I am trying to fit a distribution of the gene expression (DESeq Normalized counts) of each sample from a RNA Seq data. And I came across this paper where they describe that the expression levels of genes have a Bimodal distribution. .
I was wondering would this be true for the count data as well or do we expect any specific distribution for Counts that will have a change of skew that may arise due to the length factor that comes in RPKM?
Can some one help me here on how to test this for count data?