Check for allele with same description when entering in Canto

[ValWood]

Since we still get new additions into the chado_checks.duplicate_allele_descriptions log

https://curation.pombase.org/dumps/latest_build/logs/log.2024-06-25-21-01-00.chado_checks.duplicate_allele_descriptions

And, since this is going to cause problems once the permanent allele IDs are implemented, we should make a check in Canto so that we cannot add an existing allele with a new name (prompt, did you want to use the existing allele "allele with the same description".

That way, even if we DO need to change a name, we will be alerted before we add it and you try to load it.

[kimrutherford]

Related:

• 1284

[ValWood]

needed before allele pages

[ValWood]

This will be implemented as a check to make sure that the allele is not existing with a different name already.

Exampes: swi5-39

[kimrutherford]

I've implemented this such that it checks for duplicates as you type. It wasn't any extra effort. The "OK" will be greyed-out if the description is a duplicate.

I've updated the test Canto with this change. Could you do a quick test of the allele dialog when you get a chance? Also, please let me know if the wording of the messages needs improving.

https://curation.pombase.org/test/curs/4666975359de04dd/genotype_manage

[ValWood]

It works, but I think we shoud change the messaging.

In this instance I did a new name, and then a description, and get:

[ValWood]

It probably isn't very obvious what the user needs to do. Perhaps the message should say: This allele is already in PomBase. Use the existing allele and add your name as a synonym?

[kimrutherford]

We need to explain that it's the description that's already in PomBase not necessarily the whole allele.

[kimrutherford]

It could say something like: An allele with this description already exists in PomBase (A bit long though)

[ValWood]

In this scenario, I added the allele description first , which I often do, and got the message before I added a name:

[ValWood]

We need to explain that it's the description that's already in PomBase not necessarily the whole allele.

I don't understand? - Alleles and descriptions are 1:1 so if the description is in PomBase , the allele is too. It seems that to the users just saying the allele already exists makes more sense (although internally we are sure about this because we already have the description)

[kimrutherford]

Use existing allele for T4118TT ( name).

I don't understand what should happen after the message is shown. If there isn't a button how can the user make Canto "Use existing allele"? There needs to be some way of having that happen doesn't there?

Can we talk on Zoom/Skype about this sometime?

[PCarme]

The check does not seem to work when renaming an allele in "Edit genotype details". For example here, I could rename the allele as pcf1-PIP*

although there is already an allele with the same description in Canto/PomBase, called pcf1-PIPmut.

For context, the allele was called pcf1-PIPmut when I started approving the session, I changed it to pcf1-PIP* to fit the naming used in the paper without realising that the author used an existing allele in his curation.

[ValWood]

Message: Add your name as a synonym for existing allele (blah) ? [Add]

[PCarme]

The check does not seem to work when renaming an allele in "Edit genotype details". For example here, I could rename the allele as pcf1-PIP*

For context, the allele was called pcf1-PIPmut when I started approving the session, I changed it to pcf1-PIP* to fit the naming used in the paper without realising that the author used an existing allele in his curation.

After checking, this also happened on the test Canto (see the screenshot where I could bypass the check and create two alleles with the same description).

[kimrutherford]

Hi @PCarme

Thanks for testing. I can't work out how to reproduce the bug you found. I'll get you to show me how to do it on the next call.

[kimrutherford]

Hi @PCarme

I think I've fixed the bug with copy-and-edit. But I'm not sure about the other problem you demonstrated. When you have time could you try the test Canto again? Thanks.

https://curation.pombase.org/test/curs/4666975359de04dd/genotype_manage

[PCarme]

Hi @kimrutherford The bug with the copy-and-edit is fixed. The other bug is still there though. To reproduce it, you have to create the existing allele in the session, then you can change its name while still keeping the same description via the "Edit genotype details" action.

[kimrutherford]

Hi Pascal. I'm having trouble reproducing the name change problem. Could you give me a demo on the next call?

[PCarme]

Sure, no problem !

[kimrutherford]

The cause of this problem might be that we can edit the details of existing alleles when we go back to the Genotype edit page.

[PCarme]

I think it is good to be able to edit existing alleles, but only to change the new synonyms and expression status.

[kimrutherford]

I think it is good to be able to edit existing alleles, but only to change the new synonyms and expression status.

Good point. I'll have a think about how that could work.

[kimrutherford]

The cause of this problem might be that we can edit the details of existing alleles when we go back to the Genotype edit page.

I had forgotten the details about this. The current Canto release has bugs when dealing with existing alleles. These problems are fixed in the new version for the stable allele identifiers project.

• https://github.com/pombase/pombase-chado/issues/986#issuecomment-1963361242

When we switch over to the new allele system the bugs should be fixed.

I'll leave this open until the next Zoom to remind us to have a chat about it.

We (I) need to get on with implementing the new system!