#431 subsets: Make some subsets visible for browsing, but don't allow their selection for annotation

[Valerie Wood]

Some grouping terms are defined in GO as a grouping_term should not to be used for annotation (for example nuclear part?)

We will also use the same subsets for phenotype terms where the terms should not be used for direct annotation. In these cases the "proceed" button should not be visible and a message should say

"This term should not be used for direct annotations, please select a child term"

[Midori Harris]

Message edited, mainly to correct punctuation (but also to make the notion of "direct annotation" a little clearer):

"Genes should not be annotated directly to this term. Please select a child term ."

[Midori Harris]

GO and FYPO now both use the same subset for this:

subsetdef: high_level_annotation_qc "High-level terms not to be used for direct annotation"

[Valerie Wood]

This will be useful, but not necessary for general CC release so lowering priority

[Kim Rutherford]

If a term is marked a not for direct annotation, does that mean all it's parents shouldn't be used for direct annotation either?

[Kim Rutherford]

See also: [curation-tool:#430]

[Midori Harris]

Good question - I'd never properly thought about it before.

My gut feeling is that you'd want to disallow annotations to is_a ancestors, but for other relationship types you can't make a blanket statement.

For example, Antonia pointed out that "cytoplasmic part" is do-not-annotate, but its part_of parent "cytoplasm" is fine (as it should be).

[Kim Rutherford]

My gut feeling is that you'd want to disallow annotations to is_a ancestors, but for other relationship types you can't make a blanket statement.

So, in the general case, to find out if a term shouldn't be used for annotation, we need to see if any of it's ISA descendants shouldn't be used for annotation. That sound like a job for Oort/OWLTools. Do you know if Oort can do such things? Or is it another question for the OWLTools folk?

[Midori Harris]

We chatted about this on skype - owltools should be able to do the relevant query (get all terms in the subset, then get their is_a ancestors), although I don't know the exact syntax.

I'm also not sure whether Oort can add ancestors to a subset by magic, so if it becomes important, ask Chris/Heiko.

[Midori Harris]

update: the subset IDs have changed

in FYPO:
subsetdef: qc_do_not_annotate "Term not to be used for direct annotation"
subsetdef: qc_do_not_manually_annotate "Term not to be used for direct manual annotation"

in GO:
subsetdef: gocheck_do_not_annotate "Term not to be used for direct annotation"
subsetdef: gocheck_do_not_manually_annotate "Term not to be used for direct manual annotation"

[Valerie Wood]

also
conditions defined the subset grouping_terms for terms should not be visible in the tool

[Midori Harris]

[curation-tool:#428] is also related (tho it appears to include an unrelated comment)