Ok, the first annotations are up (crudely). The choice of what to start with was fairly arbitrary, though these had the advantage of being very simple. The paper cited is just a survey of Tetragnatha species that turned up first in the alphabetized list of publications. I have about 150 more annotations sitting in a spreadsheet waiting for the infrastructure to settle down. Of course there is also a lot more to do with both the OWL generation (publications currently have nothing but their doi’s or internal ids) and the display (e.g., each of the labels correponds URI that could be linked, though OntoBee pages aren’t the most friendly).
Getting this table to work is the end result of a lengthy SPARQL session on Saturday, mostly spent building up a query that captured the taxon, anatomy, behavior and publication as it crawled up a deeply nested RDF hierarchy. Once I had the query working, it was back to the javascript to update the little utility that transforms the JSON-ized SPARQL query results into a table (which now has more than one column) and then fix the cases where an NCBI id was available (e.g., Habronattus) which just returns the name and the identifier, and failing properly for taxa not supported (e.g., Homo sapiens), as well as non-taxa (e.g., SPARQL injection attempts).
The server is a bit slow, meaning I’m close to hitting the limits of the AWS micro-instance I’ve been serving from for the past 18 months. Probably time to start planning the upgrade (which also may involve switching from Ubuntu to Debian).
For now though, I’ll enjoy what’s there and add more goodies within the limits.
An Arachnolingua Blog 