This event has concluded
Verifiable Private Genomic Computation Stream
July 20–21, 2026 · 8:00–12:00 ET both days
A two-day blue-sky exploration of what zero-knowledge cryptography could unlock for genomic data sharing.
Challenges in existing infrastructure limit the growth of genomic data sharing. Adding institutions is still cumbersome. Federated systems stay separate because neither side wants to cede authority. Auth often sits with a broker. Patients sign a one-time release and rarely see how their data is used afterward.
Federated networks, trusted research environments, and TEEs already help keep data in place — and they can return narrow results from local computation. What they don't dissolve is the cost of growing and unifying the network: committees, DUAs, and audits for each relationship; single roots of trust; brokered rails; and a consent conversation that ends at the signing of a release form.
What if trust composed across accredited institutions, overlapping networks could form without one federation ceding control, and no one owned and operated the auth stack? Researchers broadcast a computation under policy; participants run it locally and return results with cryptographic proofs. Patients see the inquiry first — the program, the intended use, the data that will be revealed, and who it will be sent to — so consent is informed and ongoing.
You're here because you know the pain — the bureaucracy, the dead ends, the patient pushback, the agreement that never gets signed. Zero-knowledge cryptography could be a powerful tool for addressing those issues. Over two mornings we'll try to do four things:
What zero-knowledge unlocks for genomic data sharing — proving a computation ran correctly without revealing the underlying data.
Walk a concrete ZEENOME user journey and surface the assumptions that don't hold up in the real world.
What classes of problems is this especially well suited for, and what is it not suitable for?
Envision the next generation of genomics data-sharing — how would that change your work?
Two mornings, one question: what could privacy-preserving verifiable genomic computation actually unlock — and where does it break? Thank you to everyone who explored it with us.