These are the priorities for Sprint number 1, which runs from Feb 12, 2024 to March 8, 2024. In this sprint, the Lobster build is in CI and the Pre-Lobster build (based on the Feb 6 release) is in Test. Each priority has a list of subtasks that are to be completed within this sprint.
- Determine 2-3 specific use cases for a Pathfinder query that provide value to a user AND have known inputs/expected outputs.
- Use those specific use cases as the basis for discussion. Brainstorm to identify knowns and unknowns (eg. we know TRAPI and UI changes will be needed - what else needs to be considered?)
- Identify next steps and map out what can be completed in the upcoming development cycles
- Discuss TRAPI changes and either outline a plan for those changes AND/OR set planning goals for TRAPI for Octopus Backend Development cycle
- Discuss UI changes and either outline a plan for those changes AND/OR set planning goals for upcoming development cycles
- Push to CI
- Push to Test
- Test for controversial items
- Test for child terms
- TRAPI defines syntax for bypassing caches in tools
- Send concurrent queries (MVP1 & 2) in batches of 5, 10, 15, 20, 25, 30
- Send concurrent queries (MVP1 & 2) in batches of 5, 10, 15, 20, 25, 30
- ARAs review the changes needed to continue use of treats
- Document usage of "Drug" vs "Procedure" vs "Treatment" categories
- Provide edge examples in Biolink documentation for "treats" predications
- Complete the "treats" refactor source transformation spreadsheet
- KPs implement Biolink 4.x including the refactoring of treats edges
- Extend CQS to support new MVP workflows
- Develop valid TRAPI queries and deposit to CQS repo
- Identify relevant tests assets and deposit to Testing Team's G-sheet/repo
- Direct CQS tests in DEV using test assets
- Push initial workflows (Paths A, B, E) to TEST (end of cycle) contingent on CI tests
- Hook up CI instance of CQS to ARS in CI
- Direct CQS CI tests using test assets with new Q-A pairs
- ARS CI tests using Information Radiator
- Determine 2-3 specific use cases for a MultiCURIE query that provide value to a user AND have known inputs/expected outputs.
- Use those specific use cases as the basis for discussion. Brainstorm to identify knowns and unknowns (eg. we know TRAPI and UI changes will be needed - what else needs to be considered?)
- Identify next steps and map out what can be completed in the upcoming development cycles
- Discuss TRAPI changes and either outline a plan for those changes AND/OR set planning goals for TRAPI for Octopus Backend Development cycle
- Discuss UI changes and either outline a plan for those changes AND/OR set planning goals for upcoming development cycles
- Address Confidence (aka g()) and Ordering (aka f()) scores for ordering / confidence score ties for MVP1 and MVP2
- Clarify testing process (data, metrics, pass/fail)
- Each project specifies its benchmarks
- Make CoLab code ready for O&O tests
- Add tags to the information radiator to make clear which ARAs are failing
- Pull tagged information from the radiator into a spreadsheet grouped by ARA
- Decide per-team targets (how many failing tests need to be fixed per team) based on information in the spreadsheet
- ARAs implement fixes so that no ARA has more than 100 failing tests or more than 100 tests that do not return results
- Draft documentation / implementation guidance
- Outreach to KPs to collect concerns / feedback
- Outreach to ARAs / O&O about use cases and utility of current terms
- Identify 5-10 issues that can be fully resolved
- Work to resolve these issues
- Review progress at TAQA meetings