PR #620 covers adapting the generation of test data from the existing (and out of date) predicates/MKG data. This PR attempts to replace that with a new process that gets the Biolink class hierarchy and relations directly from the database, allows us to regenerate (and hopefully fix MKG issues like #585) in one fell swoop.
PR #620 covers adapting the generation of test data from the existing (and out of date) predicates/MKG data. This PR attempts to replace that with a new process that gets the Biolink class hierarchy and relations directly from the database, allows us to regenerate (and hopefully fix MKG issues like #585) in one fell swoop.
WIP