I want to correct for batch effects in my normalized BS and oxBS array data, before calculating the 5hmC / 5mC levels. However, comBat outputs only a beta matrix, so I don't have comBat-adjusted meth & unmeth values to input into oxBS.MLE. Would it be correct to input the normalized meth & unmeth values along with the comBat-adjusted betas to the oxBS.MLE function? Or perhaps is it possible to run MLE following normalization, then comBat on the 5hmC / 5mC levels? I'd love to know how other people are handling this!