ALBUQUERQUE, N.M. — A final step for both the University of New Mexico (UNM) and Keck projects is analyzing, archiving, and annotating the data to determine gene relationships. Many approaches have been taken in these analyses.
For instance, researchers from UNM, led by biology professor Maggie Werner-Washburne, are using high-speed computers at the Computing Education and Research Center (HPCERC) to analyze the data with statistical techniques, Bayesian networks, and vector-support machines. The Sandia team is using tools from researcher David Haaland’s chemometrics research, linear and nonlinear classifiers and vector support machines and Sandia’s VxInsight data-mining software, which has also been successfully used at other genomic research centers.
“We’ve used VxInsight for a long time at Sandia for various projects,” said Sandia researcher George Davidson. “To make it applicable to genomics, however, the software had to be made even more sophisticated. Some of that development was the direct result of working with Maggie. Other parts sprang from our collaboration with Stuart Kim at the Stanford Medical School, who uses VxInsight for his genomic research, and many of the recently added capabilities were required to analyze the leukemia data. These same techniques will be useful in the new Genomes to Life research that has just been awarded.”
Davidson said he believes the Keck project — well into its second of three years — has accomplished much and is the start of bigger things.
“Sandia is poised to begin making major contributions in biology; the researchers in New Mexico and California have a real opportunity now that we have been awarded the Genome to Life grant,” Davidson said. “Certainly, the larger Sandia experience and capabilities grew from many sources, but David“s and my research benefited from the UNM collaborations.”
University of New Mexico Contact: Steve Carr, scarr@unm.edu, (505) 277-1821