Accelerating Protein Structure Prediction with FPGA
Author Information
Author(s): Xia Fei, Dou Yong, Lei Guoqing, Tan Yusong
Primary Institution: National University of Defense Technology
Hypothesis
Can FPGA technology significantly speed up the GOR algorithm for protein secondary structure prediction?
Conclusion
The FPGA implementation achieved over 430 times speedup compared to the original GOR-IV software version.
Supporting Evidence
- The FPGA implementation achieved a speedup factor of more than 430x over the original GOR-IV version.
- Power consumption of the FPGA accelerator is only about 30% of that of current general-purpose CPUs.
- The study utilized a dataset of 6,028,192 sequences and 1,207,406,759 amino acids.
Takeaway
This study shows that using special computer chips called FPGAs can make predicting protein structures much faster than regular computers.
Methodology
A fine-grained parallel hardware implementation on FPGA was developed to accelerate the GOR-IV package for 2D protein structure prediction.
Limitations
The study does not address the scalability of the FPGA implementation beyond the tested configurations.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website