A hierarchical model of expression across multiple genes within a cluster. Each row represents one gene (gene names are to the right) and each box within that row represents a biological replicate. Data are represented as black points, the shaded area represents 95% confidence interval and a solid line represents a posterior mean function. The left-most box on each row shows the inferred function for each gene g
(t), and the bottom row shows the mean function for the cluster h(t) (left) and the predicted function(s) for a hypothetical new gene.