A multi-objective evolutionary algorithm for improving the iterative search of MAFFT
Mengxi Gu, Xinyue Zhang, Zhengxin Huang, Demin Cao, Xiaoxue Zhang, Parvaiz Ahmad Naik · Innovation and Emerging Technologies · 2025
The multiple sequence alignment (MSA) problem is a central issue in bioinformatics, with significant applications in the evolutionary analysis of biological sequences, identification of conserved regions, protein structure prediction, gene annotation, and function prediction. MAFFT is currently one of the most widely used tools for solving the MSA problem. This article aims to improve the limitations of MAFFT in solving the MSA problem by using a multi-objective evolutionary algorithm (MOEA) to enhance the quality of the iterative search. To this end, a new three-objective model is defined, and a novel MOEA is proposed to optimize this model. Four mutation and two crossover operators are designed to generate promising offspring individuals during the evolutionary process. To evaluate the effectiveness of the proposed method, computational experiments are conducted on 60 instances randomly selected from the BAliBASE 3.0 database. The results show that the proposed method can improve the quality of the iterative search and output alignment of MAFFT in terms of the Q and TC metric scores. This study provides new insights into how to better utilize existing tools to solve the MSA problem and offers a promising direction for improving the quality of their output alignments in future research.