Parallelization of the Smith-Waterman Algorithm to Accelerate DNA Sequence Alignment

The genome is a rapidly evolving topic that constantly offers researchers with new biological and computational application challenges. One of the biggest challenges is biological sequence comparison, which involves identifying similarities or identical segments between a query sequence and a reference sequence. Global and local alignments are two common types of sequence alignments. Global alignment identifies the optimal alignment across entire sequences, while local alignment focuses on matching specific segments rather than the entire length. Local alignment algorithms offer faster performance compared to global alignment since they only align specific portions of sequences. In this study, the Smith-Waterman method would be executed serially and in parallel for comparison and analysis. The results reveal that parallel execution has a significant advantage over serial execution, with a maximum improvement of $78.5 \%$ above serial performance. Furthermore, increasing the matrix size enhances parallel execution performance over serial processing.