Title: Covariate-dependent graphical models
Authors: Yang Ni - Texas A&M University (United States) [presenting]
Francesco Stingo - University of Florence (Italy)
Veerabhadran Baladandayuthapani - University of Michigan (United States)
Abstract: Covariate-dependent graphical models are developed. The proposed model allows the graph structure to vary with covariates. We construct conditional independence function that maps from covariate space to sparse graph space. We will present both directed graph and undirected graph cases. Applying the proposed method to multiple myeloma data, we find interesting subject-level gene expression networks.