Skip to main content

Log-linear models for Gene Interactions

We describe a Bayesian approach to the analysis of gene and network interaction data based on log-linear models. The goal of our approach is inference regarding causal connections among genes and functional pathways, and is based on a Bayesian model selection algorithm applied to data that has been reduced to ordered categories using ranks of observations within both subjects and genes or network components. This is joint work with Jianhua Hu.