Ontology meeting 2016-05-05

From GO Wiki
Jump to navigation Jump to search

Attendees:DavidH, Paola, Melanie, Heiko, Tanya, Val (special guest), Harold, DavidOS, PaulT

Minutes:DavidH


Val's requests - discuss term specificity

https://github.com/geneontology/go-ontology/issues/12416

The basic issue with this is that when curators use different specificities of the terms like the protein kinases, they are displayed as if they have different activities. But isn't it informative to know if something is a serine/threonine kinase? Pombase would capture this in the annotation. How far are we willing to go to represent mechanism? Does it matter to know if it modifying a tyrosine or a serine or a threonine? The most important thing is the substrate. There is some disagreement about whether this is useful information. Not sure how much discussion this is worth. People refer to it in the literature and it is not unreasonable to represent. From the GO point of view, we don't really care about the amino acid residue. Is this just a PomBase issue? Don't we annotate with respect to the information that is represented in a paper? Isn't that why we want these? Val will think of a way to do it display-wise. We can close this one. We will not add residue-specific terms, but will leave the rest as is.


https://github.com/geneontology/go-ontology/issues/12417

These terms are poster children for why LEGO will be a clearer representation. These ubiquitin ligases have hundreds of substrates. Do we want to go the route of enumerating all of the things that this regulates? We should just start doing this with LEGO. Many of these terms when broken down into models will be much clearer. We need general grouping terms however. Lots of discussion. In process, we want to represent biological modules. We don't need to invent generalized grouping classes. Lots of discussion. The primary term name should be the term a biologist would use. Anything else should become a synonym. Keep the APC term and use the cyclin as the input. Get rid of the first term. Instead we will merge the terms.

https://github.com/geneontology/go-ontology/issues/12418

https://github.com/geneontology/go-ontology/issues/12419

https://github.com/geneontology/go-ontology/issues/12420

https://github.com/geneontology/go-ontology/issues/12427