An Efficient Parallel Algorithm for Constraint Multiple Sequence Alignment

Abstract

The concept of constrained sequence alignment is proposed to incorporate the biologist's knowledge into sequence alignment such that the user-specified residues/nucleotides are aligned together in the computed alignment. Tang et al. were first to investigate the constrained multiple sequence alignment problem. Their algorithm for two sequences alignment with constraints runs in O(αn4) time and needs O(n4) space. Later, this result was improved by two groups of researchers independently to O(rn2) time and space using the same approach of dynamic programming. Recently, Lu and Huang designed a memory-efficient algorithm to improve the two sequence alignment with constraints (CPSA) by adopting the divide-and-conquer approach, this algorithm for solving CPSA problem can run in O(rn2) time and only need O(un) space, where u is the sum of the length of constraints and usually u<<n in the practical applications. In this paper, we design an efficient parallel algorithm for the constrained multiple sequence alignment based on the memory-efficient algorithm designed by Lu and Huang and the progressive strategy.