SSW is a fast implementation of the Smith-Waterman algorithm, which uses the Single-Instruction Multiple-Data (SIMD) instructions to parallelize the algorithm at the instruction level. SSW library provides an API that can be flexibly used by programs written in C, C++ and other languages. We also provide a software that can do protein and genome alignment directly. Current version of our implementation is ~50 times faster than an ordinary Smith-Waterman. It can return the Smith-Waterman score, alignment location and traceback path (cigar) of the optimal alignment accurately; and return the sub-optimal alignment score and location heuristically.
0 commit comments