Bayesian learning of multiple essential graphs

Structural learning of graphical models is a well-established approach to the identification of complex dependencies in biological networks. We here present a Bayesian methodology for learning directed networks from observational data when distinct subgroups of a population are observed.