Dear all, 

I want to use deconvolution methods to estimate the proportions of different cell types in my RNA-Seq samples. In this post ( https://www.biostars.org/p/121286/ ), it's mentioned that "signals from different cell-types/tissues will sum more linearly in microarrays than RNAseq, where the sum is highly non-linear" and  "Any paper talking about signal separation will likely mention that the signals need to be independent for optimal performance, which they self-evidently aren't in RNAseq." Could someone please explain to me why in RNA-Seq samples the signals from different cell-types/tissues are not independent, or why the signals don't sum linearly?

Also, if I do decide to go ahead with using deconvolution methods, should I apply the deconvolution methods to raw RNA-Seq counts, log(CPM) transformed data, or voom transformed data?

Thanks. 

Paul

Don't have thorough answers for you yet, but I did poke Devon over at Biostars to clarify the "self-evident" nature of that statement.

A few other notes:

• It has been suggested/shown that deconvolution performs better on the natural scale (ie. you should not operate on log-transformed data)
• There already is a bioconductor package that does deconvolution using RNA-seq data: DeconRNASeq
• I'm not aware of any published manuscripts that have actually used any such deconvolution methods using RNA-seq data. If you have any on hand, I would be happy to check them out.