Technical brief · Version 1.0 · 2026-04-27

Genetic History Module — Technical Brief

For: clinical-lab technical evaluator

1. What it is

EU-based B2B push-model software module installed in clinical / medical laboratory infrastructure for Y-DNA / mtDNA ancestry-narrative generation — partner customer data never transfers out.

The module turns NGS pipeline output (VCF) into a research-grade haplogroup narrative plus per-subclade geographic heatmap. Partner delivers the narrative to its patient via the partner’s existing patient-communication channel.

2. Deployment form

Docker image, published to a container registry (GHCR primary; private registry on request). Multi-stage build, distroless base. Single command:

docker pull ghcr.io/genetichistory/module:v1.0
docker run --network none -p 8080:8080 -e PARTNER_KEY=... \
  ghcr.io/genetichistory/module:v1.0

--network none is supported and recommended — the module operates fully offline at runtime (see §6).

3. Input contract

4. Output contract

5. Integration surface

End-to-end:

curl -X POST http://localhost:8080/v1/analyze \
  -H "Authorization: Bearer $PARTNER_KEY" \
  -H "Content-Type: application/json" \
  --data @sample.vcf.json

6. Data locality (architectural invariant — do not weaken)

Partner customer data never transfers to Genetic History S.L. infrastructure. No egress. No telemetry carrying sample identifiers. The module operates fully offline after the container starts; zero call-home, zero usage analytics, zero runtime logs sent to Genetic History.

Verifiable by the partner: docker run --network none ... confirms a full report served without any external call. Any opt-in metadata (e.g., aggregate library-version statistics) is documented, off by default, and contains no patient identifiers.

7. Library update protocol

Library refresh ships as a new image tag (v1.0.x). Partner pulls on a partner-controlled cadence (recommended weekly / monthly / quarterly per partner stability preference). No silent auto-update; partner controls every version transition.

8. Resource footprint

Dimensionv1.0
Image size≈ 500 MB – 1 GB (assets bundled, fully offline)
RAM1 GB minimum, 2 GB recommended
CPU1 vCPU minimum
Disk2 GB free (image + scratch tmpfs)
Runtime per sampletypically 5–30 s

9. Supported coverage

10. Compliance posture

DPA template + sub-processor registry available on request — see contact §12.

11. Sandbox availability

v1.0: prospect technical evaluation via dedicated staging endpoint, provisioned per pilot. Request access via §12 contact. v2.0: self-service sandbox planned post-Wave 1 validation.

12. Versioning + contact


Genetic History S.L.U. operates the YFull bioinformatics service (yfull.com) under the same corporate perimeter; B2B clinical-lab integration is the focus of this brief.

Direct contact: urasin@yfull.com