RegiSTORM: channel registration for multi-color stochastic optical reconstruction microscopy

Area of Science:

• Microscopy
• Biophysics
• Computational Biology

Background:

• Stochastic optical reconstruction microscopy (STORM) enables super-resolution imaging but suffers from sample drift during long acquisitions.
• Sequential channel acquisition in multi-color STORM leads to interchannel misalignment, hindering the study of biological interactions.
• Existing drift correction methods do not fully address interchannel misalignment in multi-color STORM.

Purpose of the Study:

• To develop and validate a software solution, RegiSTORM, for accurate channel registration in STORM microscopy.
• To overcome the limitations of interchannel drift in multi-color STORM imaging.
• To enhance the characterization of biological interactions using precise multi-color STORM data.

Main Methods:

• RegiSTORM utilizes fiducial markers within STORM localization data as landmarks for precise channel registration.
• Fiducials are identified based on their non-blinking characteristics in the STORM data.
• The software was tested on fiducial-only recordings, cell samples, and lipid nanoparticles.

Main Results:

• RegiSTORM significantly reduced target registration error across all tested channel combinations in fiducial recordings.
• The software demonstrated effective channel registration in multi-stained cell samples (tubulin).
• RegiSTORM successfully registered two-color STORM recordings of lipid nanoparticles, even without fiducials, showing broad applicability.

Conclusions:

• RegiSTORM accurately registers multiple STORM channels, effectively mitigating interchannel misalignment.
• The software is freely available as open-source, facilitating its adoption in research.
• RegiSTORM enhances the reliability of multi-color STORM for complex biological studies.