Question: Can limma's voom take normalized counts as input?
gravatar for vkartha
3.8 years ago by
United States
vkartha0 wrote:


     I am currently analyzing some RNASeq data from TCGA - which is data that has been processed by RSEM and normalized using a basic quantile normalization. Having this data, I wanted to perform some differential expression testing comparing two groups  and thought of using limma to serve this purpose. 


However, I am currently adding a pseudocount of 1 to all my normalized counts prior to taking the log2 transform. While this is an easy solution - it is not always the best one. I realize voom transforms are meant to pre-process the RNASeq counts prior to running a linear model, but can you give voom normalized counts instead of raw counts? The documentation makes me think that it only takes in raw counts, and does the normalization internally. 


Any feedback would be much appreciated



rna-seq limma • 3.7k views
ADD COMMENTlink modified 3.8 years ago by fubar1.1k • written 3.8 years ago by vkartha0
gravatar for fubar
3.8 years ago by
fubar1.1k wrote:

From my reading, although the online docs for voom (like edgeR) indicate that it wants raw counts, in the package author says "the RSEM expected counts from the TCGA project will work fine with either limma-voom or edgeR" .

So expected counts will apparently work - if that's not exactly what you plan to use, I'd recommend you check the experts at for an authoritative answer - they don't generally read this Galaxy specific forum ?

ADD COMMENTlink modified 3.8 years ago • written 3.8 years ago by fubar1.1k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 173 users visited in the last hour