Single Pass, BLAST-Like, Approximate String Matching on FPGAs

Abstract

Approximate string matching is fundamental to bioinformatics, and has been the subject of numerous FPGA acceleration studies. We address issues with respect to FPGA implementations of both BLAST-and dynamic-programming-(DP) based methods. Our primary contributions are two new algorithms for emulating the seeding and extension phases of BLAST. These operate in a single pass through a database at streaming rate (110 Maa/sec on a VP70 for query sizes up to 600 and 170 Maa/sec on a Virtex4 for query sizes up to 1024), and with no preprocessing other than loading the query string. Further, they use very high sensitivity with no slowdown. While current DP-based methods also operate at streaming rate, generating results can be cumbersome. We address this with a new structure for data extraction. We present results from several implementations.

Reconfigurable acceleration of genetic sequence alignment: A survey of two decades of efforts

“…The sensitivity of the prefiltering approach is tuned to exceed that of the NCBI BLAST implementation to ensure identical results. Experimental results show that, compared with NCBI BLASTn, the speedup is greater than 12× when pre-filtering and accelerator in [59] are used in execution.…”

“…In particular, Herbordt et al [59], [60] propose the use of a DP approach on FPGA to emulate the seeding and extension phases of BLAST. This algorithm, named TreeBLAST, can improve the performance of the database search by 400× on Virtex-4 LX160 FPGA compared to multiple-pass NCBI BLASTp on Xeon 2.8 GHz.…”

The Research on BLAST Accelerator with Parallel Multi-Seeds Detection and Extension

“…活的算法适应性和较低的硬件代价成为理想的算法加速平台。为了提高搜索效率，基于 FPGA 平台的并 行 BLAST 算法成为研究热点。 近年来备受关注的研究成果有 Mercury BLASTn [9]- [11]、 Mercury BLASTp [12]、RC-BLAST [13]、Tree-BLAST [14]、FPGA/FLASH [15]、Multi-engine BLASTn [16] …”

A Banded Smith-Waterman FPGA Accelerator for Mercury BLASTP

“…In addition to the S-W accelerators described above, recent literature reports several accelerators for BLAST-like seeded alignment algorithms [7,8]. None of these accelerators include a S-W stage, though [8] plans one for future work 1 .…”

“…None of these accelerators include a S-W stage, though [8] plans one for future work 1 . Moreover, none of them provide sensitivity measurements for their implementations versus the software that they are designed to replace.…”