The average <i>p</i>-values obtained using Wilcoxon rank sum test on the areas under <i>β</i>₀ curves for small (<i>p</i> = 10) and large (<i>p</i> = 100) network settings.

Here, the columns 6 networks, 8 networks, 10 networks, and 12 networks indicate the number of networks that we considered for both the groups. The p-values smaller than 0.01 indicate that our method can identify network differences at a 99% confidence level.