Founding Lab programme
Ten labs. Ninety days. Your own data.
We are inviting ten HLA laboratories to run HLA-Verify on their real typing data before anyone else. You get the full Lab tier free for 90 days, with no card and no contract to start. If it earns its place, you keep it at 60% off for two years.
The offer
What a founding lab gets
| During the trial | After the trial | |
|---|---|---|
| Length | 90 days from your key | 24 months at the founding price, then the list price |
| Price | $0, no card | $119.60/mo instead of $299 (60% off Lab) |
| Calls per day | 10,000 | |
| Typings per batch call | 5,000 | |
| Support | Direct line to the person who built the engine, next business day, and every report reviewed on support | |
The founding price is fixed before the trial starts, so keeping it is a yes, not a negotiation. If you do not convert, the key simply stops at day 90: nothing to cancel.
See it on your own data
A report on your typings in minutes
Day one, run the trial report on any LIMS export or typing file you already have. It finds every HLA allele name in it and checks each one against the pinned IPD-IMGT/HLA release in batches of 5,000.
What it shows
- Names that do not exist in any release: typos and fabricated names
- Deleted and renamed alleles, with their current names
- Outdated spellings: colon-less names, Cw*, legacy forms
- Null alleles, ambiguity codes and group names worth a second look
- How long every name in the file took to check
What leaves your lab
Only the allele names themselves, deduplicated. The report is a small open-source script that runs on your machine: patient names, record numbers, dates and every other column stay where they are. HLA-Verify processes the names in memory and stores none of them. Read the script before you run it: poc_report.py.
What we ask
What we ask in return
- A named contact who will actually run it, and a 20-minute call at day 30 and day 90.
- Success criteria agreed at the start: for example, "check our last 12 months of typings and show us what needs correcting".
- If it works for you: permission to list your lab as a founding lab and a short quote. If it does not, tell us why; that is just as useful.
- Reports of anything that looks wrong, through support. Every one is reviewed, and confirmed errors are fixed and published.
Before you apply
What HLA-Verify is, and is not
- Deterministic. No model in the loop: every answer is computed from the official IPD-IMGT/HLA files and reproducible from the release it names. See how it works and the public benchmark.
- Not a medical device. It is a research-and-evaluation tool that supports your validated procedures and does not replace them. The terms of use and Trust page cover data handling.
- Ten places. We take labs in the order they are ready to start, so the trial is used, not parked.