Underdispersion issue in RNA-Seq data

Question

[Naim.Mahi]

Dear Altruists,

In most of the Rna-seq studies, people consider the data as overdispersed. I was looking for a paper that deals with underdispersion as well, but couldn't find any.

My question is, why people usually do not consider underdispersion in rna-seq data? Why they assume that Rna-seq count data is overdispersed? Is this because biological samples always shows overdispersion in the data? If so, then why?

Any reference would be really helpful.

Thanks in advance.

Answer 1

[Chris Miller]

I think your last point is the most salient - we just don't observe underdispersion in practice. Start with the first answer on this previous question and see if it provides some clarity on why this is the case:

Why Does Rna-Seq Read Count Fit Poisson Distribution?

[Naim.Mahi]

Thanks Chris.