Hi everyone, I did a DGE on a count matrix of 7 sample classes (6 Tumor and 1 Healthy) and N Genes. I used edgeR and normalized with TMM. I know that in edgeR we do normalize the library and not the counts but for my further analysis that is Differential Co-Expression I need the normalized count matrix. I actually used the following :

counts.norm <- cpm(y,prior.count=1,log = TRUE)

Where y is the DGEList to which I have previously applied :

y <- calcNormFactors(y,method = "TMM")

The final results after all analysis and classification are great , but I'm scared that the task "get the normalized count matrix" was not respected because for what I know cpm(DGEList, .... ) is used only for visualization purposes and I dont know If I had to use it in that way. Because for my task (Differential Co-Expression Analysis) I need the normalized count matrix , or at least I think so.