A collaborative framework for 3D alignment and classification of heterogeneous subvolumes in cryo-electron tomography

Area of Science:

• Structural Biology
• Biophysics
• Microscopy

Background:

• Low electron doses in cryo-electron tomography limit resolution.
• Sub-volume averaging is crucial for enhancing signal but challenging with multiple species.
• Conformational heterogeneity complicates alignment in biological specimens.

Purpose of the Study:

• To develop a novel method for conformational separation and alignment of sub-volumes in cryo-electron tomography.
• To overcome the "curse of dimensionality" in standard pairwise comparisons for complex biological samples.
• To enable accurate identification of viral particles based on subtle structural differences.

Main Methods:

• A collaborative approach utilizing the nuclear norm as a similarity measure for sub-volume alignment.
• Exploiting symmetry early in the processing pipeline to improve efficiency.
• Validation using mixtures of simian immunodeficiency viruses (SIV mac239 and SIV CP-MAC).

Main Results:

• The nuclear norm-based collaborative alignment method effectively separates and aligns structurally distinct sub-volumes.
• Subtle conformational differences in envelope glycoproteins were successfully resolved.
• Accurate genetic identity assignment of individual virus particles was achieved solely from surface glycoprotein structures.

Conclusions:

• The developed method offers a powerful solution for analyzing heterogeneous biological samples in cryo-electron tomography.
• This approach enhances the ability to discern fine structural details and differentiate between closely related biological entities.
• It opens new avenues for high-resolution structural analysis of complex biological systems using cryo-electron tomography.