Potential functions for hydrogen bonds in protein structure prediction and design

Area of Science:

• Biomolecular Modeling
• Computational Chemistry
• Structural Biology

Background:

• Hydrogen bonds significantly influence the stability of biological macromolecules.
• Traditional models often simplify hydrogen bonds as electrostatic dipole-dipole interactions.
• The true nature of hydrogen bonding involves complex quantum mechanical effects.

Purpose of the Study:

• To explore the fundamental physical nature of hydrogen bond formation.
• To evaluate different modeling approaches for hydrogen bonds in biomolecular systems.
• To assess the impact of accurate hydrogen bond descriptions on computational biology tasks.

Main Methods:

• Review of existing knowledge-based and empirical molecular mechanics models.
• Application of quantum mechanics-based electronic structure calculations, including ab initio methods.
• Analysis of hydrogen bond geometries from structural databases of proteins and small molecules.

Main Results:

• Hydrogen bonding is fundamentally a quantum mechanical phenomenon with electrostatic, covalent, polarization, and charge transfer contributions.
• Ab initio calculations accurately predict hydrogen bonding energy landscapes.
• Orientation-dependent hydrogen bonding potentials enhance protein structure prediction, refinement, and docking.

Conclusions:

• A comprehensive understanding of hydrogen bonds, incorporating both electrostatic and covalent aspects, is vital for accurate biomolecular modeling.
• Advanced computational methods, particularly ab initio calculations, provide accurate descriptions of hydrogen bonding.
• Improved hydrogen bond potentials significantly advance protein structure prediction, protein-protein interactions, and protein design.