Abstract
Motility of bacteria is often recognized by the visual inspection through optical microscopy. It is known as something different from Brownian motion induced by thermal fluctuation. Diffusion coefficient is a typical measure for Brownian motion, but it tells only the overall diffusive speed. Therefore, another measure is necessary to detect the nature of random yet self-propulsive motion, especially when comparing the motions having the same diffusion coefficient. We propose a novel approach to detect and characterize motility, based on the large deviation principle in mathematics. Our proposed method does not require the chemical stimuli to induce chemo-taxis, i.e., the motility does not have to be in the specific prescribed direction. Furthermore, the method does not need the information of orientation, which is advantageous for the application to amoeboid cells as well. Thus, the method is applicable to various kinds of self-propulsive motion including unknown samples from environment.
