BEYOND BELL'S LAW: MODEL-FREE TRANSITION RATE ESTIMATION FROM NONEQUILIBRIUM TRAJECTORIES*

Nonequilibrium systems regulate the rate at which they transition from one long-lived functional state to another by channeling chemical and mechanical energy from external forces and noisy environments. Here we discuss a concrete tradeoff, imposed by statistical physics, between the extent to which such transitions can be sped-up and the work required to do so [1]. We use a model cooperatively (un)folding protein to illustrate how such a guiding principle can be put to use inferring transition rates from nonequilibrium force spectroscopy experiments and molecular simulation.  Our general, model independent framework reduces to Bell’s phenomenological rate law in the appropriate limit, and goes beyond it, allowing for arbitrary time-dependent driving forces without the assumption of quasi-equilibrium. Finally, we show how this dissipative rate bound can be used as a variational principle to iteratively design optimal forcing protocols to minimize artifacts from experimental apparatus, and better recapitulate unbiased transition paths from steered molecular dynamics [2].