Related Experiment Videos
ConsDock: A new program for the consensus analysis of protein-ligand interactions
1Bioinformatic Group, Laboratoire de Pharmacochimie de la Communication Cellulaire, UMR CNRS 7081, Illkirch, France.
Proteins
|May 10, 2002
Summary
A new consensus docking approach (ConsDock) combining three tools significantly improves the accuracy of predicting protein-ligand interactions. This method enhances lead optimization by reliably ranking the most likely molecular poses.
Area of Science:
- Computational chemistry
- Drug discovery
- Structural biology
Background:
- Protein-based virtual screening aids in identifying novel molecules for macromolecular targets.
- Limitations in docking and scoring accuracy hinder its application in lead optimization.
Purpose of the Study:
- To develop and evaluate a consensus docking approach (ConsDock) for improved accuracy in predicting protein-ligand interactions.
- To enhance the reliability of virtual screening for lead optimization.
Main Methods:
- ConsDock integrates three widely used docking tools: Dock, FlexX, and Gold.
- A four-step sequential consensus analysis is applied to poses generated by multiple docking programs.
- Hierarchical clustering, consensus pair definition, and mean structure clustering are employed for pose ranking.
Main Results:
- ConsDock significantly outperforms single docking methods in ranking accuracy for protein-ligand poses.
- In 60% of cases, ConsDock identified a correct pose within 2 Å RMSD of the experimental structure as the top solution.
- The approach demonstrated superior performance on a test set of 100 Protein Data Bank complexes.
Conclusions:
- ConsDock serves as an effective postprocessing filter for single or multiple docking programs.
- This method can prioritize three-dimensional guided lead optimization by identifying the most probable docking solutions.
- The consensus strategy enhances the precision of virtual screening for drug discovery.