OPEN MP-BASED PARALLEL AND SCALABLE GENETIC SEQUENCE ALIGNMENT

Abstract

In bioinformatics, sequence alignment is a common and insistent task. Biologists align genome sequences to find important similarities and dissimilarities in them. Multiple heuristics and dynamic programming based approaches are available for sequence alignment. Smith-Waterman (SW), an exact algorithm for local alignment, is the most accurate of them all. However, the space and time complexity of the SW algorithm is quadratic. It is imperative to use parallelism and distributed computing techniques in order to speed up this process. In this paper, we discuss and evaluate an OpenMP based implementation of SW algorithm. All the experiments have been performed on a Linux based multi-core machine thereby reducing the overall complexity of the SW algorithm from quadratic to linear. The results obtained with various input sequences demonstrate that the parallel version of the SW algorithm runs 2.63 times faster than its sequential counterpart.