Title: Parallel multiple DNA sequence alignment using genetic algorithm and asynchronous advantage actor critic model

Authors: Khadidja Belattar; El-Amine Zemali; Sabrina Baouni; Souhila Dehni

Addresses: Department of Computer Science, University of Algiers, Algiers, 16000, Algeria ' Department of Computer Science, University of Algiers, Algiers, 16000, Algeria ' Department of Computer Science, University of Algiers, Algiers, 16000, Algeria ' Department of Computer Science, University of Algiers, Algiers, 16000, Algeria

Abstract: Multiple sequence alignment plays a crucial role in bioinformatics field. In this process, sequences are compared by establishing residue-residue correspondence among related sequences. In this paper, we present a novel framework that uses genetic algorithm and asynchronous advantage actor critic model to look for the optimal multiple DNA sequence alignment. In the first step, we generate initial alignments by applying the T-COFFEE method. Then, we recursively use a genetic algorithm to find the optimal fragmentation of the pre-aligned sequences. At this stage, the obtained short sub alignments are introduced to the A3C model for parallel alignments. So, the optimal multiple alignment of the original sequences represents the fusion of sub-alignments. To evaluate the performances of the proposed approach, we used the Balibase 3.0 benchmark and compared to TCOFFEE, MAFFT and RLALIGN methods. The obtained experimental results show that the proposed approach can achieve solutions of competitive quality.

Keywords: parallel sequence alignment; multiple sequence alignment; bioinformatics; genetic algorithm; asynchronous advantage actor critic.

DOI: 10.1504/IJBRA.2022.128236

International Journal of Bioinformatics Research and Applications, 2022 Vol.18 No.5, pp.460 - 478

Received: 08 Dec 2021
Accepted: 15 Jul 2022
Published online: 12 Jan 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article