Public genomic evidence, gathered into versioned, ML‑ready records.
GA4GH-alignedVersioned records
What is THREAD?
Web-based application that serves as a genetic library tool for storing and sharing genetic files.
THREAD is BIOPRIUM's data foundation: a GA4GH-aligned genomic data platform. It gathers public genomic evidence into structured, versioned records, builds ML‑ready datasets from them under strict leakage controls, and feeds those records to ARIADNE.
Day to day, its workspace keeps runs, file transfers, meetings and records in one place.
From public sourcesto ML‑ready records.
An automated pipeline brings the evidence in, gives every record an accession and a version, and builds datasets under strict leakage controls. The records leave for ARIADNE through the JSON Record Manifest API.
- Public sources: ClinVar, gnomAD v4, MaveDB, ClinGen, HPO, and more.
- THREAD ingests them through an automated pipeline.
- Each record gets a structured accession, a Crockford base32 ID with a check digit, and a version.
- Datasets are built from the records under strict leakage controls, so they are ready for machine learning.
- The records leave for ARIADNE through the JSON Record Manifest API.
Built to be checked.
GA4GH-aligned
THREAD is aligned with the standards of the Global Alliance for Genomics and Health, the body that develops frameworks for sharing genomic and health data responsibly.
Leakage-safe datasets
When THREAD builds a dataset for machine learning, strict leakage controls keep the data a model is evaluated on apart from the data it learns from.
- Track runs
- Schedule when to send genetic files
- Schedule any meeting concerning the patient’s genetic info
All in 
.
Questions about THREAD?
Tell us what you are working on and what you need from THREAD.