On Efficient Implementation of Search Algorithm for Genome Patterns

The presented paper describes the implementation of the computation algorithm on modern, complex programmable hardware devices. The presented algorithm originates from computation biology and works on very long chains of symbols, which come from reference patterns of the genome. The software solutions in the field are very limited and need large time and space resources. Main research efforts have been done to investigate the properties of the searching algorithm. Especially the influence of the penalty values assigned for the mismatch, the insertion and the deletion on the algorithm has been analyzed. This allows obtaining completely new algorithm that offers extremely efficient implementation and exhibits outstanding performance. The different FPGA generations have been considered as target families for the searching algorithm based on the dynamic programming idea. The obtained results are very promising and show the dominance of the dedicated platforms over the general purpose PC-based systems.