search

geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
223 stars 40 forks source link

Taxon issues with neuroblast differentiation #27267

Open hattrill opened 8 months ago

hattrill commented 8 months ago

Taxon violations for fly annotations for GO:0014016 neuroblast differentiation and descendants: GO:0014019 neuroblast development GO:0014018 neuroblast fate specification GO:0014017 neuroblast fate commitment GO:0007400 neuroblast fate determination

coming from the relation with neuroblast (sensu Vertebrata) (CL:0000031)

From FBbt:00005146 neuroblast maps to CL:0000338 neuroblast (sensu Nematoda and Protostomia)

Is there a way that we can resolve this or do we have to make a parallel GO branch for neuroblast differentiation sensu Nematoda and Protostomia?

cmungall commented 8 months ago

We could add a generic neuroblast in CL that groups the two but I expect this was not done for a reason. For example, fly neuroblasts are stem cells, whereas vertebrate ones are not (although they are obviously not terminally differentiated).

The advantages of having parallel branches in GO is that fly biologists asking GO for genes involved in stem cell development will (correctly) get genes involved in neuroblast development.

If we don't do parallel branches, then either fly biologists will get incomplete results (or vertebrate biologists will get incorrect results, if we falsely state that its always a stem cell)

If we do make parallel then we should obsolete the groupings or place them into do-not-annotate. It should be easy for groups to automatically migrate, we would provide two mapping tables, one for fly, the other for vertebrates.

I confess I'm not sure sure where C elegans belongs in all this.

pgaudet commented 8 months ago

@raymond91105 , can you please give some insights about C elegans neuroblasts?

hattrill commented 8 months ago

Thanks for the explanation @cmungall, that makes sense. Not as many manual annotations to this term and its descendants - 127 excluding IBAs, 65 of which are D.mel (12 to c.elegans). Happy to go with the two branch solution you suggest.

raymond91125 commented 8 months ago

Sorry, I've always wondered if there is a necessary and sufficient definition of neuroblast. In the worm, other than the germline, there is no self-renewing stem cell. Although some cells, say a P1.a, can divide multiple times to produce only neurons. And its sister cell, P1.p becomes part of the hypodermis. So are P1.a and P1 both neuroblasts? Or perhaps P1's mother cell is also a neuroblast? I wonder if the difference between vertebrate and fly neuroblasts is simply an artifact of at which point of lineage they are defined. Aren't vertebrate neuroblasts produced by stem cells like https://en.wikipedia.org/wiki/Radial_glial_cell?

cmungall commented 8 months ago

Aren't vertebrate neuroblasts produced by stem cells like https://en.wikipedia.org/wiki/Radial_glial_cell?

Hmm, it seems the term neuroblast is a bit of a terminological minefield, see https://github.com/obophenotype/cell-ontology/issues/2312

hattrill commented 8 months ago

If no one wants 'neuroblast', we'll have it ;-) But in all seriousness, looks like an even stronger case for having a distinct branch, perhaps limited to arthrpods for neuroblasts that are authentic neural stem cells.