Figure 5From: A hierarchical Bayesian network approach for linkage disequilibrium modeling and data-dimensionality reduction prior to genome-wide association studiesAssociating a latent variable to any clique of variables in the undirected graph of dependence relations (a) Three pairwise dependent variables (clique). (b) Latent model: the three variables depend on a common latent variable. The dark shade indicates the latent variable designed to model the pairwise dependence between the three variables.Back to article page