Our group's work focuses on accelerating bioinformatics algorithms by exploiting their inherent parallelism. We explore ways to implement these algorithms on high-performance embedded computing architectures, such as field-programmable gate arrays (FPGAs), graphics processors (GPUs), and chip multiprocessors.
We have worked primarily on on accelerating algorithms for biological sequence comparison, including pairwise similarity search (BLAST), motif search (HMMER), and RNA folding. Our main implementation project is Mercury BLAST, an FPGA-accelerated version of the popular NCBI BLAST search tool. We consider both ways to realize existing algorithms and alternative methods with improved efficiency or better suitability for acceleration.
We also investigate many underlying issues of how best to realize computational biology algorithms on nontraditional high-performance platforms. Important research questions include:
The HPCB group is supported by awards from NIH's National Human Genome Research Institute and the National Science Foundation. Feel free to browse our research project summaries and our list of publications.