Protein-Ligand CH-π Interactions: Structural Informatics, Energy Function Development, and Docking Implementation.

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• CH-π interactions are crucial in protein-ligand binding but lack accurate models in standard computational tools.
• Existing molecular mechanics force fields and docking scoring functions do not adequately quantify CH-π contributions.

Purpose of the Study:

• To develop an empirical energy function for CH-π interactions using quantum mechanical data.
• To integrate this new term into the AutoDock Vina (AD VINA) scoring function.
• To evaluate the impact of the CH-π term on predicting protein-ligand complex structures and enhancing docking performance.

Main Methods:

• Developed an empirical energy function based on quantum mechanical data for methane-benzene interactions.
• Incorporated the CH-π term into the AutoDock Vina scoring function.
• Utilized a conformational grid search algorithm for evaluating ligand orientation.
• Tested performance against experimental protein-ligand complex structures and benchmark datasets.

Main Results:

• The CH-π term improved the prediction of preferred ligand orientations in protein-binding sites.
• Enhanced the detection of carbohydrate-binding sites exhibiting CH-π interactions.
• Demonstrated improved docking performance on the CASF-2016 benchmark and a specific carbohydrate dataset.

Conclusions:

• The developed CH-π energy term is a valuable addition to molecular docking scoring functions.
• This enhancement improves the accuracy of modeling protein-ligand interactions, particularly for carbohydrate ligands.
• The improved model aids in understanding and predicting binding modes involving CH-π interactions.