Modern medicine has graduated from broad spectrum treatments to targeted therapeutics. New treatments attack specific genetic pathways which are only dysregulated in some smaller subset of patients with a disease. Unfortunately this subset is often poorly understood. Building/discovering biomarkers to characterize this subset is key to personalizing medicine.
Our goal in this project is to screen [potentially high dimensional] biomolecular features, to find those few that are clinically informative. This work combines aspects of statistics, computer science, and biomedical science.