A survey on improving pattern matching algorithms for biological sequences

Abstract

SummaryPattern matching is a highly useful procedure in several stages of the computational pipelines. Furthermore, some research trends in this research domain contributed to growing biological databases and updated them throughout time. This article proposes an comparison and analysis of different algorithms for match equivalent pattern matching like complexity, efficiency, and techniques. Which algorithm is best for which DNA sequence and why? This describes the different algorithms for various activities that include pattern matching as an important aspect of functionality. This article shows that BM, Horspool, ZT, QS, FS, Smith, and SSABS methods employ the bad character preprocessing function. In addition, BM, SSABS, TVSBS, and BRFS methods are using two approaches in the preprocessing stage, which decreases the preprocessing time. Furthermore, KR, QS, SSABS, BRFS, and Shift‐Or are not recommended for the long pattern, whereas ZT, FS, d‐BM, Raita, and Smith are not recommended for the short pattern. This is because they are time‐consuming and certain algorithms, such as ZT and DCPM, use a lot of time and space during the matching and search process, while others, such as d‐BM and TSW, save space and time. Although DCPM, BRFS, and QS are quicker than other algorithms, FLPM, PAPM, and LFPM rank highest in terms of complexity time.