Skip to content

2025 04 17 steering group minutes

Agenda

  1. Guest: Maryanne Martone (SPARC)

  2. Copenhagen maintainer meeting

  3. Follow up on cardiac BEP

  4. BEP045 (Physio) > email from BEP045 (Physio), is input needed from steering?

  5. Maintainers Updates [name= all maintainers]

Agenda

  • Yaroslav Halchenko
  • Dora Hermes
  • Kim Ray
  • Eric Earl (Maintainer rep)
  • Maryanne Martone (SPARC)
  • Tom Gillespie (SPARC)

Meeting Minutes

SPARC

  • originally noted and some pointers in https://github.com/bids-standard/bids-2-devel/issues/62

  • History

    • Originated in 2018

    • Physiology, microscopy...

    • NWB was not mature then

    • Found BIDS useful and thorough but not enough for physio, micro... data

    • raw, derived, and primary data

    • on population, specimen, subject, sample: https://github.com/SciCrunch/sparc-curation/blob/master/docs/participants.org

    • Tom Gillespie instrumental in implementing version 2 of SPARC

    • Just released SDS 3.0 (MIS = minimal information standard)

      • includes an SDS viewer to allow seeing the over structure of a dataset (graph based)

      • allows for "flexible hierarchy"?

        • reorders the hierarchy categories to subject level

      • Sample, Subject, and manifests

    • Study > Subjects > Protocols > Subject Groups

    • SDS allows nesting of datasets to support interoperability, but they cannot have the same IDs/naming schema

    • conversion/renaming/moving process of data into SDS is often a source of error

      • solution: validation during upload to SPARC database

    • allows to adding additional "standards" a particular dataset conforms to (for example RE-JOIN/HEAL) and providing additional jsonschemas for specific files.

  • SPARC was always meant to be multi-modal

  • Tries to be "Flexible" to accommodate different requirements of multiple "actors" in a consortia

  • EX: SDS allowing embedding of BIDS: /.dss (ad-hoc prescription of a dataset standard)

  • Have an extensive list of "modalities/data types", which also have strong accent on genetic data and also "computational models"

  • Questions Yarik planned to ask but didn't

    • Any specific rules or "mapping" regex or alike for the file names?

    • Any software/libraries to deal with SDS formatted datasets?

    • If not NWB -- what file format for electrophys data?

    • Given current state of NWB, BIDS and various BEPS -- would you consider just going after BIDS?

      • What aspects of SPARC to bring in?

    • There is a validator (and also mentioned in https://github.com/con/validation/issues/1), but how "detailed" validator is, what aspects does it validate?

  • Yarik: I think we should aim to translate such presentations into specific actionable items on how to improve BIDS

  • for example here the idea from SDS to provide custom jsonschemas for extra validation of some .json files. that somewhat relates to an issue where I added a comment now: https://github.com/bids-standard/bids-2-devel/issues/74#issuecomment-2813819345

Copenhagen Meeting

maintainers need to get travel information to Cyril ASAP - goal is this week or next.

Follow up on cardiac BEP

BEP045 (Physio)

Maintainers Updates

  1. Getting BEP status updates in the next 2-4 weeks.
  2. Would like to talk about the potential Virtual BIDS Town Hall this year.

Tasks

\==Importance== (1 - 5) / Name

Steering Group

  • [ ] BIDS townhall

Maintainers

  • [ ] Send travel info to Cyril