You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For better or worse, subsets have been used for a variety of use cases (from defining upper levels in ontologies, to creating domain specific subsets, to external stakeholder usage. Mondo for example has a rich variety of subsets, including: rare, nord_rare, otar, clingen, disease_grouping, subtype etc. Uberon makes extensive use of subsets as well, relevant to groups such as hubmap or hca. I would like to ask, without really being 100% sure that I thought this through, to make it possible to index the subset remainders and take them into account during search.
Use cases
The subsets provide a load of search signal. For example, the keyword "rare" will make it possible to discover a rare disease even if the fact is not recorded anywhere else in the text. Similar, if you look for "nord", you will find diseases which related to the NORD organisation (which was requested by the NORD organisation for example).
So I would like, at least weakly, that a user typing nord would be able find terms in the nord_rare subset. Right now the index only returns classes that have a NORD xref when you search:
Describe alternatives you've considered
None.
Additional context
My biggest worry with this request is that OLS currently does not have a way to display how the hit was established. In particular, the autocomplete field does not show the field where the hit was done, like say Protege does:
This may confuse some people.
The text was updated successfully, but these errors were encountered:
New feature description.
For better or worse, subsets have been used for a variety of use cases (from defining upper levels in ontologies, to creating domain specific subsets, to external stakeholder usage. Mondo for example has a rich variety of subsets, including:
rare
,nord_rare
,otar
,clingen
,disease_grouping
,subtype
etc. Uberon makes extensive use of subsets as well, relevant to groups such as hubmap or hca. I would like to ask, without really being 100% sure that I thought this through, to make it possible to index the subset remainders and take them into account during search.Use cases
The subsets provide a load of search signal. For example, the keyword "rare" will make it possible to discover a rare disease even if the fact is not recorded anywhere else in the text. Similar, if you look for "nord", you will find diseases which related to the NORD organisation (which was requested by the NORD organisation for example).
User communities
Describe the solution you'd like
Subsets are formatted like this:
So I would like, at least weakly, that a user typing
nord
would be able find terms in thenord_rare
subset. Right now the index only returns classes that have a NORD xref when you search:Describe alternatives you've considered
None.
Additional context
My biggest worry with this request is that OLS currently does not have a way to display how the hit was established. In particular, the autocomplete field does not show the field where the hit was done, like say Protege does:
This may confuse some people.
The text was updated successfully, but these errors were encountered: