Can't display exon features in a gff3 file

[KenN]

Hi,

I've been trying to set up a simple gff file to display in IGV. While
my mRNA does display, I can't seem to make IGV display the exons. Here
is a sample of the gff file:

####################################
##gff-version 3
##sequence-region scaffold230113 1 13043
scaffold230113 gmap gene 1 13043 . + . ID=gene0001;Name=scaffold230113
scaffold230113 gmap mRNA 12342 2930 . - .
ID=mRNA0001;Parent=gene0001;Name=DCL2
scaffold230113 gmap exon 12341 12272 . - .
ID=exon0001;Parent=mRNA0001;Name=exon1
scaffold230113 gmap exon 12151 11974 . - .
ID=exon0002;Parent=mRNA0001;Name=exon2
####################################

All I can get IGV to display is the 'gene' feature and under that the
'mRNA' feature. But it won't seem to display the 'exon' features (the
gff file i'm using has 21 exons). I've looked through several GFF
format sites (e.g. http://www.sequenceontology.org/gff3.shtml) and I
don't think I'm doing anything out of the ordinary.

The reference file is just a fasta file with various 'scaffold'
sequences, and I'm trying to annotate whatever genes each may
contain.
I'm using version 2.0.23.

Any help appreciated,
Thanks!
Ken

[James Robinson]

This looks legit, could you send your gff file? Sometimes whitespace or other format problems can cause this and I can't determine if that's the case without the file.

I suggest you upgrade, many bugs have been fixed since 2.0.23, although I don't recall fixing this one. The current build is 2.0.34.

-- Jim

[Jim Robinson]

Hi,  I replied by private email but the start positions of these exons are greater than the end positions,  this is likely the problem.

[KenN]

Yes, that fixed the problem. Many thanks.

[Alex]

Ken, could you please post the correct version of GFF3 since I had
exactly the same issue. I also wonder how to correctly delineate UTRs
in GFF3 gene presentation, the GBrowse definition on that changes.
Thanks, Alexander

[Jim Robinson]

Hi Alex,

I can't really tell you the "correct" way to delineate UTRs,  I've seen this done many ways,  we try to correctly parse them all.  If you follow the "canonical gene model" as described here it should work with IGV at least http://www.sequenceontology.org/gff3.shtml.  Note that UTRs are implied here,  by combining the CDS with exons.   Alternatively you can define the UTRs explicitly.  The following terms are recognized,  I suspect they are not all legal SO terms but they have all been seen in one GFF or another.

five_prime_UTR
three_prime_UTR
5'-utr
3'-utr
3'-UTR
5'-UTR