Parallel implementation of quorum planted (l, d) motif search on multi-core/many-core platforms

摘要

Multi-core and many-core architectures are widely adopted by researchers in applied sciences and engineering, owing to their reasonable cost, and ease of access. Moreover, their painless hardware set-up process and rather simple programming paradigm attract more researchers to acquire them and implement their time-expensive computations on these platforms. Planted Motif Search problem is one of the most challenging problems in bioinformatics whose goal is to enumerate all strings of length l that are commonly planted in a given set of DNA sequences. In this paper, we propose an efficient method of thread parallelization to accelerate the latest Quorum Planted Motif Search algorithm (qPMS9) on multi core and many-core systems. Our contribution towards dynamic scheduling of threads and parallelization of loops in the proposed method outperforms previous sequential and parallel algorithms. (C) 2016 Elsevier B.V. All rights reserved.