The RPKM and TPM measures correct for read depth and gene lenght, and work reasonably well when dealing with single copy genes. However, what if I want to mapp transcripts to a repetitive element (e.g. transposon)? How do I calculate transcript abundance in a way that I'm able to compare it between samples? I guess dividing the RPM values by the transposon length may not be OK because there are actually thousands of copies of this TE throughout the genome. Some of them may be transcriptionally active, others not. Should I summ all TE sequences lenght and use this for correction? Or maybe using just RPM values is fine in this case?
All ideas are wellcome. ;)