Normal mode analysis with molecular geometry restraints: bridging molecular mechanics and elastic models

Area of Science:

• Structural Biology
• Computational Chemistry
• Biophysics

Background:

• Normal mode analysis (NMA) is crucial for understanding protein dynamics.
• Existing NMA methods, like molecular mechanics (MM) and elastic network models (ENM), have limitations.
• Bridging MM and ENM representations can enhance NMA accuracy.

Purpose of the Study:

• To introduce a novel method for NMA of all-atom structures using molecular geometry restraints (MGR).
• To evaluate MGR's performance against established NMA techniques.
• To demonstrate MGR's applicability to diverse structures.

Main Methods:

• Developed MGR potential with short-range (geometry-based) and long-range (ENM-like) terms.
• Used a single force constant per term, fitted against known structures.
• Applied MGR to protein and non-protein datasets.

Main Results:

• MGR produced low-frequency eigenvectors closer to all-atom force-field methods than conventional ENMs.
• The "tip effect" artifact in ENMs was significantly reduced by MGR.
• MGR demonstrated applicability to structures with imperfections, like missing atoms.

Conclusions:

• Molecular geometry is critical for low-frequency motions, alongside molecular shape.
• MGR effectively bridges MM and ENM approaches in NMA.
• MGR offers a robust and versatile tool for analyzing molecular dynamics, especially for low-resolution data.