Classification-based motion analysis of single-molecule trajectories using DiffusionLab

Abstract

Single-particle tracking is a powerful approach to study the motion of individual molecules and particles. It can uncover heterogeneities that are invisible to ensemble techniques, which places it uniquely among techniques to study mass transport. Analysis of the trajectories obtained with single-particle tracking in inorganic porous hosts is often challenging, because trajectories are short and/or motion is heterogeneous. We present the DiffusionLab software package for motion analysis of such challenging data sets. Trajectories are first classified into populations with similar characteristics to which the motion analysis is tailored in a second step. DiffusionLab provides tools to classify trajectories based on the motion type either with machine learning or manually. It also offers quantitative mean squared displacement analysis of the trajectories. The software can compute the diffusion constant for an individual trajectory if it is sufficiently long, or the average diffusion constant for multiple shorter trajectories. We demonstrate the DiffusionLab approach via the analysis of a simulated data set with motion types frequently observed in inorganic porous hosts, such as zeolites. The software package with graphical user interface and its documentation are freely available.