GalaxyDock BP2 score: a hybrid scoring function for accurate protein-ligand docking

Area of Science:

• Computational biology
• Structural bioinformatics
• Drug discovery

Background:

• Protein-ligand docking is crucial for understanding protein function and designing novel molecules.
• Accurate identification of ligand binding poses is essential for successful docking.
• Existing scoring functions often focus on binding affinity, not directly on pose prediction.

Purpose of the Study:

• To develop a novel docking scoring function, GalaxyDock BP2 Score, specifically trained for binding pose prediction.
• To create a hybrid scoring function combining physics-based, empirical, and knowledge-based approaches.
• To evaluate the performance of the new scoring function against existing methods.

Main Methods:

• Developed GalaxyDock BP2 Score, a hybrid scoring function trained directly on binding pose data.
• Integrated the scoring function into the GalaxyDock2 protein-ligand docking program.
• Tested performance using decoy pose discrimination and benchmark docking datasets (Astex diverse, Cross2009, Astex non-native).

Main Results:

• GalaxyDock BP2 Score demonstrated significant improvement in decoy pose discrimination compared to existing functions.
• The GalaxyDock2 program, utilizing the new score, outperformed other state-of-the-art docking programs on multiple benchmark sets.
• The scoring function effectively balances physics-based, empirical, and knowledge-based components.

Conclusions:

• The novel GalaxyDock BP2 Score enhances the accuracy of protein-ligand docking pose prediction.
• This advancement offers improved atomic-level understanding of protein-ligand interactions.
• GalaxyDock BP2 Score and GalaxyDock2 provide a powerful, freely available tool for computational drug discovery.