IsoQuant icon indicating copy to clipboard operation
IsoQuant copied to clipboard

Using IsoQuant counts with DESeq2?

Open gaoxiangcao opened this issue 5 months ago • 4 comments

Hi, thanks for the great tool! Can I use the transcript-level counts from IsoQuant as input for DESeq2 for differential expression analysis? Thanks!

gaoxiangcao avatar Aug 07 '25 23:08 gaoxiangcao

Dear @gaoxiangcao

Yes, I don't see any reasons why not. Just keep in mind that compared to short reads, long reads don't require gene length normalization when counting TPMs or other normalized values.

Best Andrey

andrewprzh avatar Aug 08 '25 20:08 andrewprzh

Dear Andrey,

Thanks for the reminder! Yes, as far as I understand, DESeq2 does not perform gene length normalization by default.

Best, Gaoxiang

gaoxiangcao avatar Aug 08 '25 23:08 gaoxiangcao

Hello @andrewprzh,

While reporting TPM values does isoquant normalize only on sequencing depth. Could you please confirm that it does not normalize on gene/transcript length?

Thanks Sagnik

sagnikbanerjee15 avatar Sep 10 '25 04:09 sagnikbanerjee15

Dear @sagnikbanerjee15

Yes, you are right. IsoQuant only normalizes by depth.

I think the common practice is that, unlike for short reads, counts obtained with long reads do not require gene/transcript length normalization, as one read typically represents one mRNA molecule.

Best Andrey

andrewprzh avatar Sep 10 '25 10:09 andrewprzh