For researchers

Send us a paper

Every one of the 2,971 variant pages and 83 condition pages here was written from published research — most of it yours. Until now the traffic only went one way.

Start by looking for your own work.

889 papers hold this site up and yours may already be one.

Is your paper here? → Go straight to the form →

We read it, and we connect it

That is the whole of it. What comes out is a page in our own plain words — what the position is, what each genotype means, what the evidence does and does not support — and an edge in the Gene Knowledge Graph: this variant to that condition, with your paper named as the reason the line is drawn.

A finding is not anybody's property. The sentences you wrote are, and we do not take them — no quotation beyond a citation, no figure reproduced, no abstract pasted. Nothing is published without your paper cited and linked, and we are not a journal: no review, no acceptance, no promise to write anything at all.

The most useful things to send

A submission need not name a variant. Guidelines, prevalence estimates and methodological work are all in scope.

If you send a copy

Sending a reader a copy of your own paper is a reprint request, which authors have done for a century. We treat it as one.

Being credited

If we use it we say so, on the page, by name, with your paper linked. Tick the box and we ask you first — every time, before your name appears anywhere.

A credit here is not a publication and is not presented as one. It is a public, dated and verifiable record of a contribution to an open reference.

It also travels further than the page. The structured data is served to anything that asks — an OpenAPI spec, llms.txt, schema.org on every page — so when an assistant answers from this site, the paper it draws on is named, together with its full author list rather than the three shown on the page.

Or take the data instead

The other direction has always been open: the structured fields are CC BY 4.0, free for any use including grounding a model, with attribution. API · OpenAPI · snapshots. Every condition page carries its open_questions — what we could not resolve. Answering one of those is the most direct contribution there is.

Terms, plainly

You keep every right you have; nothing transfers and we license nothing onward. What we publish is written here from the facts, in our own words. Your address is used to reply and nothing else. Ask us to remove your name or your submission and we will. Nothing here is peer review, endorsement, or medical advice.

The form

Name and email are optional unless you want the credit. Everything sent here is read by a person, and none of it is published without asking you.