Docs · OGN platform
GPU-native genomics operating system
From raw reads to GIAB-validated variant calls in a continuous GPU pipeline. This is the control surface for the engine: CLI, pipelines, benchmarks, and deployment runbooks.
CUDA 12+Hopper · AmpereGIAB-validated flowsSchemas stable
Viewing
Paid pilot offer
Snapshot Report offer
The Snapshot Report is a fixed-scope engagement designed to answer a single question: should you adopt OGN for GPU-accelerated variant calling on your data?
Deliverables (what you get)
- A reproducible run on a representative dataset (yours or a public substitute you approve).
- A shareable proof bundle artifact folder you can forward internally:
vcf(+ optional index)provenance.json(engine version, container digests, hashes)metrics.json(runtime / throughput / cost hooks)logs.ndjson(downloadable + streamable)summary.md(human-readable recap + rerun commands)
- Runtime + cost comparison against your stated baseline (DRAGEN / Parabricks / Sentieon / GATK, etc.).
- Written acceptance criteria agreed up front (accuracy, runtime, cost, reproducibility).
Inputs (what we need)
- Dataset type + size (WGS / exome / panel, sample count, typical coverage).
- Baseline target (what you want us to beat, and the constraints you care about).
- Deployment shape (cloud/on‑prem, GPU model, storage).
- Any red lines (PHI handling, data residency, networking constraints).
Timeline (typical)
- Week 1: intake, baseline capture, environment constraints, acceptance criteria doc.
- Week 2: runs, proof bundle delivery, runtime + cost comparison, readout.
Pricing model
- Fixed Snapshot Report fee + compute pass-through (scoped to a fixed number of runs / iterations).
- Enterprise licensing is discussed only after the Snapshot Report produces a deterministic proof bundle.
Next steps
- Submit a request at
/company/contact. - If you want to talk first, include two meeting times (with timezone) and we’ll confirm quickly.