Figure 7.

Statistical significance testing of per-term evaluation. Each panel shows a directed graph in which nodes are methods and a directed edge from node A to node B indicates that method A performs significantly better than method B according to the Z-test described in the Statistical testing section of Materials and methods. Because Z-tests are transitive - unlike other tests such as the Wilcoxon signed-rank test - we represent the graphs as transitive reductions, that is, removing edges that are already implied transitively by other edges. BP, biological process; CC, cellular component; MF, molecular function; Ont, ontology.