Accelerating protein tertiary structure analysis based on FPGA

Abstract

In the era of tertiary protein structure prediction, the most successful protein structure prediction methods involve sampling protein conformations which is a computational problem due to the increasing scale of the protein database. Recently FPGA chips have emerged as one promising application accelerator to accelerate bioinformatics algorithms by exploiting fine-grained custom design. In this paper, we propose a complete fine-grained parallel hardware implementation on FPGA XC5VLX330 chip to accelerate the BackboneDBN program for sampling realistic protein conformations using local structural bias. The experimental results show a speedup factor of more than 20× over software version running on a PC platform with Intel E7400 dual-core. However, the FPGA's power consumption is only about 30% of that of current general-purpose CPUs.