Title:Generalized Boltzmann Distribution for Systems out of Equilibrium

Abstract:The theory of statistical mechanics pioneered by Boltzmann has enabled predictive understanding of a wide range of physical systems that are at thermal equilibrium. Understanding how living systems function at the microscopic scale requires an equally powerful theory of far-from- equilibrium behavior. Such a theory has so far been lacking. Here, I generalize the Boltzmann distribution to enable calculation of microstate probabilities and fluxes for systems that are driven, time-dependent, and/or are not at steady state; at equilibrium, the equation reduces to the Boltzmann distribution. The derived equation is an isomorphic mapping between any network of microstates to an equivalent electrical circuit with resistors and voltage- dependent voltage sources. This mapping allows for the application of powerful theorems from circuit theory to nonequilibrium statistical mechanics, from which they take on new meaning, such as maximum efficiency and reciprocal relations arbitrarily far from equilibrium. These relations are obeyed by driven steady state systems in general, and therefore govern biological performance, response, and evolvability. The resistors can be systematically coarse-grained to simplify the analysis of complex systems, as well as to deduce microscopic thermodynamic information like free energy barriers from measurements of macroscopic observables like dissipation rate or motility. As concrete examples, the single-cycle microstate topology with one and two driving sources, which describe protein catalysis and activation, respectively, are solved analytically. In these cases, being driven far from equilibrium allows function to be optimized independently of stability.