Rapid flexible docking using a stochastic rotamer library of ligands

Area of Science:

• Computational Biology
• Molecular Modeling
• Drug Discovery

Background:

• Existing flexible docking methods inadequately capture ligand and receptor conformational changes.
• Simultaneous modeling of both ligand and receptor flexibility is crucial for accurate docking predictions.

Purpose of the Study:

• To introduce MedusaDock, a novel flexible docking approach.
• To evaluate MedusaDock's efficiency and accuracy in self- and cross-docking simulations.
• To assess MedusaDock's performance in virtual screening of kinase targets.

Main Methods:

• Developed MedusaDock, a flexible docking approach using discrete rotamers.
• Implemented an algorithm for on-the-fly ligand rotamer library generation.
• Benchmarked MedusaDock against rigid-receptor methods in self-docking, cross-docking, and virtual screening.

Main Results:

• MedusaDock demonstrated rapid sampling efficiency and high prediction accuracy.
• Significant improvements in virtual screening enrichment were observed for flexible kinase targets.
• Cross-docking performance indicated strong predictive power for virtual screening applications.

Conclusions:

• Simultaneous modeling of ligand and receptor flexibility is essential for effective computational docking.
• MedusaDock offers a promising tool for virtual screening and drug discovery.
• The approach advances flexible docking methodologies by integrating ligand and receptor conformational dynamics.