Turn messy biomedical metadata into clean, standards‑compliant records in just a few lines of code.
The Netrias Harmonization platform provides REST endpoints, a Python CLI, and a prototype user interface for:
- CDE discovery – automatically find the best Common Data Element (CDE) for an arbitrary table column.
- Value harmonization – map free‑text cell values to controlled vocabularies.
- End‑to‑end pipelines – batch‑convert entire spreadsheets into standards‑ready JSON.
Follow this sequence for a smooth on‑boarding. Each step links to a dedicated page with examples and API snippets.
| Step | Topic | File | Why read it first? |
|---|---|---|---|
| 1 | What We Harmonize | what-we-harmonize.md |
Learn the core concepts & data models (CDEs + Schemas). |
| 2 | Request an API Key | requesting-API-key.md |
Get your credentials to call the service. |
| 3 | Install the CLI Tool | apiclient.md |
Quick local setup for scripting & command‑line experimentation. |
| 4 | Install the UI Tool | data-chord.md |
No-code user interface for guided harmoization. |
| 5 | CDE Recommendation | cde-recommendation.md |
Discover which CDEs match your columns. |
| 6 | Value Harmonization | harmonize.md |
Convert raw values into permissible values. |
| 7 | Advanced Pipeline | example-use-cases.md |
Automate steps 4‑5 end‑to‑end on a whole table. |
| 8 | Swagger / OpenAPI | https://apiserver.netriasbdf.cloud/docs | Interactive playground & full endpoint reference. |
| 9 | Submit Your Own CDEs | requesting-data-be-added.md |
How to get your custom data loaded into the platform. |
Tip: Bookmark the Swagger docs - they’re always up to date with the latest versions and error codes.
pip install git+https://github.com/netrias/bdf_harmonizationSee requesting-API-key.md and email the necessary info to us. Store the key and URL in an environment variable:
export NETRIAS_API_KEY="<YOUR_KEY>"
export HARMONIZATION_API_URL="https://apiserver.netriasbdf.cloud/v1/harmonize"apc harmonize 1006 "nf" | jq .(Maps the string “nf” against the diagnosis CDE in Sage Bionetworks' Neurofibromatosis (NF) data model and prints the ranked options.)
- 🔍 /mapping endpoints for programmatic CDE‑ID lookup.
- 🔄 Versioned CDE support - store multiple historical CDE releases and let clients specify which version to harmonize against.
- 🗂️ Bulk upload of TSV/CSV/Excel files.
Generously supported by ARPA-H via funding from the Biomedical Data Fabric (BDF) Toolbox program.
We gladly accept pull requests that improve docs, examples, or client code. Please open an issue first if you plan a large change.
© 2025 Netrias LLC - Released under the Apache 2.0 license.