Interested in submitting a GO annotation for non-model organism

[manyasi06]

Greetings,

I am emailing before I begin to setup GO-annotation reference set for Channel Catfish. We additionally plan to update the data as well after the initial submission.

Regards

Bryan Musungu

[vanaukenk]

Hi Bryan,

Thanks very much for contacting GO about your submission.

Can you tell us a bit more about what types of annotations you would like to submit (e.g. InterPro2GO-based annotations, literature-based annotations) and what primary identifiers you'll be using to associate the genes/gene products with GO terms? That will help us get started.

Best,
--Kimberly Van Auken

[manyasi06]

Hello Kimberly,

It will be doing the Interpro2go based annotation and I will be using NCBI accession as the identifiers if that is okay. If not I am not opposed to using different identifiers.

Regards,

Bryan

[selewis]

We'd prefer the UniProtKB identifiers because that is what we use for the reference proteomes of each organism ("model" or not). Hope that's possible, but since you're using interpro2go, seems even easier.

…

-S

On Wed, Oct 25, 2017 at 12:43 PM, manyasi06 ***@***.***> wrote: Hello Kimberly, It will be doing the Interpro2go based annotation and I will be using NCBI accession as the identifiers if that is okay. If not I am not opposed to using different identifiers. Regards, Bryan — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1670 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABcuECe--advn40YJNzr7H-pXtVBQd7gks5sv49cgaJpZM4QGL4H> .

[vanaukenk]

Hi Bryan,

Thanks again for your interest in GO annotations for channel catfish.

As Suzi mentioned, for non-model organisms, we typically work with UniProt as they have an established InterPro2GO pipeline, as well as other automated annotation pipelines, for a large number of proteomes.

Also, from querying the UniProt database for 'channel catfish' it seems that there are already some manual (i.e. literature-derived) GO annotations from curation groups such as AgBase. Is your group also interested in making manual GO annotations?

Have you had a chance to compare your sequences to what already exists in UniProt, and if so, how complete is the current coverage? Once we have a better sense of how your reference sequences compare to what is currently in UniProt we can come up with a more concrete plan on moving forward.

Best,
--Kimberly

[manyasi06]

Hello Kimberly,

The agbase currently only has about 6221/47978. I am not too particularly sure how recent the update has been because the genome was updated last year. Another question I had is the current genome assembly does not appear to link with uniprot because I am unable to look ids from NCBI or is it my side.

Best,

Bryan

[vanaukenk]

Thanks, Bryan.

I think the best thing to do here would be to put you in direct contact with the UniProt group so that we can work together with them to get the channel catfish proteome into the UniProt databases and, ultimately, into the automated GO annotation pipelines.

I will send an email and cc you.

[pgaudet]

@vanaukenk OK to close ?

[ValWood]

It really bothers me that we no longer have an easily accessible helpdesk. In fact, it seems inexcusable that the GO does not have a helpdesk.

I'm sure that our funders and advisors will look very unfavourably on this. Yes, the number of queries has dropped (probably by 99%, I am guessing, but I did see all the help-desk queries so I think this is a ball park underestimate).

I do not think that this
amigo -> data health link->
http://help.geneontology.org/
Can possibly work if users need to register on Github to report a problem. My concern is certainly justified because nobody has yet reported a problem via this route (is that correct?).

Think about it....... We cannot even easily get the UniProt curators who use GO to get a GitHub ID, so why should we expect a random GO user or contributor to do this GO ?

Our priorities need to be based on what users say and need. It might seem to be a pain, but if we listen to what our users need, rather than what we think they might need, we will be better.

We exist to support biologists. We cannot possibly justify not enabling users to contact us easily with questions , problems or suggestions.

Think about how frustrating this is for users.

Rant over.

[ValWood]

@RLovering did a massive amount of work to get users to submit information, and this has effectively been stymied by the decision to remove the GO helpdesk.

To evaluate, please go to AmiGO and follow the instructions from the "GO health link",
http://help.geneontology.org/

or better still, sit with somebody whois not connected to GO and ask them tell you what they would do if they got to this link. It is actually quite pointless.....