Solving the α-helix correspondence problem at medium-resolution Cryo-EM maps through modeling and 3D matching

Area of Science:

• Biophysics
• Structural Biology
• Computational Biology

Background:

• Cryo-electron microscopy (cryo-EM) is advancing macromolecular structure determination.
• Medium-resolution cryo-EM maps (4-10 Å) reveal secondary structures but lack direct atomic detail.
• Matching detected secondary structures to protein sequences remains a challenge.

Purpose of the Study:

• To develop an automated framework for solving the alpha-helix correspondence problem in 3D cryo-EM density maps.
• To enable accurate atomic model building from medium-resolution cryo-EM data.

Main Methods:

• Modeling protein sequences and transforming the alpha-helix correspondence problem into a weighted bipartite graph matching problem.
• Utilizing a novel correlation-based scoring function for graph weighting.
• Employing Greedy and Improved Greedy local optimization algorithms for correspondence finding.

Main Results:

• The automated method achieved 86.25% accuracy with an 11.3% error rate on a dataset of 20 cryo-EM maps.
• The algorithm demonstrated high efficiency and robustness.
• The method operates independently of cryo-EM skeletonization.

Conclusions:

• The proposed automated framework effectively solves the alpha-helix correspondence problem in cryo-EM density maps.
• This method enhances the interpretation of medium-resolution cryo-EM data, facilitating atomic model construction.
• The approach offers a fast, accurate, and robust solution for structural biologists.