GeauxDock: A novel approach for mixed-resolution ligand docking using a descriptor-based force field

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• Molecular docking is crucial for computer-aided drug discovery.
• Existing methods face challenges in accurately predicting protein-ligand binding poses.

Purpose of the Study:

• Introduce GeauxDock, a new molecular docking approach.
• Enhance the accuracy of identifying native-like binding conformations.
• Improve efficiency in drug discovery pipelines.

Main Methods:

• Developed a descriptor-based scoring function integrating evolutionary and physics-based energy terms.
• Utilized a mixed-resolution molecular representation for protein-ligand complexes.
• Implemented an efficient Monte Carlo sampling protocol.
• Optimized the scoring function to correlate pseudoenergy with native-likeness.

Main Results:

• GeauxDock demonstrated a strong capacity to identify near-native conformations.
• Achieved an area under the receiver operating characteristic curve of 0.85 in benchmarking.
• Maintained accuracy at lower homology levels by balancing evolutionary and physics-based contributions.

Conclusions:

• GeauxDock offers a robust and accurate method for molecular docking.
• The approach enhances the identification of biologically relevant binding poses.
• GeauxDock is a valuable tool for accelerating computer-aided drug discovery.