Assign MONDO ontology terms to diagnosis groups

[allyhawkins]

Looking at these diagnosis groups, I was wondering if we wanted to use MONDO terms here too (maybe not for display, but we could include them in the table)? My general thought is that if we can use ontology-based groups, we can more easily justify them.

We could use:

• MONDO:0021248: nervous system neoplasm for Brain and CNS
• MONDO:0002334: hematopoietic and lymphoid system neoplasm for leukemia
• MONDO:0044334: connective and soft tissue neoplasm for sarcomas
  And then leave an "other" category.

This would change a few categories, but I am had questions about some of them them anyway:

• is there a reason you don't have neuroblastoma and retinoblastoma in the Brain and CNS category?
• in the ontology, rhabdoid tumor falls under connective and soft tissue neoplasm; I'm not sure if we would want to maintain that...

This makes the categories more unbalanced, which could create a different problem... we might have to revert to making separate bar plots (though I would try using patchwork to join them, as I think you can get easier control over placement).

All of this I might say to push off to a separate PR. The current version strikes me as good, so the real question is whether we want to use Ontology-based groupings, and then we might have consequences that flow from that.

Originally posted by @jashapiro in #13 (comment)

We should make the following decisions:

• If we want to use the MONDO terms and provide the diagnosis group table as a supplemental figure? Note that if we do this, we will use Sarcoma instead of the connective tissue group.
• I already moved retinoblastoma, but we should decide if we want to move neuroblastoma.