Bayesian Approaches for Regulatory Networks in Cancer

• Activity Number: 58
• Type: Topic Contributed
• Date/Time: Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
• Sponsor: Biometrics Section
• Author(s): Francesco Stingo*+
• Companies: MD Anderson Cancer Center
• Keywords: Genomics ; Graphical model ; Bayesian statistics ; Biological networks

Abstract

Recent advancements in Gaussian graphical models for the analysis of biological networks will be illustrated. First, in order to take into account the known biological nonlinear interactions, it will be shown how to construct gene regulatory networks that allow for both linear (Gaussian) and non-linear protein interactions. Then, a modeling approach for joint inference of multiple networks will be introduced. This approach allows us not only to share information between sample groups when appropriate, but also to obtain a measure of relative network similarity across groups. The application of proposed methodologies to the inference of protein networks will be illustrated, where the protein expressions were obtained using reverse phase protein arrays (RPPA).