Using RPKM to compare gene expression between samples

Question

[kz]

Hi all, wondering if normalisation using log2(rpkm + 1) is suitable for comparing gene counts between samples and groups. I see this quite often in pubs, e.g.:

But I came across this, which says RPKM is for "gene count comparisons between genes within a sample; NOT for between sample comparisons or DE analysis"

Thanks!

Answer 1

[dsull]

The pubs that do things like that are doing things incorrectly. If you're comparing between samples, there are many programs written for that purpose (e.g. deseq2, sleuth, edgeR, etc.)