GMOD / Apollo

Genome annotation editor with a Java Server backend and a Javascript client that runs in a web browser as a JBrowse plugin.
http://genomearchitect.readthedocs.io/
Other
128 stars 85 forks source link

Improving capture of gene ontology (GO) data in Information Editor #1134

Closed monicacecilia closed 5 years ago

monicacecilia commented 8 years ago

Discussing the ability to capture ontology terms as well as evidence codes and supporting literature. It includes Gene Ontology as well as various anatomy and phenotype ontologies (e.g. Uberon, OMP, mouse anatomy, etc.).

Gene ID --> ontology term (e.g. GO ID) --> evidence code (e.g. ECO ID) --> PMID

Bringing @satta @cmungall @kltm to this discussion.

nathandunn commented 8 years ago

Should look something like this.

apollo-mockup-go-capture_2016-06-16 (edited)

monicacecilia commented 8 years ago

To start with.. and we may add other fields.

monicacecilia commented 8 years ago

Collection of GH issues related to this improvement:

selewis commented 8 years ago

On export the annotation ("GAF") file should include the user ID as well as the source.

-S

On Fri, Jun 17, 2016 at 12:50 AM, Monica Munoz-Torres < notifications@github.com> wrote:

Collection of GH issues related to this improvement:

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/GMOD/Apollo/issues/1134#issuecomment-226624248, or mute the thread https://github.com/notifications/unsubscribe/ABcuEIWp6FKqzQaYmaX9EeEDhckRKYRcks5qMcUPgaJpZM4I3kr6 .

nathandunn commented 8 years ago

Okay @selewis I added it to the relevant issue.

obsidian83 commented 6 years ago

Any update to this?

hexylena commented 6 years ago

Unrelated to obsidian83's query, since I'd never seen this issue before, wanted to share that we had a nice UI for this, but we built it as a standalone service that was iframe'd in to apollo. https://github.com/TAMU-CPT/cacao-frontend/

nathandunn commented 6 years ago

@obsidian83 Unfortunately not. Do you want to describe your use-case and I can make some recommendations or you just want to provide an actual GO annotation.

If you can use what @erasche and @MoffMade built in the interim that would be great.

@erasche How does this export GO annotations with the rest of the GFF3?

hexylena commented 6 years ago

@nathandunn it writes out a GAF file since we needed to do more than just adding GO xrefs into our gff3 would allow.

obsidian83 commented 6 years ago

@nathandunn the use-case is to be able to use one piece of software to capture all an annotator/curators needs to provide good functional and structural annotation while providing the evidence for them.
@erasche it looks like a nice piece of software, I will try it out.

hexylena commented 6 years ago

@obsidian83 let me know if you have any questions/problems. It proably isn't super easy to set up and it probably hard codes a lot of assumptions from our old workflow which was built around supporting the gonuts program. But maybe you can make it work for you.

nathandunn commented 6 years ago

@erasche / @MoffMade Does it use a separate backend? Is there a rough install guide for how to integrate it or a high-level (this is what it is)? Would this be appropriate to potentially integrate?

I think the GAF output is fine, but I agree that the current GO annotations implementation (essentially a key/value) was pretty dated even 10 years ago and would love to see something like this integrated.

cmungall commented 6 years ago

Expressivity: GAF < GPAD < GO-CAM

(with a large difference between GPAD and GO-CAM)

It wouldn't make sense to reinvent a GO-CAM editor but you could federate with Noctua

For a simple bag or terms editor it would be better to target GPAD, capture full provenance of annotation

nathandunn commented 6 years ago

@cmungall I like the idea of federating with an existing editor, but will this change the lack of richness in terms of how we represent evidence (ECO code, etc.)?

nathandunn commented 6 years ago

I think there was some confusion as we don't allow a ready way to provide evidence. @obsidian83
My advice would be:

1 - use https://github.com/TAMU-CPT/cacao-frontend/ or some version of it 2 - provide comments that indicate the GO annotation evidence that get mapped back together when reprocessing the GFF3 3 - export annotations to Chado or GFF3 and use another tool to manually provide the functional annotations. I'm unsure if something in Galaxy (https://github.com/galaxy-genome-annotation) or if you could use Artemis (which does these annotations correctly) would be useful.

To be clear, would love to implement this (and should have been done awhile ago), but we don't have the funding to do it in the way we would like. If its something you'd be interested in developing I'd be glad to help.

hexylena commented 6 years ago

@nathandunn

@erasche / @MoffMade

@elenimijalis did the vast majority of the work on CACAO, not @moffmade (Hence the edit to your comment.) That said, @moffmade is the current maintainer since he's working for the CPT.

Does it use a separate backend?

Yep, https://github.com/tamu-cpt/cacao-backend

Is there a rough install guide for how to integrate it or a high-level (this is what it is)? Would this be appropriate to potentially integrate?

Not really. There are containers and some instructions for various sub-components, but no overall guide since it was primarily focused on the CPT's use case.

@cmungall GO-CAM

We considered integration with noctua somehow but could not find a good way to do this in the context of: 1) apollo, 2) the gonuts/cacao programme we were trying to integrate with. We also considered GPAD but the existing pipelines that were being used accepted GAF only. Definitely good plans for future work.

nathandunn commented 5 years ago

image