Count Vs Dispersion Plot

Question

[Eric]

I am analyzing my RNA-seq data with cummeRbund. I have four conditions, 5 biological replicates per condition. Why do the count vs dispersion plots look more or less identical for all four conditions? I can see from the cummeRbund vignette that this is supposed to be the case; I just don't know why.

Any insights would be appreciated.

Thanks.

Answer 1

[Sean Davis]

The count-vs-dispersion relationship is roughly governed by the biological counting process used to quantify RNA-seq data. You might take a look at papers by Simon Anders and Wolfgang Huber (authors of DEXSeq and DESeq) for some background and details.