About
About DNADojo
A small, deliberately narrow toolbox for genetic genealogy — built so that privacy is a property of the architecture rather than a paragraph in a policy.
Updated 2026-09-13 · 5 min read · Genealogy & education only
Why this exists
Tens of millions of people have taken a consumer DNA test, and a large share of them end up staring at a number they cannot interpret. "You share 857 cM with this person." What does that mean? Who is this? The testing platforms answer with a vague band and a prompt to subscribe.
Meanwhile, the events of recent years have made a second point hard to ignore: a company holding your genetic data is only as stable as its balance sheet. Ownership changes. Policies change. Your genome does not.
DNADojo is the response to both: make the interpretation genuinely good and free, and make the privacy structural rather than promised.
Principles
- Local-first, by architecture. If there is no server that can receive your DNA, the promise is hard to break by accident.
- Honest uncertainty. A cM number produces a shortlist, not an answer. Every result says so.
- Versioned science. Every figure carries a dataset version and a source ID, so a result you saved last year can be traced back to the numbers it was based on.
- A clear medical boundary. Genealogy and education only. No disease risk, no drug response, no nutrition advice — no exceptions for "informational purposes".
- Free where it matters. The tools that answer "what relationship is this?" stay free. Payment, if it comes, is for saving and organising research.
What we are not building
- A testing company. DNADojo does not sequence anything.
- A cross-user matching database. There is no pool of strangers' DNA here to search, sell or subpoena.
- A health interpretation service.
- A clone of the excellent tools that already exist. Where a good tool exists, DNADojo links to it rather than reimplementing its data.
How it is built
| Layer | Choice | Why |
|---|---|---|
| Site | Astro + TypeScript, statically generated | Fast, crawlable, tiny attack surface, no database to maintain |
| Calculators | Pure TypeScript, no framework | A few kilobytes of JavaScript per page instead of a hydrated app |
| Raw file parsing | File API + Web Worker, streaming | Large files without freezing the page, and nothing leaves the device |
| Data | Versioned JSON with a CI validation gate | Science can be updated without touching product code |
| Hosting | Static CDN | Near-zero running cost, which is what keeps the tools free |
Roadmap
| Phase | What ships | Gate to move on |
|---|---|---|
| Phase 1 | Five calculators, relationship guides, learning content | Real people complete real calculations; the dataset passes scientific review |
| Phase 2 | Local DNA Explorer with format detection and SNP search | No raw file ever leaves the browser, verified in the network panel |
| Phase 3 | Pro research workspace — match notebook, hypotheses, projects | Demonstrated willingness to pay, and a completed legal review |
Everything in Phase 3 syncs research metadata only. Raw genotypes stay local no matter how the product grows.