A2 - Phylogenetic Tree Reconstruction Using FPGA

Virology is an important subject in medical research. It helps to understand the origin and nature of different illnesses. Based on the understanding of a virus, scientists can produce new drugs to fight against it. This in turn could help control its associated illnesses.

One way to investigate a virus is to analyze its DNA structures. However, the analysis is often complex and time consuming. Traditionally, analysis of a DNA sequence can take hours even by using the most powerful computers. This is very inefficient and would seriously affect the progress in virology advancement.

To circumvent this problem, we have developed a mixed hardware/software computational platform using FPGA for real-time searching of structures in DNA sequences. Compare with existing software methods, our system can perform 100 times faster.

Applications

• Development of an efficient phylogenetic tree reconstruction tool
• The tool offers support to DNA research in virus identification and drug discovery

Features

• First hybrid solution, by integrating software flexibility with FPGA hardware speed
• Massive parallelism for maximum likelihood calculation
• Current work offers potential for large-scale tree reconstruction

Principal Investigator
Prof. K. P. Lam
Department of Systems Engineering and Engineering Management