Single-particle tracking (SPT) is a central technique in biophysics for studying molecular motion, but it remains highly challenging due to motion blur, blinking, out-of-focus movement, and ambiguities in linking trajectories. Many existing SPT methods also struggle under conditions of high particle density and low signal-to-noise ratios, limiting their applicability to complex biological data sets. A Bayesian alternative, Reversible Jump Markov Chain Monte Carlo (RJMCMC), directly addresses these challenges by performing localization and trajectory inference simultaneously. Unlike fixed-dimensional methods, RJMCMC enables dynamic model selection by allowing the dimensionality of the parameter space to change between iterations. This feature naturally accounts for photophysical events such as blinking and emitter appearance or disappearance. The resulting output is a statistically principled posterior distribution over possible trajectories, with the maximum a posteriori solution representing the most probable tracks. We present RJTrack.jl, an open-source Julia package implementing an RJMCMC framework for SPT. Julia was chosen for its unique combination of high-level usability and low-level computational efficiency. RJTrack leverages Julia’s speed to efficiently explore high-dimensional posterior spaces while remaining flexible for method development. Our package implements a suite of reversible moves including birth/death, grow/shrink, split/merge, swap, and parameter updates that allow efficient navigation of trajectory space. The RJMCMC formalism provides a rigorous statistical framework for uncertainty quantification in trajectory reconstruction.
Habibi et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: