The mission
One rare disease is rare. All rare diseases are not.
MISHA Foundation accelerates rare genetic disease research by funding the labs that move fastest, building family-controlled patient registries, and putting computation to work where it moves biology forward. It started with one child's hearing loss. It is built to outlast his specific cure.
There are approximately seven thousand rare diseases and 350 million people afflicted by them worldwide. Approximately 95 percent of rare diseases do not have a single FDA-approved therapy because they are so poorly understood. Simply put, you can't create a targeted treatment when you don't know what needs to be targeted.
Why this exists
When you are told your two-year-old will not hear the way other children hear, you are also told, gently, to wait. Wait for research no one is funding on your timeline. Wait for a trial that does not exist.
STRC sits in a blind spot: common enough to matter, rare enough for commercial pipelines to ignore. The biology is tractable. The missing piece is coordination, money moving to the right benches, and families who can find each other.
MISHA stands for Medicine, Intelligence, and Science for Human Action. The honest expansion is shorter: it is my son's name. Named for one child, scoped for the thousands who carry the same and adjacent variants. I do not own the science. I fund it, connect it, run the computation, and keep the receipts in public.
Egor Lyfar, founder and Misha's father
How a small foundation moves a whole field
A specific, borrowed playbook. Four moves, in order.
Start where the gap is
Hearing-genetics research clusters in the West. We put the work in Asia, where Mainland trial pipelines meet Western lab quality.
De-risk with computation
Rank the hypotheses, model the structures, find the druggable pocket in silico first. The first wet-lab experiment should be the cheapest one, not the first guess.
Seed the labs that move
A 50k to 150k grant buys the preliminary data that unlocks a grant ten to fifty times larger. Repeated, that leverage is how the SMA and cystic fibrosis foundations reached their first therapies.
Keep the receipts in public
Every decision, failed application, and ranking delta goes on the public wiki. For a one-person foundation, legibility is how you earn the right to hold real money.
Not an advocacy-only group, and not a single-target biotech. Closer in spirit to the early SMA Foundation, the Rett Syndrome Research Trust, and the Usher 1F Collaborative: small, technical, and pointed at tools the whole field can use.
Where computation actually helps
A specific bet about which parts of rare-disease research got cheap, used to spend wet-lab money last, not first.
Structure used to take years
Predicting a protein's shape once meant years at a crystallography bench. AlphaFold has released more than 214 million predicted structures, and its authors shared the 2024 Nobel Prize in Chemistry. Inference now runs in hours.
AlphaFold DB, Nucleic Acids Research 2024; Nobel Prize 2024
Rare disease is mostly monogenic
About 72% of rare diseases are genetic, usually a single gene. That is the tractable case: model the variant, rank the hypotheses, find the pocket, before anyone pipettes anything.
Nguengang Wakap et al., 2020
Small cohorts add up when federated
One rare-disease registry is too small to power a study. Many registries sharing a common structure become a research surface big enough to matter, and the FDA accepts natural-history data as an external control arm.
FDA Natural History Studies guidance, 2019
Sources (4)
- Varadi M, et al. AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Research, 2024.
- Abramson J, et al. Accurate structure prediction of biomolecular interactions with AlphaFold3. Nature, 2024.
- The Nobel Prize in Chemistry 2024 (Baker; Hassabis and Jumper).
- Rare Diseases: Natural History Studies for Drug Development. FDA draft guidance, 2019.
A bridge, and a guard on the bridge
Families carry the case. Labs can test it. Between them sits a gap that wastes both. We close it, and we guard it: a family's case becomes a rigorous, computation-verified hypothesis, and only the serious ones reach a researcher's bench, never unfiltered noise. That is the whole job.