Three-dimensional localization refinement and motion model parameter estimation for confined single particle tracking...

Area of Science:

• Biophysics
• Computational Biology
• Cellular Dynamics

Background:

• Confined diffusion models biological macromolecule motion in cellular environments.
• Single particle tracking (SPT) is crucial for studying these dynamics.
• Analyzing SPT data is challenging due to complex motion and camera noise.

Purpose of the Study:

• To enhance the Sequential Monte Carlo - Expectation Maximization (SMC-EM) technique for analyzing confined diffusion.
• To adapt SMC-EM for the double-helix point spread function (DH-PSF) and scientific CMOS (sCMOS) cameras.
• To evaluate SMC-EM's performance against standard 'localize-then-estimate' methods, particularly in low signal conditions.

Main Methods:

• Extended SMC-EM to incorporate DH-PSF for 3D particle localization from 2D images.
• Accounted for pixel-dependent noise characteristic of sCMOS cameras.
• Compared SMC-EM with Gaussian fitting, Maximum Likelihood Estimator (MLE), and Mean Squared Displacement (MSD) fitting via simulations.

Main Results:

• SMC-EM outperformed standard methods in low signal-to-background regimes.
• SMC-EM and MLE-based methods showed comparable performance at higher signal levels.
• SMC-EM demonstrated superiority when analyzing motion models dominated by nonlinearities at smaller confinement lengths.

Conclusions:

• The enhanced SMC-EM method provides a robust approach for analyzing confined diffusion from SPT data.
• SMC-EM offers significant advantages over traditional methods, especially under challenging low signal and nonlinear motion conditions.
• This technique advances the study of macromolecule dynamics within crowded cellular environments.