I will modify the gff file as you mentioned and update galaxy.

Thanks a lot.



Yec'han LAIZET
Plateforme Genome Transcriptome
Tel: 05 57 12 27 75
Equipe Genetique
69 route d'Arcachon
33612 CESTAS

Le 29/10/2012 15:59, Jeremy Goecks a écrit :
Whatever the file type I set for the gff file (gff3, gff or gtf), I get the transcript_id error:

Traceback (most recent call last):
File "/home/pgtgal/galaxy-dist/lib/galaxy/datatypes/converters/interval_to_fli.py", line 91, in
File "/home/pgtgal/galaxy-dist/lib/galaxy/datatypes/converters/interval_to_fli.py", line 30, in main
   for feature in read_unordered_gtf( open( in_fname, 'r' ) ):
File "/home/pgtgal/galaxy-dist/lib/galaxy/datatypes/util/gff_util.py", line 375, in read_unordered_gtf
   transcript_id = line_attrs[ 'transcript_id' ]
KeyError: 'transcript_id'

This was due to an incomplete feature. Turns out that GFF support hadn't been included in feature search; I've added it in -central changeset fa045aad74e9:


Is the gff file not correct?

I believe there is an issue with your GFF: it is using non-standard identifiers in the attributes (last) column. To the best of my knowledge, 'name' is not a valid field for connecting features in GFF3 (which is my best guess for the file version), but your GFF uses this field anyways.

To fix this issue, I replaced 'name' with 'ID' (which is compliant GFF3) from the command line:

% sed s/name/ID/ ~/Downloads/test.gff > ~/Downloads/test_with_ids.gff

and this fixed the issue.

Finally, there is a sed wrapper in the toolshed should you want to do this conversion in Galaxy:



Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:


Reply via email to