Figure 3

Varying the relative expected degree between sets A and B. Here we are varying the expected degrees of sets A and B. The GeneRank algorithm provides slight improvement over pure expression based ranking when the expected degree of boths sets is equal. As the expected degree of set A becomes larger compared to that of set B, the improvement observed over expression ranking increases. We are measuring the AUC of the ranking (max = 1), averaged over 5 experiments. The lines on the diagrams indicate constant values of the AUC. The black cross indicates where the maximum AUC occurs, and hence shows for what values of d and relative connectivity the method works best.