Schematic model of regulome construction. Two input sources are selected, e.g a set of TF tracks and a set of disease tracks. For each combination, the pairwise relation model, in this case the number or genes containing TF binding locations, is evaluated and subsequently differentiated against the full matrix of counts. The main output is an interactive heat map of over-/under-representation that for each entry also includes detailed information and links to follow-up analysis. The regulome construction is performed by a web-based system, the Genomic HyperBrowser , that allows input data, a pairwise relation model and a measure of deviation to be selected.