Comparison of multiple sequence alignment approaches based on heuristic methods for DNA sequences

Abstract

Purpose: This study addresses the multiple sequence alignment (MSA) problem by proposing four heuristic algorithms-GA, DE, GASA, and DESA-that reorder DNA sequences and apply the Needleman-Wunsch algorithm to improve alignment quality. The proposed methods are compared with ClustalW in terms of alignment score, runtime, and computational complexity. Theory and Methods: Bioinformatics applies mathematical and computational methods to analyze biological data. To address the complexity of sequence alignment, this study proposes four heuristic algorithms-GA, DE, GASA, and DESA-that integrate the Needleman-Wunsch algorithm to optimize DNA sequence alignment. The proposed methods are evaluated against ClustalW, particularly for datasets with equal-length sequences. Results: The proposed algorithms outperformed ClustalW in alignment scores on four of six datasets (trnN-GUU_rps12, rrn4.5_rps12, rrn5_rps12, psbT_pbf1). GA achieved the fastest runtime, while DE and DESA produced the highest alignment scores with longer execution times. Although ClustalW showed the best productivity (alignment score/runtime), the proposed methods provide advantages when alignment accuracy is prioritized, particularly DE and DESA on equal-length sequence datasets. Conclusion: The proposed heuristic algorithms (GA, DE, GASA, DESA) achieved higher alignment accuracy than ClustalW in most datasets. Although ClustalW is faster and computationally efficient, it delivered lower alignment quality in several cases. These methods are particularly suitable for bioinformatics tasks requiring high alignment accuracy.