Single-particle tracking (SPT) enables the investigation of biomolecular dynamics at a high temporal and spatial resolution in living cells, and the analysis of these SPT datasets can reveal biochemical interactions and mechanisms. We have developed a new SPT analysis framework, NOBIAS, which applies nonparametric Bayesian statistics and deep learning approaches to thoroughly analyze SPT datasets. We utilize a Hierarchical Dirichlet process Hidden Markov Model (HDP-HMM) to infer the number of diffusive states and the associated dynamics, populations and step labels for each diffusive state, then we apply a Recurrent Neural Network (RNN) to classify the diffusion type of each diffusive state. We further validate the performance of NOBIAS with simulated tracks and the quantify diffusion of single outer-membrane proteins in Bacteroides thetaiotaomicron.
