A key biological question is to uncover the regulatory networks in a cellular system and to understand how this network varies across individuals, cell types, and environmental conditions. In this talk I will describe work that uses machine learning techniques to reconstruct gene regulatory networks from gene expression data. Specifically, we exploit novel forms of Bayesian regularized regression to enable transfer between multiple related learning problems, such as between different individuals or between different cell types. We demonstrate applications in two domains: understanding the effect of individual genetic variation on gene regulation and its effect on phenotypes including human disease; and understanding the regulatory mechanisms underling immune system cell differentiation.