Accelerating a mutational robustness algorithm using FPGAs

Robustness is a  fundamental property of biological systems which expresses a system’s ability to maintain function in the face of mutational or environmental challenges. From the genomic evolution standpoint, the robustness of a DNA strand is measured by computing the Gibb’s Free Energy of folding on evolving populations of strand permutations. On a typical mutational robustness algorithm, a large number of arithmetic operations (O(n^2)) must be performed for each population and many populations must be generated, thus requiring considerable execution times on sequential algorithm implementations. The increasing quantity and availability of genomic data makes faster implementations of these algorithms a necessity. Computations on population members are data independent, which make these algorithms good candidates for improvement by using the parallelism available in field programmable gate arrays (FPGAs).  In our presentation we will discuss our initial efforts to implement time-critical parts of a mutational robustness algorithm to FPGAs.