Peptides at the transcript end not excluded

[zhuchcn]

Based on a discussion that we had a while ago (maybe > a year) that we want to exclude the peptides that are at the very end of the transcript if the end of the transcript is not confirmed (with mRNA_end_NF). But there are two cases we need to consider:

1. Coding transcripts that end is confirmed and has a stop lost mutation
2. Noncoding transcripts also with end is confirmed.

For the above two situations, should we also ignore the peptides at the end of the transcript?

[lydiayliu]

So we are talking about coding and noncoding transcripts with known end but stop lost mutation?

I feel like 1) is worth a run-through until the end of the transcript.

[zhuchcn]

The question is about whether we should include the very last peptide or not. I guess it's a yes because the transcript sequence is known and it's also known that the transcript stops at the given position?

[lydiayliu]

I think overall if it's just the very last peptide it's not going to make a huge difference.

But there are a few different cases that we have to treat differently?
a) there is a single stop codon downstream
b) there are a few stop codons downstream
c) there are no stop codons downstream but there is no UTR
d) there are no stop codons downstream but there is UTR

I think c) and d) are the cases where we are worried about the last peptide?

for a) and b) we need to decide whether we want to stop at the stop codon

[zhuchcn]

For a peptide at the very end of the 3' end, there isn't anything downstream at all. For for coding transcript, the case would be, if there is a stop lost mutation that causes the 3' UTR being translated, should we include the peptide translated from the 3' end. For noncoding, there is not UTR at all, so the issue would be if there isn't any stop codon after the last start codon, should the very last peptide be included.

One situation that I can think of to not inclulde the last peptide is circRNA. In mpg, we represent the circRNA as linear graph with the sequence copied 4 times. But the very last peptides should not be included.

Agreed that this is not going to change much! But it's currently not handled consistently so I want to clean it up.

[lydiayliu]

OK we can separate coding and noncoding

Coding

• right now, if there is a stop lost mutation, do we stop at the next stop codon if there is one in the 3'UTR?
• if there is no stop codon at all in the 3'UTR, we translate all the way to the 3' end. I think it's fine to include the last peptide?

Noncoding

• we are talking about the case where there's no stop codon until the end of the transcript. I don't have an opinion, I think it is fine to include the very last peptide

[zhuchcn]

That aligns with my thoughts, too! How about circRNA? Do you think we should also include the very last peptide?