Assessing the ligand native-like pose using a quantum mechanical-derived hydropathic score for protein-ligand...

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Accurate prediction of ligand-protein binding poses is essential for drug development.
• Identifying the near-native binding pose in docking simulations remains a significant challenge.

Purpose of the Study:

• To evaluate the effectiveness of hydropathicity-based scoring functions in identifying near-native ligand-protein binding poses.
• To assess the performance of 3D hydropathicity atomic descriptors in distinguishing bioactive from decoy poses.

Main Methods:

• Utilized a benchmarking dataset of 1000 ligand-protein complexes.
• Assessed 3D hydropathicity atomic descriptors derived from empirical and quantum mechanical models.
• Incorporated the influence of conformational stress into the analysis.

Main Results:

• Hydropathicity-based scoring functions demonstrated significant potential in identifying near-native poses.
• Achieved a predictive accuracy of approximately 90% when considering conformational stress.
• Highlighted the importance of nonpolar/polar chemical features for pose discrimination.

Conclusions:

• Simple hydropathicity scoring, when accounting for conformational stress, can effectively predict near-native ligand-protein binding poses.
• Understanding nonpolar and polar interactions is crucial for successful drug-target recognition and optimization.
• This approach aids in guiding hit-to-lead optimization in drug discovery.