Fig. 2. Performance comparison for all competing algorithms on the yeast dataset.

a The performance of the methods on the task of predicting regulating TFs: for each algorithm, we compute for each gene the rankings of gold standard edges ${\overline{rank}}_i^g$ adjacent to it, sort them in descending order, and plot the sorted average rankings. b The performance of the methods on the task of predicting regulated genes using a measure similar to in (a) but focusing on genes regulated by TFs. c The performance of the methods on the task of predicting regulating TFs that are not observed in prior data. The procedure is the same as in (a) but only the gold standard edges that are not included in the prior knowledge are used for the evaluation. d The performance of the methods on the task of predicting regulated genes not observed in prior data (similar to (c) but focusing on genes regulated by TFs). e Overlap between priors. f Venn diagram for gold standard dataset and the union of the three prior datasets. g Summary of the average rankings of each algorithm for tasks reported in panels (a–d). The lowest scores are highlighted in bold red.