Markov chain models of emitter activations in single molecule localization microscopy

Area of Science:

• Biophysics
• Microscopy Techniques
• Computational Biology

Background:

• Accurate modeling of data movies is crucial for single molecule localization microscopy (SMLM).
• Existing models often lack a detailed representation of emitter activation dynamics.
• Developing robust models is essential for advancing SMLM applications.

Purpose of the Study:

• To develop a Markov chain model for emitter activation processes in SMLM.
• To analyze emitter activation under both cycled and continuous illumination schemes.
• To integrate activation models with data frame models for comprehensive data movie simulation.

Main Methods:

• Proposed a two-phase Markov chain to model emitter activation with cycled illumination.
• Derived stationary state distributions and analytical formulas for the Markov chain.
• Extended Markov chain analysis to continuous illumination scenarios.
• Simulated 2D and 3D SMLM data movies by integrating activation and data frame models.

Main Results:

• The proposed Markov chain model accurately captures emitter activation dynamics.
• Analytical formulas derived from the model precisely predict simulation outcomes.
• Simulated data movies generated by the model demonstrate high fidelity.
• The model successfully synthesizes data movies for both illumination types.

Conclusions:

• The developed Markov chain models provide a well-reasoned approach to SMLM data movie generation.
• These models are valuable for creating training data for neural networks in SMLM.
• The models facilitate the rigorous evaluation of SMLM localization algorithms.