Abstract:
|
DNA microarray studies provide an opportunity to study thousands of genes at the same time. Gene-Set Analysis is a popular approach to examine the association between gene expression of a predefined gene-set and a phenotype. However, often not all the genes within a significant gene set contribute to its significance. Identifying the core subset enhances our understanding of disease biological mechanism, provides additional insights into disease progression and treatment strategies. Many methods have been proposed for a binary outcome (diseased versus disease-free subjects), but only a few for continuous phenotype (tumor size). The challenges consist of large number of genes in a set, small sample size and accommodating correlations between genes across the set. We developed a powerful method to reduce gene-sets associated with a continuous phenotype. The method is based on the Linear Combination Test (LCT) for gene-sets, which incorporates gene expression covariance matrix into the test statistic, via a shrinkage estimation approach. We apply our method on real microarray data to extract core subsets.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.