Single-gene expansion reveals candidate pathway components. We compared the distribution of rankings from the single-gene expansion method and co-expression analysis. A. Genes were added to the network one at a time in all possible putative positions and the Bayesian addition scores (y-axis) were calculated for each new network (x-axis). The top scores for each gene were plotted from best to worst (grey line). The top 5% of the distribution (black line) consist of 209 genes. B. Similarity between genes and network members was computed by calculating all pairwise Pearson's correlations and the maximal correlation score (y-axis) was used to rank each new gene (x-axis). Using the cutoff determined by the single-gene expansion method, we focused our comparison on the top 5% of the distribution (black line).