BDIViz
BDIViz is an interactive web-based visualization tool developed as part of the ARPA-H ASKEM/BDF project. It supports schema matching and value mapping tasks in biomedical data integration, helping researchers align their raw datasets with standardized formats such as the Genomic Data Commons (GDC) and Proteomic Data Commons (PDC).
Open Source MIT Licensed.
Why choose BDIViz?
BDIViz is a visualization-driven, expert-in-the-loop tool designed to simplify and improve schema matching and value mapping in biomedical data integration. Explore the key features that make BDIViz intuitive, powerful, and research-ready.
Interactive Heatmap
Visualize schema matching candidates across all columns using an intuitive heatmap interface. Color-coded cues help quickly identify strong and weak matches.
Value Comparison
Easily compare unique values between source and target columns using fuzzy matching and histograms to validate semantic alignment.
LLM Agent Explanations
Get contextual explanations for candidate matches powered by large language models. Understand the rationale behind every recommendation.
Timeline & Undo/Redo
Track every decision with a timeline view and easily reverse or reapply actions using undo/redo support.
Export & Integration
Export your final mappings as CSV or JSON to use in downstream tools, or integrate with other pipelines such as Beaker or BDIKit.
Built for Biomedicine
Tailored for biomedical researchers, BDIViz supports common data commons like GDC and PDC, and was co-designed with domain experts through the ARPA-H ASKEM project.