Tracking variable number of multiple subcellular structures in 3D.

Area of Science:

• Cell Biology
• Biophysics
• Computational Biology

Background:

• Advancements in electronic imaging and green fluorescent protein (GFP) tagging necessitate automated tools for live cell dynamics analysis.
• Studying subcellular structures requires methods capable of handling dynamic changes and multiple objects.
• Existing methods may lack the ability to track a variable number of structures or complex motion patterns.

Purpose of the Study:

• To develop an automatic quantitative data analysis tool for live cell dynamics at the subcellular level.
• To propose a novel sequential Monte Carlo (SMC) method for tracking multiple 3D subcellular structures with variable numbers.
• To address challenges in tracking structures that appear or disappear during live cell imaging.

Main Methods:

• Representation of multiple subcellular structures using a joint state.
• Efficient sampling of the dimension-changing joint state via reverse jump Markov chain Monte Carlo (RJMCMC).
• Implementation of RJMCMC with specific moves: update, identity switch, disappearing, and appearing moves.

Main Results:

• The proposed SMC method successfully tracks multiple 3D subcellular structures.
• The method demonstrates efficacy in handling structures with varying motion modalities, including appearance and disappearance.
• Experimental results validate the capability of the RJMCMC-based approach for dynamic tracking.

Conclusions:

• The developed SMC method provides a robust solution for quantitative analysis of live cell dynamics at the subcellular level.
• This approach is critical for advancing the study of complex biological processes involving dynamic subcellular structures.
• The method's ability to handle variable numbers and changing dimensions of structures enhances live cell imaging data analysis.