Knowing how our bodies are genetically composed is extremely important to keeping ourselves healthy. The ability to translate DNA information into a quantitative prediction of the risk to disease or observed phenotype is critical.
Professor Jinbo Bi is currently leading a project collaborating with Professor Sanguthevar Rajasekaran entitled A High Performance Computing Foundation to Whole-Genome Prediction to address the urgent need for new statistical models and high performance computing foundations that allow the concurrent use of millions of genetic markers and a large variety of heterogeneous variables describing a disease or a breeding target.
Precision medicine can benefit from the prediction of an individual’s genetic risk to disease. Animal and plant breeding programs can benefit from the selection of individuals based on genomic information. There are, however, technical barriers in the prediction using a whole-genome sample of genetic markers. Professor Bi’s solution is an integrative approach to combine and develop techniques for dimension reduction, parallel and distributed computing and Bayesian inference. Her proposed solutions will be tested in the analysis of large-scale biological and genomic datasets, including a dairy cattle study database collected by the U.S. Department of Agriculture and a disease database aggregated from multiple genetic studies of human diseases.
The project, funded with a $750,000 grant from the National Science Foundation Directorate for Computer and Information Sciences and Engineering, will systematically examine the feasibility and usability of genomic risk prediction, and also yield user-friendly software tools that will be broadly disseminated to the life science communities.