Differential Gene Expression Analysis From TMM normalized counts

Question

[boyinepallyki]

Hi,

I understand that DGE analysis software like EdgeR or DEseq take in raw counts as input and then calculated a normalized factor for TMM. The data set I have only has TMM normalized counts and I do not have access to the original counts. Is there a way to bypass EdgeR's internal normalization? Can I use these TMM values for differential expression analysis and if so, how?

Thank You

Answer 1

[Gordon Smyth]

It depends on what you mean by "TMM normalized counts". If you have CPM values or log-CPM values output by edgeR's cpm function, then you could conduct a DE analysis with them using limma. If you don't have the original counts then you cannot use edgeR.