Aromatic interactions at the ligand-protein interface: Implications for the development of docking scoring functions

Area of Science:

• Computational chemistry
• Structural biology
• Medicinal chemistry

Background:

• Non-covalent interactions, particularly aromatic stacking, are crucial for ligand-protein binding in drug development.
• Existing computational methods for modeling aromatic interactions in docking programs have limitations in accurately representing their geometry.

Purpose of the Study:

• To conduct a large-scale statistical analysis of aromatic contacts in ligand-protein complexes.
• To evaluate the geometric accuracy of π-π stacking in current molecular docking programs.
• To lay the groundwork for developing improved computational potentials for aromatic interactions.

Main Methods:

• Utilized a two-parameter geometric model for statistical analysis.
• Analyzed both experimental and computer-generated structures of ligand-protein complexes.
• Compared interfacial π-π stacking geometries with data from simple systems like the benzene dimer.

Main Results:

• The geometry of π-π stacking in crystal structures aligns with experimental data for simple systems.
• Current docking programs, while generally effective, could refine the geometric modeling of π-π contacts in high-scoring conformations.
• A comprehensive dataset and analysis script were developed.

Conclusions:

• A detailed understanding of aromatic geometries at ligand-protein interfaces is essential for advancing rational drug design.
• Improved, type-specific statistical potentials can enhance the accuracy of molecular docking simulations.
• The findings provide a foundation for developing more precise computational tools for drug discovery.