - 1
else:
return H[v] + 1
# Add sequences to the DP matrix
for l in range(1, len(S1) + 1):
H[(l,0)] = S1[l - 1]
for l in range(1, len(S2) + 1):
H[(0,l)] = S2[l - 1]
# Set current cell, "v" to start cell
v = (1,1)
# Best score of an alignment from H0 to u
Pu = 0
H[v] = Pu
# Push current (start) cell to the queue
Q.append(v)
# while queue has elements do
while Q != []:
# remove v from queue
v = Q.pop(0)
# set best score so far to the current cell's score
Su = Pu
if H[v] + upperBound(v) >= K:
for neighbour in getForwardNeighbours(v):
if neighbour not in Q:
Q.append(neighbour)
# calculate score and assign as best
Pu = Su + getScore(v, neighbour)
# store the score of the w cell in the matrix
H[neighbour] = Pu
else:
# pick maximum score and assign as best
Pu = max(Pu, Su + getScore(v, neighbour))
# store the score of the w cell in the matrix
H[neighbour] = Pu
print(Su)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment