metadata

Information about mapping to published open schemas and ontologies.

Goals

1. Make Databrary compliant with open data schema standards, wherever possible.
   • NIH Common Data Elements (CDE). High priority.
   • DataCite. High priority.
   • Schema.org. High priority.
   • JSON-LD. High priority.
   • DataTags project at Harvard project. Moderate priority.
   • Cognitive Atlas. Low priority.
   • CSV Schema Language.
   • Brain Imaging Data Structure (BIDS), especially the Hierarchical Event Descriptors (HED) specification.
     • See Databrary-related example here.
2. Support users who want to provide data compatible with these standards to do so in straightforward ways that enable self-curation.
3. Support linking data elements to specific, version-controlled, data schemas.
4. Eventually, support linking data sets or projects to concepts, tasks, disorders, or theories defined on Cognitive Atlas using the Cognitive Atlas API.

Documents

• NIH CDE and Databrary 1.0/PLAY Project.
• Google sheet mapping existing most Databrary data elements and API calls to schema.org and datacite.org.
• Current Databrary 1.0 schema specifications
  • Volume (will be called project in Databrary 2.0) specification.
  • Metadata requirements for automated (not self-curated) ingest of data into Databrary 1.0.
• Sketch of how a dataset from PLAY might be structured using the Brain Imaging Data Structure (BIDS) format.

Tools

• databraryapi R package.
  • There is a Python version that is less well developed.
• Internal Databrary reports that use the databraryapi R package.
  • Which volumes/projects have shared sessions and of what type (overview only or full volume)?
  • What user-defined tags or keywords have been added to volumes/projects?
  • What types of files have been shared?
  • What volumes/projects provide demographic data about individual participants using the Databrary spreadsheet?
  • Data about which shared projects have indicated funding sources.