Transition-path sampling for run-and-tumble particles

We elaborate and validate a generalization of the renowned transition-path-sampling algorithm for a paradigmatic model of active particles, namely, the run-and-tumble particles. Notwithstanding the nonequilibrium character of these particles, we show how the consequent lack of the microscopical reversibility property, which is usually required by transition-path sampling, can be circumvented by identifying reasonable backward dynamics with a well-defined path-probability density. Our method is then applied to characterize the structure and kinetics of rare transition pathways undergone by run-and-tumble particles having to cross a potential barrier in order to find a target.