I am using DESeq2 for a while now, but I never see people talking about the use of the function "estimateSizeFactors" with the option "controlGenes".
For what I understand, it is supposed to be used with spike-ins, in which case DESeq2 will estimate the size factors only with the spike-ins and use it to normalize the counts of the other genes.
In my case, I am suspecting a global shift in gene expression, so is it correct to use internal controls (instead of spike-ins) in the option "controlGenes"? Or would you say that a global shift negates the premise of the algorithm?