Next Motion AI

THE PLATFORM

Building a learning platform for biology.

We’re building a reinforcement learning platform to help explore large biological search spaces through sequential choices and feedback. Our peptide–MHC binding project is the first place we’re testing that approach.

THE BIOLOGICAL QUESTION

What can the feedback tell us?

The platform starts with a biological question, an eligible candidate space, and a clear account of what the available feedback means. Our first project, peptide–MHC binding, gives us a concrete way to develop and test that loop. Computational scores can guide exploration, but they do not establish what happens in a cell.

THE LEARNING LOOP

How we test adaptive search.

We’re developing a process for choosing candidates, reading feedback, and updating priorities. Each project must test whether those adaptive choices improve on simpler strategies.

Conceptual feedback loop: define the question, consider eligible candidates, choose one, read available feedback, then update priorities for the next choice.
Define what the feedback can tell us before searching. Then compare adaptive choices with a fixed ranking under the same candidate set and evaluation budget. Computational feedback guides priorities; biological outcomes require separate evidence.

EVIDENCE AND ITS LIMITS

A promising score is a starting point.

Computational predictions can guide priorities, but they are not measurements of biological outcomes. Fair comparisons must show whether adaptation helps, and biological claims require the right experimental evidence.

Explore our first project