Figure 6
From: Assessing the functional coherence of modules found in multiple-evidence networks from Arabidopsis
Schematic diagram showing how entropy provides a useful metric of fragmentation of a given GO term across clusters. If 20 proteins are associated with a given GO term and they all are in the same cluster then the entropy (H) is zero. If most (16) of the proteins are in one cluster and the remaining proteins are in separate clusters H = 0.338. However, as the proteins get more evenly distributed across clusters the entropy increases.