Single particle maximum likelihood reconstruction from superresolution microscopy images

Area of Science:

• Biophysics
• Microscopy
• Structural Biology

Background:

• Superresolution microscopy achieves nanoscale precision, but structural modeling of small objects is challenging due to localization uncertainty.
• Existing methods often require particle averaging, limiting single-particle analysis.

Purpose of the Study:

• To develop a method for accurate structural modeling of small biological objects using superresolution microscopy data at the single-particle level.
• To overcome the limitations of localization uncertainty in reconstructing structural models from superresolution images.

Main Methods:

• Implemented a Maximum Likelihood Reconstruction (MLR) method tailored for the stochastic nature of superresolution imaging.
• Applied MLR to both simulated and experimental photoactivated localization microscopy (PALM) data of Human Immunodeficiency Virus type 1 (HIV-1) immature virus-like particles.
• Validated the method for single-particle analysis without the need for particle averaging.

Main Results:

• MLR enabled precise measurement of individual virus radii with nanometer accuracy.
• The method confirmed incomplete closure of the viral protein lattice in HIV-1 particles.
• Quantitative results align with previous cryoelectron microscopy findings.

Conclusions:

• MLR provides a robust framework for determining structural parameters from superresolution data at the single-particle level.
• This approach is particularly advantageous for analyzing heterogeneous biological structures.
• The study establishes a new standard for structural analysis in superresolution microscopy.