CHAPTER 22. Machine-learning Routes to Dynamics, Thermodynamics and Work Cycles of Biological Nanomachines

Abstract

The behavior of a wide variety of biological nanomachines in function depends on the energy landscapes explored during their work cycle, with states corresponding to high energy barriers playing a critical, rate-limiting role. The application of machine-learning to the rapidly growing body of experimental data from X-ray free electron lasers and cryogenic electron microscopy is providing deep insights into energy landscapes, and the conformational changes involved in nanomachine work cycles. Here, I review some recent applications of machine-learning to such data, and assess future prospects.