A novel dynamic part-wise direct and complementary template matching approach for DNA sequence similarity identification.

Understanding DNA sequence similarity is essential for uncovering evolutionary relationships and functional insights across diverse biological systems, particularly in the era of rapidly expanding genomic data, DNA sequence similarity analysis plays a central role in comparative genomics, evolutionary biology, and phylogenetic reconstruction. However, the rapid expansion of genomic databases has made large-scale sequence comparison increasingly challenging for traditional alignment-based approaches due to their high computational cost and limited efficiency for highly divergent sequences. Alignment-free methods provide an attractive alternative, but many existing techniques still face limitations in phylogenetic accuracy and computational efficiency. In this study, we propose a novel alignment-free method for DNA sequence similarity analysis based on dynamic template matching and subsequence similarity representation. The approach generates a sub-sequence similarity score number (SSSN) vector using direct and complementary nucleotide matching and reconstructs phylogenetic relationships from the resulting feature vectors. The method was evaluated using two benchmark datasets and four standard biological datasets. Experimental results show that the proposed approach achieves high phylogenetic accuracy (95–100%) while significantly reducing computational requirements. In particular, it is 55–1747 times faster than the MEGA tool and requires 30–99% less memory than several existing approaches. Overall, the proposed method provides an efficient and scalable framework for DNA sequence similarity analysis and phylogenetic inference in large genomic datasets. These findings demonstrate that the proposed method provides an accurate and computationally efficient framework for large-scale DNA sequence analysis, with strong potential for applications in comparative genomics, evolutionary studies, and future high-throughput genomic research. The datasets and source code supporting this study are publicly available at the provided https://github.com/machbah/DPTM_Seq_Sim Github repository.