Bayesian Graphical Models for Gene X Environment Interaction

• Activity Number: 522
• Type: Topic Contributed
• Date/Time: Wednesday, August 7, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Paola Sebastiani*+
• Companies: Boston University
• Keywords: Bayes theorem ; directed acyclic graph ; graphical models ; marginal independence ; conditional independence

Abstract

Bayesian directed graphical models have been described as multivariate models that accommodate many interacting variables and therefore can be useful to describe many interacting genetic and non-genetic variants and their association with a complex genetic trait. We review different concepts of interactions and show that directed graphical models can conveniently represent biological interactions. We show how reading off these relations from a directed graph uses conditional independence between variables, and we review simple algorithms to check if two variables in a directed acyclic graph are conditionally independent given a third variable.