Meet The Speaker
“Computational Challenges in Molecular Medicine”
Donald Geman received a B.A. degree in Literature from the University of Illinois and a Ph.D. in Mathematics from Northwestern University. He was Distinguished Professor at the University of Massachusetts until 2001, when he joined the Department of Applied Mathematics and Statistics at Johns Hopkins University, where he is a member of the Center for Imaging Science and the Institute for Computational Medicine. He is an IMS and SIAM fellow.
Seminar Abstract
“Computational Challenges in Molecular Medicine”
The goal of translational medicine is to carry fundamental research into clinical practice. This is very ambitious and rarely realized. I will talk about computational barriers in attempting to carry out this program in the context of statistical learning applied to cancer genomics with high-throughput “omics” data. In particular, I will focus on learning predictors of disease phenotypes, tumor sites-of-origin and pathway deregulation with limited sample sizes and massive study-to-study (non-biological) variability. Finally, I will argue that low-dimensional order statistics are easy to interpret, can account for combinatorial interactions among genes and gene products and might even support a mechanistic interpretation for the underlying decision rules. Most of this is joint work with Nathan Price, Luigi Marchionni, David Simcha or Bahman Afsari.
Note: Light lunch will be served starting at 1:00pm. The presentation will begin at 1:30pm.
