problem with translations reverse CDS features from GFF imports

[Martin H]

When importing a genome in gff format, IGV seems to have problems rendering the translation for CDSs with a reverse orientation:
It seems only to want to translate regions which have a '0' as phase attribute in the GFF record
It seems to do the translation with that phase 0 from the left hand side of the sequence, which results in wrong (frame shifted) translations
if the intron/exon boundary does not coincide with a codon end.

[Jim Robinson]

This was a bug, its now fixed (IGV release 2.0.33). Please let me
know if you continue to have problems. I used the attached gff file as
the test case (Plasmodium 6.1).

Jim

[Martin H]

Hello JIm,

Great - this works now as expected.

Thanks, Martin

[saxxse]

Hello Jim,

we are using IGV version 2.3.72 and this problem is still present. Was this bug re-established in a later version?

Thanks in advance,

Thomas

[Jim Robinson]

Hi, I don't know, you realize you're referencing a conversation that is 4 years old?   

Inspecting the code I can see a potential problem if the total length of the coding nt sequence is not a multiple of 3.   I can't imagine a legitimate situation where it would not be,  but that could cause a problem.   Could you open a git issue describing the problem as you see it, along with a preferably small gff file to reproduce the problem?  You can open the issue here:  https://github.com/igvteam/igv

Jim