Data manager for GTF/GFF files

[frederikcoppens]

There should be a data manager for genome annotation files to populate the .loc file (e.g. to use for featurecounts).

[frederikcoppens]

@bgruening IUC interest?

[mvdbeek]

There is some work ongoing/stalled here.
Don't be mislead by the tophat2 title, this is actually universally applicable.

[bgruening]

@frederikcoppens featurecounts is not using loc files, isn't it?
I'm not so keen on supporting data managers for gtf annotations. These data change to often imho and it's a good thing that people know about ensembl.org or ucsc.
I would like to see a data-source implementation for esembl so that we can get data like from UCSC. But I failed to get in contact with the EBI folks.

[frederikcoppens]

I did not check, I should correct myself, I thought the data tables wrote to the loc files but that's not the case, right?
I think it's essential to have a gtf annotation manager. If I add a new genome, I want to immediately and easily add the annotation too. I would like to make this as easy as building a new index. I rarely update the annotation (maybe particular to the plant field? or am I just lazy) after that.
Also for e.g. trackster this could facilitate showing gene annotations.

As plants are not in ucsc, this adds to my problem in Galaxy. So a link with ensembl (all the "flavours" such as gramene) would be quite useful. I guess this was a subtle way to see if we can leverage some ELIXIR contacts?

[bwlang]

Björn

I'm not so keen on supporting data managers for gtf annotations. These data change to often imho and it's a good thing that people know about ensembl.org or ucsc.

I'm not sure I agree... in the spirit of reproducibility we often standardize on a specific GFF for the duration of a project, one fairly frequent (and worse, occult) error is that users sometimes choose the wrong gff from the shared data.