Percent methylation per condition

[Alicia Vaquero]

Hi, is there a way to have the percent methylation per condition instead of the percent methylation per samples (I have 3 samples for each condition) ?

I want to do a scatter plot (% CpG methylation Control vs % CpG methylation Treatment) and have my 20% threshold difference lines, and to get to % methylation per condition I tried doing :
mean(% meth control sample1, %meth control sample2, %meth control sample3) and mean(%methylation treatment sample1, %methylation treatment sample2, %methylation treatment sample3)

The problem is that the difference of the calculated % methylation per condition is not the same as the percent methylation difference contained in the DMRs output file, and there is a big difference between the amount of DMRs from Methylkit vs the amount considered "DMR" only from the difference of the means that I calculated

If anyone knows how to calculate it, please tell me ! thank you