Bioschemas community call - Join us on Monday 2025.09.11 at 17:00 CEST

Dear all at Bioschemas community,

We have an invented talk for our next community call. Join us if you can!

*What:* Bioschemas community call - Invited talk "The FAIRSCAPE Digital
Commons for Ethical AI in Translational Medicine"

*When:* Monday 15.Sep.2025 at 17:00 CEST

*Where:*
https://us02web.zoom.us/j/81552565788?pwd=2yI4nyybEdCbYAq1yGkdD2U8loQ5EJ.1

*Who:* Tim Clark, University of Virginia School of Medicine, Department of
Public Health Sciences

*Abstract:* Biomedical and translational medicine applications of AI impose
strong explainability (XAI) requirements for reliable research results and
ethical translation to the clinic. These include a need for robust
pre-model XAI to prevent black-boxing of data acquisition and preparation
events. FAIRSCAPE is an open-source digital commons platform developed to
support these needs in collaboration with University of Virginia and
University of California San Diego biomedical researchers as part of the
$130M NIH Bridge2AI program.  It provides a digital commons environment in
Python implementing the Bridge2AI AI-readiness principles on biomedical
datasets and provides ethical FAIRness with deep semantic provenance graphs
on components such as datasets, software, computations, runtime parameters,
environment and personnel involved in a computational analysis. FAIRSCAPE
generates detailed human- and machine-readable biomedical Datasheets using
JSON-LD serialized schema.org and Evidence Graph Ontology metadata packaged
in RO-Crates, including Croissant and Croissant RAI metadata for direct
interface to Kaggle and other AI/ML Ops packages, and provides a rich
environment for biomedical AI researchers. Remote client tools provide
packaging of these metadata with datasets, or dataset references, and their
schemas, for upload to a server running under Kubernetes. Data is held in
S3-compliant buckets and metadata is stored in MongoDB. REST APIs are
provided along with a REACT-based Web interface to the server and export
interfaces to NIH-recommended long-term archives.

*Short bio:* Clark’s work intersects several fields, including public
health, data science, and neurology. He teaches in Public Health Sciences
and Neurology in the School of Medicine, and in the School of Data Science.
Clark’s research interests include biomedical informatics, neuroscience,
Alzheimer’s Disease and disorders of cognition, knowledge representation
and integration, digital commons frameworks, ontologies, FAIR data, FAIR
software, FAIR computation, argumentation frameworks, and evidence graphs
*Agenda:*
https://docs.google.com/document/d/1kd5F97ogdiPNhLTnkei-RVR8TC8Ohpc5QSPX3KsfDrk/edit?tab=t.0#heading=h.juzanf5x2abw

Kind regards,

On behalf of the Bioschema Steering Committee