Related Experiment Videos
Comparative evaluation of eight docking tools for docking and virtual screening accuracy.
Esther Kellenberger1, Jordi Rodrigo, Pascal Muller
1Bioinformatics Group, Laboratoire de Pharmacochimie de la Communication Cellulaire, CNRS UMR7081 Illkirch, France.
Proteins
|September 2, 2004
Summary
This study compared eight molecular docking programs for accuracy in predicting ligand poses and screening enzyme inhibitors. Top-performing programs like GLIDE, GOLD, and SURFLEX excelled in both tasks, identifying key factors influencing docking success.
Area of Science:
- Computational chemistry
- Structural biology
- Drug discovery
Background:
- Molecular docking is crucial for virtual screening and drug design.
- Accurate prediction of ligand binding poses is essential for reliable virtual screening.
- Evaluating and comparing docking software performance is vital for optimizing drug discovery pipelines.
Purpose of the Study:
- To compare the performance of eight docking programs (DOCK, FLEXX, FRED, GLIDE, GOLD, SLIDE, SURFLEX, QXP).
- To assess their ability to recover experimental X-ray poses for 100 small-molecule ligands.
- To evaluate their capacity to distinguish known enzyme inhibitors from non-inhibitors in virtual screening.
Main Methods:
- Comparative analysis of eight docking programs.
- Evaluation of pose prediction accuracy using 100 small-molecule ligands with known X-ray structures.
- Virtual screening of thymidine kinase inhibitors against a background of random molecules.
Main Results:
- Docking accuracy and virtual screening success were correlated.
- GLIDE, GOLD, and SURFLEX demonstrated the highest docking accuracy.
- These top programs were also most effective in ranking known inhibitors during virtual screening.
- Physicochemical properties of ligands and binding sites influencing accuracy were identified.
Conclusions:
- The study highlights the superior performance of GLIDE, GOLD, and SURFLEX in both pose prediction and virtual screening.
- Successful virtual screening relies on accurate pose prediction.
- Understanding ligand-protein physicochemical interactions is key to improving docking and scoring accuracy.