Read coverage bias

Question

[ajaymaran97]

I have a list of 12000 transcripts out of which 1500 are novel. I performed an HTSeq count to get the read counts for all the 12000 transcripts. I segregated the 1500 novel counts and other non novel. Normalized both using TPM, however I find that the novel list has far higher TPM values compared to the non novel ones. Why is my result biased?

Answer 1

[Trivas]

You need to calculate your TPMs before splitting your novel transcripts from non-novel transcripts. This will allow you to directly compare the abundance between them.

[ajaymaran97]

Thanks a lot Trivas. My results seem to be right now.