For my research purpose I wrote a small simulation to compute an empirial p-value based on an input gene list and a reference gene list. The idea is to check for each gene in the input gene list, the associate score in the reference gene and to do the sum. After that to create N=100000 simulated list of the same size as the input gene list by picking randomly genes in the complete gene list, and to compute their associate score. At the end I can compute a p-value by counting the number of simulated list having a score superior or equal to the input gene list, then divide by N=100000.
Now the problem, is that I have several input gene lists with different number of genes; and different reference gene list with different number of genes. How can I compare the results of these different simulations ( I know that p-values can not be compared .. )
I saw that effect size can help me; and saw this figure in "Effect size, confidence interval and statistical significance: a practical guide for biologists", Nakagawa et al. How can I plot such graphic with my different input gene list an reference gene lists