Hi,
I would like to know if it is possible to use EdgeR for differential expression analysis starting from data that are no raw count, but are normalized counts.
I just try to skip the command:
d <- calcNormFactors(d)
is that way correct? or there is another way?
Thanks!
Kindly,
Enrichetta
To be on the safe side, I'd assume you can't use edgeR unless you have raw counts.
If you want to walk on the wild side, you apparently can use edgeR with normalized counts -- if you have the right type of normalized counts:
A: Can I feed TCGA normalized count data to EdgeR for differential gene expression
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
To be clear, you don't necessarily need raw integer counts, but you do need a measure that is on the same scale as raw counts. So your data cannot be normalized for gene length or for sequencing depth. If they have been, you would need to "undo" those normalizations before passing the data to edgeR.