Incorporating Gene Networks into Statistical Tests for Genomic Data via a Spatially Correlated Mixture Model

• Activity Number: 32
• Type: Contributed
• Date/Time: Sunday, August 3, 2008 : 2:00 PM to 3:50 PM
• Sponsor: ENAR
• Author(s): Peng Wei*+ and Wei Pan
• Companies: The University of Minnesota and The University of Minnesota
• Keywords: ChIP-chip ; Conditional autoregression ; Markov random field ; Microarray ; Mixture model ; Spatial statistics

Abstract

It is a common task in genomic studies to identify genes satisfying certain conditions, such as differentially expressed genes. Most existing approaches treat genes identically and independently distributed a priori, testing each gene independently. However, it is known that the genes work coordinately as dictated by gene networks. We propose incorporating gene network information into statistical analysis of genomic data. Specifically, we assume that gene-specific prior probabilities are correlated as induced by a gene network: those neighboring genes in the network have similar prior probabilities, reflecting their shared biological functions. We applied both standard mixture model and the proposed method to a real ChIP-chip dataset (and simulated data). The new method was found to be more powerful in discovering the target genes.