NIH | National Cancer Institute | NCI Wiki  



Safran, Tracy

Wynne, Robert    NIH/NCI [C]x
Ong, Kim L ISx
Lucas, Jason R ISx
Bauer, Scott  Mayox
Stancl, Craig
Endle,  CoryMayox



 Update on Value Set Loading (Tracy and Rob)

  • Use of Updated mass loader
  • Use of OWL2 source
  • Cross Value set Search failure


Discussion Points:
  • Use of updated mass loader - Update
    • Scott's latest load had 4 issues (VS that are published, but there is no source to run against).
      • This was on OWL2 source.
  • Loading discussion
    • Load all definitions based on a file. Then resolve.
      • If a vs doesn't resolve, we could remove them.
    • How does the browser know what to display?
      • Browser returns all value set definitions
    • How do we validate if all the value set definitions were created and resolved?
      • We could provide queries to the DB to determine if they were. 
        • This could create a report of URIs  (and description) that didn't load.
        • Rob doesn't want to match the URI with an entity name.
      • We could automate it by creating a folder of the failed value set definition.
    • Rob/Tracy would like a list of file names that fail to resolve
      • Scott suggested that we could return a list of URIs  and descriptions.
      • This could be located in the LexEVS Load Log.
    • We are suggesting that Rob/Tracy use the script that Scott created to load multiple definitions and resolutions
      • Rob/Tracy are using their single scripts on Data QA - LexEVS
      • Rob/Tracy are using their single scripts on DEV - LexEVS
      • They haven't tested with OWL1 source


Decision Points:

  • Scott will test OWL1 source loading value set definitions and resolved value sets on DEV.
    • DEV has 17.01e (marked PRODUCTION)
    • Scott will run some tests to verify.
    • Scott will need to wait until Kim is finished testing.
  • Scott to create a JIRA issue to track how to verify the value sets definitions/resolutions.

Future Architecture approach

  • Determine next steps
Discussion Points:

Decision Points:

  • Craig to set up a weekly discussion meeting for this.
  • No labels