Fig. 8

Summarizing the simulation and predictions of POWEREDBiSeq. Ai) A QQplot comparing the read depth (RD) of 10,000 simulated DNAm points to 10,000 randomly sampled true DNAm points from an RRBS dataset. Aii) The proportion of DNAm points remaining in the RRBS dataset with read depths >x. B) A QQplot comparing the DNAm of 10,000 simulated DNAm points to a 10,000 randomly sampled true DNAm points. C) The relationship between the difference in power predicted by POWEREDBiSeq at different minimum sample sizes (n = 2, 30 and 60) as the minimum read depth threshold is increased, with a mean difference between groups of 0.06. D) The relationship between the increase in power to detect a mean difference in DNAm between groups of 0.06 predicted by POWEREDBiSeq at a read depth of 75 compared to power at a read depth of 1 as a function of the number minimum of samples per group