FPGAAccelerator for DNA Sequence Alignment Based on an Ef ficient DataDependent Memory Access Scheme
Hasitha Muthumala Waidyasooriya, Masanori Hariyama, Michitaka Kameyama · 2014
The mapping of millions of short DNA fragments to a large genome is a very important aspect of the modern computational biology. However, software-based DNA sequence mapping takes many days to complete. This paper proposes an FPGA-based hardware accelerator to increase the mapping speed. We apply a data encoding scheme that reduces the genome size by 96%, and propose a hardware decoder to decode the data in single clock cycle. We also design customized data paths to increase the speed for random data access. According to the experimental results, the speed-up is 15 times compared to its equivalent software application.