Entering edit mode
8.5 years ago
tonja.r
▴
600
I was using RUVseq for normalizing my data and got stuck at the interpretation step. RUVg normalizes the data against the empiricial or negative control genes. So, that RLE will be similar across the samples. Till this point it is clear for me. My problem starts with RUVs and RUVr.
RUVr normalizes using residuals but it assumes that the true biological effects are larger than the unwanted variation. Why do we do this assumption when working with the residuals?
RUVs normalizes the data within each condition and against the negative control samples. And in fact RLE will be similar within the conditions. Is it right?
