Searching for robust Turing patterns through evolution and AI - Prof. Robert Endres

We are pleased to confirm that Prof. Robert Endres will be presenting on Wednesday, January 28th 2026.

Zoom meeting: https://ed-ac-uk.zoom.us/j/85128349903

Join instructions: https://ed-ac-uk.zoom.us/meetings/85128349903/invitations?signature=SnfcsOWcBfKPWrZsqm2T0BQj93TzpYo9NOAexhWprzk

See the Join instructions or Subscribe to our mailing list for the password of the meeting.

Title: Searching for robust Turing patterns through evolution and AI

Abstract:

Turing’s reaction–diffusion theory offers a compelling explanation for biological pattern formation, yet it suffers from a persistent paradox: although Turing patterns are mathematically generic, they are experimentally fragile and difficult to realise. Why, then, do robust patterns exist in nature at all? In this talk, I argue that robustness emerges not from fine-tuned mechanisms, but from search processes acting on pattern-forming systems. I first show how evolutionary and dynamical constraints — mediated by common limit cycles — can bias how living systems explore parameter space toward rare but stable Turing regimes. I then demonstrate how neural networks can be used to learn and design robust reaction–diffusion circuits, providing new tools for synthetic biology. More broadly, the work highlights parallels between biological evolution and AI as problem-solving strategies for navigating complex, high-dimensional design spaces.