Our Research Group
High-dimensional inference for modeling cancer
We study high-dimensional (genomic) data in order to develop predictive anddescriptive models for cancer. There are two main components to our research: 1) statistical inference of the geometric/statistical dependence structures inherent in the data; and 2) use of these dependency structures to both predict disease or treatment outcomes and understand the biological mechanisms that underlie the course of disease progression.
The high-dimensional system we focus on is cancer. Specifically we focus on the combination of different types of high-throughput data, which include (but are not limited to) data on copy number variation, gene expression variation, and sequence variation. We are currently using these data sets to develop models for tumorigenesis, drug sensitivity, and other complex phenotypes using methods from Bayesian statistics, machine learning, differential geometry, topological data analysis, and computational harmonic analysis.
Our long term goal is to develop a modeling framework that not only explains variance in the occurrence and progression of disease, but also addresses the underlying mechanisms through which complex traits arise.
Thanks for visiting our website! Please contact us if you are interested our software, our methods, or pursuing collaborative research. If you are a prospective graduate student or post-doc, please check out the information page in the "Lab members" section.
I am an Assistant Professor with appointments in
Department of Statistical Science
Institute for Genome Sciences & Policy
Department of Computer Science
Department of Biostatistics and Bioinformatics.