Detecting directed motion and confinement in single-particle trajectories using hidden variables

Abstract

Single-particle tracking is a powerful tool for understanding protein dynamics and characterizing microenvironments. As the motion of unconstrained nanoscale particles is governed by Brownian diffusion, deviations from this behavior are biophysically insightful. The stochastic nature of particle movement and the presence of localization error, however, pose a challenge for robust classification of non-Brownian motion. Here, we present aTrack, a versatile tool for classifying and extracting key parameters of tracks in Brownian, confined or directed motion. Our tool quickly and accurately determines estimates motion parameters from individual tracks and categorizes its motion state. Further, our tool can analyze populations of tracks and determine the most likely number of motion states. We demonstrate the working range of our approach on simulated tracks and show the application of our tool for characterizing particle motion in cells and biosensing applications. Our tool is implemented as a stand-alone software package, making it simple to analyze track data.