Accelerating the Next Generation Long Read Mapping with the FPGA-Based System.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Comparing newly determined DNA sequences to database sequences is crucial in bioinformatics.
• Current aligners efficiently handle short sequence reads but struggle with long reads (>400 bp) from next-generation sequencing (NGS).
• Increasing read lengths and volumes in NGS necessitate improved long-read alignment performance.

Purpose of the Study:

• To propose and evaluate a novel FPGA-based system for enhancing the efficiency of long DNA sequence read mapping.
• To address the performance bottlenecks in aligning long reads generated by advanced sequencing technologies.

Main Methods:

• Development of a novel Field-Programmable Gate Array (FPGA)-based system.
• Comparison of the FPGA system's performance against the state-of-the-art long-read aligner, BWA-SW.
• Experimental evaluation of the system for reads ranging from 512 to 4,096 base pairs.

Main Results:

• The FPGA-based system achieves a 10x-48x speedup on the software's bottleneck for long reads.
• The overall mapping procedure shows a 1.8x-3.3x speedup compared to BWA-SW.
• The system maintains high performance with sensitivity comparable to BWA-SW.

Conclusions:

• The proposed FPGA-based system offers a significant performance enhancement for long DNA sequence read alignment.
• This acceleration drastically reduces the time required for complex genomic analyses, from weeks to days.
• The developed platform provides a viable solution for the growing challenges in processing large-scale genomic data.