Question about using callDMR or callDML for paired design? #40
Description
@haowulab
Hello, I am performing a paired design. Do we not use callDMR or callDML for paired design?
I noticed that my dmlTest data frame already has chr,pos,stat,pvals, and fdrs (and it is given by loci), and I can sort this dataframe by the significant sites. Do you think it would be appropriate to perform "BH" method for adjusting p-values?
I tried using "dmrs <- callDMR(dmlTest)" and I got the following Warning message: "In callDMR(dmlTest) : No DMR found! Please use less stringent criteria." Is it my data here, or is the DMR function not made for paired design? (I am only doing the analysis on one chromosome at a time due to my dataset being very large, so maybe I'll find DMR in other chromosomes).
Thank you for your tool!
Kristen