imported annotations not populating annotations tab

[jaspwn]

Hi,

I am very new to apollo but would like to improve annotations of my favourite gene families, but when I add a gff track from an online database the annotations within are not appearing in the searchable annotations menu. Only the annotations that I have dragged up to the user-created annotations track appear.

My workflow at the moment is import my organism genome, add a gff annotation track, add RNAseq evidence track and then annotate.

Here is the gff track configurations:

{"storeClass":"JBrowse/Store/SeqFeature/GFF3", "urlTemplate":"raw/1718121519122424116.gff", "label":"1718121519122424116", "type":"JBrowse/View/Track/HTMLFeatures", "style":{"className":"feature"}, "key":"1718121519122424116", "topLevelFeatures":"mRNA", "apollo":{"source":"upload", "type":"GFF3", "topType":"mRNA"}}

Any help would be much appreciated.

[hexylena]

They're not annotations yet, just a track of data. If you right click/create annotation (or drag them up to the yellow bar labelled "User Created Annotations") then they'll become proper annotations.

[jaspwn]

Thanks for the feedback, is there a way import the data track as annotations so I do not have to drag each one up? Or alternatively is there a way to search the names in the data track so I can find the gene models that I want to fix more easily?

[hexylena]

I've automated that in the past with scripting, I'm not sure there's a built in way to do that. You should be able to search track values if they've been indexed (sorry not super helpful)

[garrettjstevens]

You can load a GFF3 from the command line using the arrow organisms add_organism command. See here for more details: https://github.com/galaxy-genome-annotation/python-apollo/blob/master/docs/commands/organisms.rst

[jaspwn]

Hi, sorry to get back to this after all this time but I was wondering if you could provide more insight to how you did this via scripting @hexylena ?

Or @garrettjstevens I am not sure how I would use arrow organisms add_organism to load a gff3 file? These are the options I can use for that command.

`
--blatdb TEXT Server-side path to 2bit index of the genome for Blat
--genus TEXT Genus
--species TEXT Species
--public Should the organism be public or not
--metadata TEXT JSON formatted arbitrary metadata
--suppress_output Suppress output of all organisms (true / false) (default
false)

-h, --help Show this message and exit.
`

[garrettjstevens]

@jaspwn I can't remember what I was trying to say in my last comment, but what you actually probably need is arrow annotations load_gff3. You can use arrow organisms add_organism to add the organism first, or you can use an organism you've already created in the UI. You can see some examples of load_gff3 here.

[hexylena]

Yep, @garrettjstevens's got it, you need the load_gff3 subcommand for that.

[jaspwn]

Thanks both. Perhaps this needs to be another thread (or forum search) but as I’m running this as a docker container could you provide advice about how to run that command or what script I might add it to?

Sorry, I’m very much a novice when it comes to this.

[hexylena]

Sure @jaspwn you can just do, no worries, good questions!

arrow init

to setup the connection, it can be run from outside of the container. You'll need your apollo address where it's accessible and username/password. After that you can invoke the arrow organism load_gff3 <path-to-gff3> command which will use your connection details to talk to your apollo server.

[jaspwn]

Ah amazing! Works exactly how I need it! Thanks @hexylena & @garrettjstevens for the step through!

For future novices I had to install python-apollo

pip install apollo

Have my apollo docker container running, then run

arrow init

One trick was getting the url exactly right - mine was run locally so I had to input

http://localhost:8888

Then finally

arrow annotation load_gff3 ORGANISM /path/to/file.gff

And now it my annotations appear in the User-created Annotations track and their names are now searchable in the Annotations tab!

[hexylena]

Fantastic, glad you got it working @jaspwn !