PyOPA - Optimal Pairwise Alignments

This python package provides a fast implementation to compute

• optimal pairwise alignments of molecular sequences
• ML distance estimates of pairwise alignments.

The implementation uses Farrar's algorithm <http://bioinformatics.oxfordjournals.org/content/23/2/156.abstract>_ to compute the optimal pairwise alignment using SSE vectorization operations. This package implements the Smith-Waterman and Needleman-Wunsch algorithm to compute the local and global sequence alignments.

Installation

You can install the package using pip:

pip install pyopa

We provide wheels for Linux, MacOS (x86_64) and (arm64) for cpython.

Example

import pyopa
log_pam1_env = pyopa.read_env_json(os.path.join(pyopa.matrix_dir(), 'logPAM1.json'))
s1 = pyopa.Sequence('GCANLVSRLENNSRLLNRDLIAVKINADVYKDPNAGALRL')
s2 = pyopa.Sequence('GCANPSTLETNSQLVNRELIAVKINPRVYKGPNLGAFRL')

# super fast check whether the alignment reaches a given min-score
min_score = 100
pam250_env = pyopa.generate_env(log_pam1_env, 250, min_score)
pyopa.align_short(s1, s2, pam250_env)