Predicting Ligand Binding Modes by Scaffold-Guided Structure Refinement

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Structure-based drug design requires accurate ligand binding poses, which are often experimentally unavailable.
• Current binding mode prediction methods, including co-folding, frequently lack sufficient accuracy.
• Identifying conserved binding modes based on shared molecular scaffolds in ligand series is a promising avenue.

Purpose of the Study:

• To develop a novel method for refining ligand binding poses and improving structure-based drug design.
• To leverage information from known binders within a ligand series to predict binding modes more accurately.
• To address the limitations of existing binding mode prediction tools by incorporating scaffold information.

Main Methods:

• Scaffold-Guided Structure Refinement method developed.
• Utilizes molecular docking against diverse target conformations.
• Identifies optimal conformations allowing consistent scaffold placement, favorable interactions, and low ligand strain.

Main Results:

• Successfully optimized models from various initial sources, including inaccurate co-folding models.
• Accurately reconstructed near-native ligand binding modes in three case studies.
• Identified critical induced fit effects, achieving scaffold root-mean-square deviation (RMSD) values of at most 2.2 Å.

Conclusions:

• Scaffold-Guided Structure Refinement is effective for optimizing ligand binding poses.
• The method enhances the accuracy of structure-based drug design by leveraging scaffold conservation.
• This approach offers a valuable tool for improving drug discovery pipelines.