Your family is not the first.
Families with a rare diagnosis often work alone, because the registries, studies and tools they need sit under a neighbouring disease with another name. Constellate is a map of rare diseases that finds those connections and shows the proof for every one of them.
How it works
- 01
Search a diagnosis
By disease, gene or symptom, in any of its names or IDs.
- 02
See its neighbours
Diseases are linked by shared genes and symptoms. Every line is a piece of evidence you can open.
- 03
Compare
What transfers from a related disease, what differs, what needs an expert. Every sentence cites its source.
How sure is a line?
Every link carries one of four evidence levels, a confidence, its sources and a counter-evidence field that may be empty but never missing.
observedClinically or experimentally observed, with a primary sourceClinVar, HPO annotations with frequency, trial recordscuratedStatement from an expert-curated databaseOMIM (via HPO), Orphanet, Reactome, MONDOtext_minedExtracted from an abstract by a language model, not yet verifiedPubMed / Europe PMCinferredComputed by Constellate, formula and inputs shownSymptom similarity, shared genes, clustering
Honest gap
When there is no trail, we say so
Where no source supports a link, Constellate does not guess. It shows what was searched and when, which communities are nearest and what would close the gap.
Data sources
- HPOfree with attribution
- MONDOCC BY 4.0
- OrphanetCC BY 4.0
- ClinVarpublic domain
- ClinicalTrials.govpublic domain
- Europe PMC / PubMedpublic domain
- NIH RePORTERpublic domain
- Open TargetsCC0
- ReactomeCC0
- MonarchCC BY 4.0
OMIM content enters only indirectly, via HPO annotations and Orphanet cross-references.
AI inside
Language models extract claims from abstracts, match organisation names to diseases and explain a comparison in plain language. Every such use is marked, every evidence item names its model, and every explanation sentence must cite an evidence line.
Team
Hanno Kügler, Osman and Recep.
Source code on GitHub ↗