FPGA acceleration of short read alignment with high-level synthesis

Field-programmable gate array (FPGA);Hardware acceleration;High-level synthesis;OpenCL;Short read alignment

Chen, Daniel E ; Chen ; Deming

关键词: Field-programmable gate array (FPGA);    Hardware acceleration;    High-level synthesis;    OpenCL;    Short read alignment;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/97641/CHEN-THESIS-2017.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

With the introduction of next-generation sequencing (NGS) technologies, DNA sequencing is becoming an increasingly widespread process. When performed on human patients, it can allow for the prediction and prevention of diseases. An essential part of this bioinformatics pipeline is short read alignment}, which refers to aligning short fragments of DNA to the large and expansive reference genome. This can be a very time-consuming process with much room for improvement. This thesis improves on Bowtie 2, an aligner that is already very popular and high-performing. Through the use of OpenCL, it is possible to parallelize this application for both GPU and FPGA by using the same code. Several different levels of parallelism are implemented in order to achieve speedup on Bowtie 2.