Tuesday, 2022/03/15
10:00 am ET
Terrell Russell, Kory Draughn, Alan King, Daniel Moore, Dave Fellinger, Simona Stoica (Groningen), Paul Borgermans (KU Leuven), Mike Conway (NIEHS), Mohammad Shaikh (Bristol Myers Squibb)
Update/Discussion:
- Goal: Paper for UGM2022, put this group's learnings out in the world
- Shared document, outline first, work outside of this meeting time
- Captured minutes into 'outline', now in 'whitepaper' branch
- KU Leuven
- Building a portal
- Looking at a metadata template / schema editor, and using require/optional
- Mixed mode of collection and data object level
- And enforcing validation
- BMS, Dataflow, drug discovery pipeline, from/through many teams
- Scientists want to have the data local for analysis
- But no concept of checkin/checkout
- To manage data as an asset
- 'Data bundles' - defined by shared metadata / AVUs
- A system for 'locking' as well as auditing/reporting about activity on a data object
- Perhaps a new / next working group - "Pipeline/Sharing"
- Scientists want to have the data local for analysis
- administrative/descriptive/datatypes distinction in the templates themselves?
- Define the magic fairy dust - best practices - examples - domains...
- Next meeting
- April 2022