immune cell type hierarchy is fragmented due to import of BTO cell type terms

[dosumis]

e.g.

Action:

1. review whether BTO terms are necessary for scAtlas
2. If BTO is necessary, remove mappings to BTO cell types from Zooma & investigate adding a test to the build pipline to prevent this recurring (should be possible with Sparql)

[dosumis]

• @dosumis to provide list of BTO terms that are duplicate with CL

• @Anja - scAtlas curators to fix manually and remove form the Zooma mapping file.

• Cell line terms from BTO are also used. These need to be discussed separately as there are many used. Choice of ontology needs to be co-ordindated with HCA/DCP.