A parallel implementation of the Smith-Waterman algorithm for massive sequences searching

Efficient biological sequence searching is an important and challenging task in bioinformatics. Among those fundamental sequence analysis algorithms, the Smith-Waterman algorithm that adopts the dynamic programming mechanism provides very high sensitivity. Unfortunately, the inefficiency in performance of this algorithm limits its applications in the real world. With the advances in the engineering technology, massive parallelism can be achieved using the FPGA(1)-based techniques. In this paper, a parallel implementation methodology of the Smith-Waterman algorithm is presented. This method provides magnificent speedup over the traditional sequential implementation, while maintain the same level of sensitivity.