PEM-fitter: A Coarse-Grained Method to Validate Protein Candidate Models.

Area of Science:

• Structural biology
• Computational biology
• Biophysics

Background:

• Cryo-electron microscopy (cryo-EM) generates volumetric images essential for modeling macromolecular assemblies.
• De novo protein modeling computationally determines molecular structures from these images.
• Evaluating numerous candidate conformations is a time-consuming bottleneck in protein modeling.

Purpose of the Study:

• To introduce a novel geometrical screening method for evaluating candidate protein structures.
• To reduce the number of native-like conformations requiring computationally expensive energy calculations.
• To accelerate the overall process of de novo protein structure determination.

Main Methods:

• Utilizing the skeleton derived from cryo-EM images for structural evaluation.
• Developing an initial version of a geometrical screening approach.
• Testing the method on two distinct datasets of protein structures.

Main Results:

• The method successfully identified the correct native structure from a set of diverse protein structures in the first dataset (10 proteins).
• The method effectively filtered slightly modified decoy conformations of the same protein in the second dataset (12 proteins).
• The efficiency of the geometrical screening method was demonstrated.

Conclusions:

• The geometrical screening method offers a faster alternative to traditional energy-based evaluations for candidate protein structures.
• This approach significantly reduces the computational burden in de novo protein modeling.
• The method shows promise for improving the speed and efficiency of structural biology research.