Ghostbuster: A phase retrieval diffraction tomography algorithm for cryo-EM.

Area of Science:

• Structural Biology
• Biophysics
• Electron Microscopy

Background:

• Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the 3D structure of biological molecules.
• Existing reconstruction methods often rely on approximations, such as the projection approximation, which can introduce artifacts, especially for thicker specimens.
• Reconstruction artifacts, termed 'ghosts,' have been observed even in samples previously considered thin.

Purpose of the Study:

• To identify and characterize the origin of ghost artifacts in cryo-EM reconstructions.
• To develop a computational method for eliminating these artifacts and improving image quality.
• To demonstrate the effectiveness of the proposed method compared to existing state-of-the-art software.

Main Methods:

• The study extends Ewald sphere curvature correction beyond the projection approximation.
• A novel algorithm, Ghostbuster, is introduced, which retrieves lost phases of electron exit waves.
• The method incorporates first Born approximation scattering within the particle to correct for artifacts.

Main Results:

• Ghost artifacts can appear in cryo-EM reconstructions of particles previously assumed to be thin.
• The Ghostbuster algorithm effectively eliminates these ghost artifacts by retrieving lost phases and accounting for scattering.
• Simulations show significant improvements in reconstruction quality compared to current state-of-the-art software.

Conclusions:

• Ghost artifacts are a significant issue in cryo-EM reconstructions, affecting even thin particles.
• The Ghostbuster algorithm provides an effective solution for removing these artifacts without modifying data acquisition.
• This method has the potential to enhance the resolution and reliability of cryo-EM structural studies.