OBI includes a minimal extract of uberon with classes like cerebellum, brain. However, these show up as a flat list, and this loses crucial parthood information between these.
If we look at a visualization in OAK using all relationship types:
obi viz cerebellum brain
In uberon it's a bit more complex:
while it's reasonable to leave out intermediate classes, the extract should still retain structural relationships, such as the cerebellum being part of the brain
Including more complete exports will help detect issues in OBI like this one:
1595
It will also help prevent being users being confused when they see a flat list of unconnected classes in browsers like OLS (but ultimately improvements in OLS may fix this issue by allowing uberon to see in a non extracted form: https://github.com/EBISPOT/OLS/discussions/595#discussioncomment-2787478)
OBI includes a minimal extract of uberon with classes like cerebellum, brain. However, these show up as a flat list, and this loses crucial parthood information between these.
If we look at a visualization in OAK using all relationship types:
obi viz cerebellum brain
In uberon it's a bit more complex:
while it's reasonable to leave out intermediate classes, the extract should still retain structural relationships, such as the cerebellum being part of the brain
Including more complete exports will help detect issues in OBI like this one:
1595
It will also help prevent being users being confused when they see a flat list of unconnected classes in browsers like OLS (but ultimately improvements in OLS may fix this issue by allowing uberon to see in a non extracted form: https://github.com/EBISPOT/OLS/discussions/595#discussioncomment-2787478)