When performing a blind=FALSE vst or rlog transformation on the data, it returns log2 values, that have been "corrected" by accounting for gene expression noise.
Similarly, lfcShrink also returns log2(FC) values that have been shrunk due to gene expression noise?
I therefore wonder, are these two processes conceptually similar? are they in fact conceptually identical? If not, I'd like to gain an understanding for what their differences are.