NSF Funds Genomics Petascale Project
By Matthew Dublin
A team of researchers led by Georgia Tech’s David Bader are poised to bring genomic evolution to petascale heights. Equipped with a four-year $1M grant courtesy of the National Science Foundation’s PetaApps program, Bader and his team are setting their sights on developing algorithms to take advantage of petascale computing platforms for studying genome rearrangement events. Bader says that they will be writing their own open source software development framework from scratch and eventually releasing it through the GNU GPU license on Sourceforge. The team says they will start off the peta project by ramping up GRAPPA, an open-source application that reconstructs the evolutionary relatedness among species first released by Bader in 2000, to take advantage of petascale computers. “GRAPPA is currently the most accurate method for determining genome rearrangement, but it has only been applied to small genomes with simple events because of the limitation of the algorithms and the lack of computational power,” stated Bader in an announcement. The teamed used the latest version GRAPPA to analyze a dataset of a dozen bellflower genomes to determin the flower’s evolutionary relatedness one billion times faster than the original, non-parallelized version.
Worth checking out is Bader’s 2007 book entitled “Petascale Computing: Algorithms and Applications,” which covers possible applications for petascale systems, including molecular dynamics and biomolecules, as well as multithreaded algorithm design, performance analysis, Charm++, and the Cactus framework.
David A. Bader
Distinguished Professor and Director of the Institute for Data Science
David A. Bader is a Distinguished Professor in the Department of Computer Science at New Jersey Institute of Technology.