Coloring features of GFF not working

[John Johnson]

Hi, I am using IGV 2.8.9 and I am trying to color features in my GFF (using gff2) but the feature isn't being colored correctly on the browser (it's still the default blue). The GFF is older and so is still in version 2 format, I'd like to keep it that way if possible without having to update to gff3.

Here is an example of how I have it formatted

##gff-version 2

##date 2021-01-25

chr2    ensembl exon    27928653        27931363        .       -       .       gene_id "TESTGENE"; transcript_id TESTGENE:1; color "#AAAAAA"

chr2    ensembl exon    27938451        27938599        .       -       .       gene_id "TESTGENE"; transcript_id TESTGENE:1; color "#AAAAAA"

When I click on the feature in IGV it pulls up a window that lists the attributes and their value and the color field is present with #AAAAAA being listed, but again it's not showing that color in IGV.

Many thanks!

[James Robinson]

What color is being shown?  Could you post a screenshot?

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/902e24f8-5750-4afd-9986-9cf349fee18cn%40googlegroups.com.

[James Robinson]

Also post an example file, whitespace gets mangled when pasting into email or a post

[John Johnson]

Here is a screenshot and an example file. The color should be grey according to the hex code, but you can see how it's still the default blue

[John Johnson]

Sorry, trying to figure out how to post an example file

[James Robinson]

Never mind, I was able to create the example file.  I will look into this.

To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/0ff4b8af-979f-4739-a1bc-ffd777eacb35n%40googlegroups.com.

[James Robinson]

The issue here stems from having 2 exons with no parent (no transcript).    IGV does not support coloring individual parts (e.g. exons) of a transcript.   To accommodate older GFF2 files IGV creates a transcript to tie them together if there is none represented in the file, but the phantom transcript doesn't get the attributes of its parts.   I think if you used GFF3 and set color on the transcript it should work.   But I'm not sure there is an easy fix for GFF2.

[John Johnson]

Thank you much! I'll give that a shot