allysonlister
commented
4 years ago Are @delphinedauga and @Drosophilic in agreement? :-)
Closed allysonlister closed 4 years ago
allysonlister
commented
4 years ago Are @delphinedauga and @Drosophilic in agreement? :-)
allysonlister
commented
4 years ago Updates for this ticket have been committed. Diff between these changes and previous version of DRAO-merged.owl is below.
Please note that for some reason the changes for the Classification ticket (#51) to DRAO-ontofox-annotation.txt were NOT propagated properly to my checkout of the master branch, and it was only the use of this diff that prevented the Classifications work from being accidentally rolled back. I'm sure it was some kind of user error on my part, but it shows how important running the diff is to each change to our ontology.
http://www.fairsharing.org/ontology/domain/DRAO.owlfile:/home/.../domain-ontology/development/DRAO-merged.owlhttp://www.fairsharing.org/ontology/domain/DRAO.owlfile:/home/.../domain-ontology/development/build/DRAO-merged.owlhttp://purl.obolibrary.org/obo/SO_0001060SO_0001060 替代术语 "VAAST:sequence_variant"
SO_0001060 FAIRsharing alternative term "Sequence variant"@en
SO_0001060 definition "A sequence_variant is a non exact copy of a sequence_feature or genome exhibiting one or more sequence_alteration."
SO_0001060 替代术语 "Jannovar:sequence_variant"
SO_0001060 替代术语 "sequence variant"
SO_0001060 inSubset "FAIRsharing"@en
SO_0001060 label "sequence_variant"
Class: SO_0001060
SO_0001060 SubClassOf sequence_feature
http://purl.obolibrary.org/obo/MI_1241variant definition "A natural change in a sequence or structure in comparison to a reference entity."
variant FAIRsharing alternative term "Sequence variant"@en
Class: variant
variant SubClassOf sequence_alteration
What are you changing?
Remove MI variant and replace it with the SO _sequencevariant.
The current hierarchy is as follows:
Why are you suggesting this change?
We make heavy use of SO for sequence-related terms, and variant is one of the few that come from elsewhere. The SO _sequencevariant hierarchy is quite rich and it makes sense to align with it. Also, the label variant could imply a more generic meaning than is actually intended. Finally, it makes sense to do this work as part of a general tidy of the sequence hierarchy (to improve searching in FAIRsharing), and we are working on #60 at the same time.
Mapping
Label
Recommendation: Sequence variant
Reasoning: Provides added context (compared to variant), and is the same as SO's current label, but with FAIRsharing style applied.
IRI
Recommendation: http://purl.obolibrary.org/obo/SO_0001060
Reasoning: Because we already use SO for most of our sequence-related terms.
Definition
Happy with the SO definition, even if it is a little formal.
Hierarchy
Recommendation: Child of _sequencefeature / sibling of region/sequence
Reasoning: To solve the issue of where to place _sequencevariant, I looked at their papers (1 and 2) but nothing I could find provided a suggestion of where in BFO this term could go. PATO has a variant class as a child of quality, but its definition doesn’t match what a sequence variant is, and so I rejected PATO. I know need to move _sequencevariant out of the _sequencealteration hierarchy_ (where MI _variant is currently residing) due to its definition, so I propose it as a sibling class of region (child of _sequencefeature), which is SO’s sequence class (this doesn’t exactly match SO’s native hierarchy, but it does keep everything grouped nicely together for searching FAIRsharing, and still fits with the definition provided by SO).