NIH | National Cancer Institute | NCI Wiki  

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

NameRolePresent
Wright, Larry NIH/NCI   x
Fragoso, Gilberto NIH/NCI     x

De Coronado, Sherri    

NIH/NCI    x

Safran, Tracy

NIH/NCI [C] x
Ong, Kim L
IS x
Lucas, Jason R
IS x
Bauer, Scott  Mayo x
Stancl, Craig
Mayo 
Endle,  CoryMayo x
Wynne, Robert    NIH/NCI [C] x
Tran, Tin    NIH/NCI [C]  x

Kuntipuram, Kumar

NIH/NCI [C]x
Haber, MargaretNIH/NCI
x 

Action Items

AssignedDescriptionDate IdentifiedDue DateDate CompletedStatus
CoryAdd link to Maven Opts for end users2016.12.21  

Open

...

Sprint Status

 

Current Sprint  Sprint 50 51 (February 16March 2, 2017 – March 115, 2017)

16X237 Agile Development - Sprint Status#16X237AgileDevelopment-SprintStatus-Sprint50Sprint51

Triple store discussion

  • Gilberto

Discussion Points:

  • Gilberto gave an overview of the current status of the triple stores.
  • Gilberto is waiting for quotes to come back.
  • Stardog has another version coming out that may improve performance.

Decision Points:

  • The team decided to go with Stardog if we can get several developer licenses for a discount.  (They charge $8600/developer license).  Otherwise we will go with Virtuoso.

DEV Server Configurations

  • DEV Blade
    • 6.4.1.3.FINAL
    Mayo DEV VM
    • 6.5.0 SNAPSHOTSeparate DB

Discussion Points:

  • LexEVS Jacob anticipates the DEV blade VM will be installed with the 6.4.1.3 .FINAL is loaded on Mayo DEV VM
  • LexEVS 6.5.0 SNAPSHOT will be loaded on DEV Blade with CentOS 7.
    • Jacob has a ticket for this work.
  • FINAL by early next week.
     
  • Cory created a ticket for Jacob to create separate We will want a separate DB ( MySQL 5.6 ) DBs for the Blade DEV blade and the VM Dev instances.

Decision Points:

Mayo team will request separate DBs for the Blade and the VM Dev instances.Send note to Jacob.

Value Set Loading Enhancements

  • Load Update

Discussion Points:

  • Scott and team have identified the current issues with the value set failures.
  • Tracy has not had an opportunity to test the mass load script yet.
  • Mayo team will meet with Jason and Kim to discuss value sets architecture tomorrow.

Decision Points:

  • LexEVS Meeting Minutes - Value Set Architecture Planning Session - 2017.03.02
  • We discussed how we could automate testing of loaded value set definitions and resolved value sets (in the meeting link above).
  • Tracy suggested starting from scratch to fix this error-prone approach.
  • Rob asked if we could we use the content of Thesaurus for populating the value set?
  • We could look into using an OWL based approach
    • Concept in subset.
  • Tracy suggested that this could be a prototype for outside groups to use too.
  • We will continue this discussion at our value set meeting next Tuesday, 3/14.

Decision Points:

  • Scott to test the value set load on the NCI DEV system.
  • Mayo dev team will look into a new architecture/design for generating the value sets.
CTS2 REST API Discussion

Discussion Points:

  • The Mayo dev team will look looked at the new 6 CTS2 JIRA issues created by Tracy.
  • We will determine if our current implementation can accommodate the request.
  • If it can't, can we implement it in CTS2.
  • If that can't be done, we can look at implementing it in a different service.
  • Larry suggested that there should be a simpler REST interface than CTS2.
    • Larry mentioned that they (customers) are coming up with some criteria and examples of what they would like. 
    • They will have some examples/criteria in the next couple of weeks.
    • This may be something like convenience methods or CTS2 extensions.

Decision Points:

  • The dev team will hold off looking at the existing JIRA issues and see if CTS2 can accommodate this in the next sprint.
  • The dev team will work on the tree extensions instead in this sprint.

LexEVS Logical Role Relationship

  • Discussion with Kim
  • issues and determined what was possible for each.

  • LEXEVSCTS2-138: This fits within the realm of the CTS2 spec and could be implemented as a part of that specification
  • LEXEVSCTS2-139, 140: These are LexEVS coding scheme metadata and don’t really have a good counterpart or fit in CTS2.  We’d likely want to provide a separate resource or service for these. 
  • LEXEVSCTS2-142: Return only the specified information on an entity
    • This would be a new call that returned a new partial resource
  • LEXEVSCTS2-143: Return a property type for an entity 
  • LEXEVSCTS2-144:  Retrieve a concept based on the presence of a restriction
    • Workflow - series of calls to get from CTS2
    • A new wrapper/extension REST call would need to be created to do this in one call.
  • Priorities (suggested by Tracy)
    • Ability to retrieve the FULL_SYN information (Larry)
    • Search on property for a given criteria
      • Give me all the concepts that have a specific property
    • Limit the result set coming back.
     

Decision Points:

Discussion Points:

  •  
    Jira
    serverNCI Tracker
    columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
    serverId7954a81f-12da-3366-a0ef-97c806660e7c
    keyLEXEVS-2764
  • We will leave this issue open until Kim can verify.
Decision Points:
  • Set up meeting to look at examples and discuss.
  • Start to add to sprints - convenience methods for
    • LEXEVSCTS2-142, LEXEVSCTS2-143

AnyURI Designation

Discussion Points:

  • There is an example in the OBI OWL2 source.
  • The AnyURI is declared as data type in the annotation property.
  • This doesn't fit into LexEVS model (an association on an association).  The LexEVS API wouldn't handle this well either.

Decision Points:

  • Scott mentioned that our load will ignore this case
  • Scott will sent a snippet source example to Gilberto.
  • Gilberto - from the snippet that Scott provided: an annotation property that doesn't have a range.
  • There have been JIRA issues that were created for the points that Gilberto discussed.
     

Decision Points:

  • Gilberto to send an OBO example code snippet for the issues he discussed
Docker - Next Steps Discussion

Discussion Points:

  • Docker Contacts: Mike Stockwell & Sue Pan

Decision Points:

Resolving all relationships for a given association

Thesaurus OWL2 Source

Discussion Points:

  • Our loads are now making a lot of calls into the ontology class.
  • This is increasing our Thesaurus OWL2 loads from 1hr to 3-6hrs.

Decision Points:

Scott will create a JIRA issue to investigate the new load performance issue.
Jira
serverNCI Tracker
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
serverId7954a81f-12da-3366-a0ef-97c806660e7c
keyLEXEVS-2771

Remote API timeout issue

  • Discussion with Kim

Discussion Points:

  • What is the use case for resolving all relationships for a given association. 
  • Is there a requirement for Enumerations
  • There is a workaround that can retrieve these relationships, but we should discuss implementing a more efficient API call
  •  Intermittent socket timeouts from the Remote API.
  • Scott is able to reproduce this issue locally.

Decision Points:

Scott will investigate why our system tests are not hitting the same issue.

LexEVS External Users

Discussion Points:

Nothing noted.

Decision Points:

Proposed Backlog Review Items

Discussion Points:

Decision Points:

Team Absences

Mayo Team

  • Cory -
  • Scott -
  • Craig - March 6-13

NCI

  • Tracy
    • March 9-17
  • Rob
    • March 20-24

...