Events

Date: 8 October 2026   Time: 13:00 - 14:00     

Location: MB-503

A living cell controls itself by turning a few knobs. Raising a concentration speeds a binding step, a catalyst lowers a barrier, a stabilized conformation slows every process that leaves it. Underneath the chemistry each of these is the same operation, a change in one rate of a Markov jump process, so the biological question becomes a general one. What happens to such a process when one of its knobs is turned? Classical response theory answers this near equilibrium and for small perturbations, and a cell honours neither restriction. I will show how much survives anyway: the whole nonlinear response curve follows from a linear measurement and one further number; a nonlinear response can never much exceed the linear response it grew from; and no change can be resolved more sharply than the observable's own fluctuations permit. Every response curve of a network then proves to be the same shifted sigmoid, with one center shared across all observables, and completion times, which look like a separate problem, turn out to be stationary currents in disguise, obeying a control budget in which acceleration anywhere is paid for somewhere else.