obophenotype / mouse-anatomy-ontology

Ontologies for mouse anatomy and development
Other
5 stars 3 forks source link

cartilage primordium #11

Open cmungall opened 9 years ago

cmungall commented 9 years ago

I noticed a grouping:

   po EMAPA:36065 ! hyoid cartilage primordium *** 
     po EMAPA:17351 ! hyoid pre-cartilage condensation
     po EMAPA:17622 ! hyoid cartilage condensation

this is the only such grouping (other bone primordia seem to be for membrane bones).

is there a general plan to add more of these. It slightly complicates integration with uberon, where we decided on this pattern:

https://github.com/obophenotype/uberon/wiki/Modeling-endochondral-elements-Design-Pattern

also, unless you're using reasoning then there is a large potential for doing a lot of manual work if you go this route...

tfhayamizu commented 9 years ago

The problem presented here arises often with curation of developmental literature; i.e. ambiguous data reporting. In this case, we need to annotate expression data reported as for the "hyoid" at ages where it could be at either pre-cartilage condensation or at cartilage condensation stage. (Expression images often do not enable more in-depth examination of histology even if we were to analyze this for ourselves, which we do not do at GXD anyway.) With the modeling pattern described, we could annotate to the parent class, but would seem to lose the information that this is a precursor structure.

cmungall commented 9 years ago

For uberon, we could introduce an extra level into the grouping, at the cost of ontology inflation. What is practice at other expression databases @ANiknejad, @ybradford? I believe at Bgee they would use the generic structure. I suppose it may be possible to annotate to the generic and then specific a stage range as being inclusive of the desired states but can see this being seen as too indirect.

ANiknejad commented 9 years ago

At Bgee we report a status for both the structure annotation and the stage annotation, with internal tags that allow to see if an annotation is done the best regarding the current status of the available ontology, and if the sampling matches completely or not or is not documented.

ybradford commented 9 years ago

At ZFIN if a structure is not available within the ontology at the stages described we go up the tree to the parent class and make a request to update the staging for the structure and/or get the children classes added or updated.