Improve support for synonymies when resolving specifiers against the Open Tree Taxonomy

[gaurav]

We currently use only the scientific name to tie specifiers to nodes on the phylogeny.

Steps to reproduce:

1. Create a clade definition that "includes Chelonia mydas and Chelus fimbriata".
2. Resolve these specifiers on the Open Tree Taxonomy:

• Chelonia mydas resolves to ott 559133 as Chelonia mydas.
• Chelus fimbriata resolves to ott 235773 as Chelus fimbriatus.

3. Reason over this clade definition. This definition does not resolve as expected, likely because the definition in JSON-LD points to Chelus fimbriata, not Chelus fimbriatus.

Possible fixes:

1. A deeper fix in the Curation Tool would fix this as well (Add support for recording cited relationships between taxonomic units klados#129). In this case, we know that C. fimbriatus in this case is a synonym of C. fimbriata, so we could add this as an annotation.
2. Instead of using the scientific name, we could use the Open Tree Taxonomy ID as an external identifier for both the specifier and the node. This would guarantee that every OTR definition resolves correctly without any side-effects because of synonymy.

[hlapp]

• In this case, we know that C. fimbriatus is a synonym of C. fimbriata, so we could add this as an annotation.

So doesn't that mean that the resolution to ott 235773 is correct? It sounded like it should have matched something else given better annotation.

• Instead of using the scientific name, we could use the Open Tree Taxonomy ID as an external identifier for both the specifier and the node. This would guarantee that every OTR definition resolves correctly without any side-effects because of synonymy.

Yes, but are OTT IDs persistent?