Events and Seminars : 2013 Seminars

Some new statistical methods for pathway analysis of high-throughput genomic data

Xi (Steven) Chen, PhD Division of Cancer Biostatistics
Department of Biostatistics
Vanderbilt University

Friday, August 16, 2013
12:00 PM – 1:00 PM
SCCC 1301 Conference Room

Hosted by Sylvester Comprehensive Cancer Center

Recently, high-throughput genomic technologies, such as gene expression microarrays, single nucleotide polymorphism arrays and next-generation sequencing have revolutionized biological and medical research by making it possible to measure thousands to millions of biomarkers across the genome simultaneously. To reduce dimensionality, and to increase statistical test power, pathway (or gene set) analysis has become increasingly popular. Instead of applying statistical tests to one gene at a time, pathway analysis takes advantages of previous biological knowledge and examines the gene expression patterns of groups of functionally related genes for their associations with disease outcomes. We will discuss several new statistical methods for pathway analysis that are based on supervised variable selection, unsupervised variable selection and ensemble tree-based approaches.