Open kallia-p opened 3 years ago
Bone density subclasses Edits to add is_about specific bones
heel bone mineral density Looking to assign heel bone mineral density to specific bone i.e. calcaneous aka heel bone. Created new EFO term calcaneus EFO:0010974 (used term from UBERON_0001450) SubClass of hindlimb bone. Without adding any axioms (as for other hindlimb bones) the subclass of anonymous ancestor is showing it as part of some 'anatomical system' 3 times, part of some 'skeletal system' 3 times and part of some 'appendicular system' 3 times is this OK? heel bone mineral density - added subclass of is_about some calcaneus and added alternative term calcaneal bone mineral density.
Hip bone mineral density - added subclass of is_about some hip.
Femoral neck bone mineral density - currently shows as a subclass of hip (subclass of animal body part and very sparsely populated but I could do some edits on hip) and then add subclass of is_about some neck of femur which I need to import from Uberon.
Bone measurement branch
Notes from @dosumis Bone - is reasonably complete, but is_about assertions could be made more detailed (e.g. to specific bones). (Note - the grouping class here is defined using OR. The Hermit reasoner is required for accurate classification under it.)
Initial strategy - focus on is_about axioms
Browse bone measurement branch (relatively small) to examine current classification hierarchy and relations used - iron out inconsistencies between children of this branch e.g. most children measurements in this branch are directly relevant to bone size, density, volume except for alkaline phosphatase measurement where SubClass appears to have been assigned based on biomarker relationship i.e. ‘bone measurement’ ‘liver enzyme measurement’ (also ‘protein measurement’) but not ‘blood (or serum) measurement’ whereas the definition is ‘a quantification of alkaline phosphatase in blood. AP is a biomarker of bone and liver disease.'
Review how related terms are modelled in HPO, MP, Uberon, (also CMO, NCIT) using OLS; GO in Protege: What terms are used. What does the classification hierarchy look like?
Increase the depth and breadth of is_about axiom coverage e.g. 'radius bone mineral density' -> add subClassOf is_about uberon:radius
Add EquivalentClass definitions to measurement grouping terms that currently lack them - then search out potential subclasses by string searching