A guide to the literature on inferring genetic networks by probabilistic graphical models

Abstract

In this chapter we discuss the advantages of the use of probabilistic graphical models for modelling molecular networks at different levels. We also provide an overview to the literature on inferring genetic networks by probabilistic graphical models. Different types of probabilistic graphical model – Bayesian networks, Gaussian networks – are introduced and methods for learning these models from data are presented. These models provide a concise language for describing joint probability distributions by means of local distributions. This fact and the possibility of reasoning inside the model, apart from their declarative nature, provide an advantage to inferring molecular networks and to transforming heterogeneous data sets into biological insights about the underlying mechanisms.