OpenEndo 🎗️ — open data hub for endometriosis

Open intelligence for a disease the world still ignores.

Endometriosis affects ~190 million women worldwide (1 in 10 of reproductive age), yet the average patient waits 7–10 years for a diagnosis, research funding is a fraction of what the burden demands, and there is still no cure.

This repository is an open, weekly-refreshed data hub: clinical trials, research papers, funding deadlines and policy — for patients, relatives, researchers, journalists and politicians.

OpenEndo (formerly endometriosis-watch): open source, MIT licensed, built for collaboration. The data is yours.

Live site

👉 https://openendo.org/ — the public UI, designed & deployed by Jaeger via Lovable (a Next.js app; it is NOT served from this repository). This repo is the canonical data & knowledge layer: every dataset, the wiki and the research program live in docs/ and are consumed from GitHub (raw.githubusercontent.com/wckdboy/openendo/main/docs/...) — see docs/llms.txt for the full agent index. Leftover docs/*.html is not the product UI (GitHub Pages CNAME-redirects and does not serve those files). The one live-fetch exception is docs/what-we-know.html — openendo.org/research loads it from this repo.

More

• docs/llms.txt — machine-readable index (agents: start here)
• BRAND.md — brand tokens & conventions, machine-readable (for AI agents)
• Live product: openendo.org · dashboard · research · access · knowledge
• docs/what-we-know.html — research explainer (live contract for /research — do not delete)
• docs/styleguide.html — visual design system (legacy HTML; no live route)
• docs/support.html — seven ways to help (legacy HTML; no live route)
• docs/ai-agenda.html — what agents/algorithms/compute can do for endometriosis
• docs/one-pager-dk.html — one-pager for Danish politicians (unique; keep)
• CONTRIBUTING.md — PR checklist

What's inside

docs/                         canonical open data + knowledge + reference pages
                              (consumed via raw.githubusercontent.com; the live
                              UI is the Lovable app at openendo.org)
  data/                           open datasets (JSON, refreshed weekly)
    trials_global_recruiting.json   recruiting trials worldwide        (ClinicalTrials.gov)
    trials_denmark.json             all registered trials in Denmark
    trials_recent.json              new/updated trials, last 7 days
    pubmed_recent.json              new papers, last 7 days             (PubMed)
    pubmed_monthly.json             papers per month, last 6 months
    meta.json                       generation timestamp + counts
    funding.json                    funding opportunities with deadlines (curated)
    content.json                    stats, problem framing, actions, resources (curated, EN/DA)
scripts/update_data.py        regenerates docs/data/*.json from the APIs, commits & pushes

How the data is refreshed

Every week an automated job runs:

python3 scripts/update_data.py

It queries the ClinicalTrials.gov API v2 and PubMed E-utilities, writes the JSON files, and pushes a commit only when something changed. The data pipeline needs no server — the JSON contract is consumed by any UI (Lovable app, notebooks, tools).

Sources

• ClinicalTrials.gov API v2 (public domain / NLM data usage policy)
• PubMed E-utilities (NCBI)
• Curated editorial content with sources in data/content.json
• WHO fact sheet on endometriosis; Danish patient organisation (endo.dk); Lægehåndbogen (sundhed.dk)

Knowledge base

OpenEndo runs a knowledge base for agents and humans — interlinked markdown with provenance, confidence markers and a review workflow:

• Agent discovery: docs/llms.txt — machine-readable index of data, knowledge and research (raw: https://raw.githubusercontent.com/wckdboy/openendo/main/docs/llms.txt)
• Conventions: docs/knowledge/SCHEMA.md (agents must read before contributing)
• Index: docs/knowledge/index.md · pages in entities/, concepts/, comparisons/, queries/
• Review workflow: agents contribute via PRs; the weekly monitor job reviews and merges
• Privacy rule: no personal data, patient stories or PII in this repo — enforced by a CI scan (.github/workflows/security-scan.yml). Family/health/business context lives in a separate private repo.

Contributing

This project exists to be used and extended. Ways to help:

• Submit leads — open an issue with funding calls, new trials or policy developments we've missed
• Translate — the site is bilingual (EN/DA); adding more languages is a small PR
• Add resources — patient organisations, diagnostics, centres (docs/data/content.json)
• Run the pipeline — fork, python3 scripts/update_data.py, PR the refreshed data
• Spread the word — this disease is ignored because it is invisible; sharing is a contribution

Good first issues are tagged good first issue.

License

MIT — reuse the data, the site and the scripts freely. Attribution is appreciated.

Disclaimer

This is research and advocacy, not medical advice. Treatment choices must always be made with a specialist. Project is independent and not affiliated with any of the organisations referenced.

Open source, independent and unaffiliated — built for patients, researchers and policymakers. The data is a public good.