[FEATURE] Beta or M values as input

[Rut-MDLV]

Hi,

I have been testing the code on my methylation data (approx. 260 patients and controls) from a complex disease, and fails to correct my batch effect (which is strong), however, Combat and RUVm eliminate it. I was wondering, why using beta values which are recommended for visualization and understanding, instead of M values which are stronger for statistical testing? Any idea on what could be happening?