Accelerating DNA Sequence Analysis using Content-Addressable Memory in FPGAs

Biological sequence alignment is a widely used technique where the sequence databases are searched to find similar sequence to the input query. In this work we focus on the most popular local sequence alignment algorithm; Basic Local Alignment Search Tool (BLAST). It is a computationally intensive operation, and with exponential growing databases, makes it further complex to execute in real time. Field-programmable gate arrays (FPGAs) provides hardware-like performance and software-like programmability which makes them the ideal candidate for computationally complex tasks. This paper presents a content-addressable memory (CAM)-based implementation of BLAST on FPGA that accelerates the alignment process using concurrent computations. The searching of the input query is performed in parallel across the database sequence to produce the result in one clock cycle. The proposed design is implemented on Xilinx Virtex-7 FPGA device XC7VX690TFFG1761. Results indicate better feasibility and accelerated performance (149–180 MHz speed) compared to the available searching algorithms.