search

pombase / canto

The PomBase community curation tool
https://curation.pombase.org
Other
19 stars 7 forks source link

cardinality for protein binding phenotype extensions #1124

Open mah11 opened 8 years ago

mah11 commented 8 years ago

In b6b9d534da028509, my protein binding phenotype annotation demanded both two assayed proteins AND two assayed isoforms, because the configuration file has (well, had) cardinality 2 for both entries. It's got a biologically nonsensical hacky placeholder for the moment.

For now, I've changed the configuration so the admin-only isoform extensions aren't required (cardinality 0,2) (and I'll go back to the session and see if it works as I hope).

But in the longer term it would be nice if Canto could be clever enough to require two extensions, which could be either generic protein IDs (pick a gene ID) or isoform-specific PRO IDs. Presumably that would require work on both configuration and code, so I don't expect it soon!

ValWood commented 8 years ago

I think thats better in the short term anyway....If PRO IDs are required we'll need to deal with them in the short/medium/long term

I thought I had opened a ticket about this, clearly not...

mah11 commented 8 years ago

update: admin-only isoform options now, well, optional in config (and temporary hack fixed in the session I mentioned, so it should be at least sort of ok for us)

mah11 commented 8 years ago

This looks like a special case of the general issue in #1068.