Variability in docking success rates due to dataset preparation.

Area of Science:

• Computational Chemistry
• Structural Biology
• Drug Discovery

Background:

• Molecular docking is crucial for predicting ligand-protein interactions.
• Accurate scoring functions and dataset preparation are vital for reliable docking results.
• Standardized datasets for benchmarking docking software are essential.

Purpose of the Study:

• To evaluate the performance of the Molecular Operating Environment (MOE) software with a new scoring function (GBVI/WSA dG).
• To assess the impact of dataset preparation on molecular docking success rates using the Astex dataset.
• To identify factors contributing to docking failures.

Main Methods:

• Utilized MOE software with the GBVI/WSA dG scoring function for cognate docking.
• Employed the prepared Astex dataset from the "Docking and Scoring: A Review of Docking Programs" session.
• Analyzed the accuracy of top-scoring poses against X-ray structures.

Main Results:

• MOE achieved top-scoring poses within 2 Å of X-ray structures for 80% of Astex targets.
• Poses within 2 Å were found in the top 30 for 91% of targets.
• Docking failures were often linked to the absence of bound waters in the dataset.

Conclusions:

• The MOE software demonstrates strong performance in molecular docking.
• Accurate dataset preparation, including bound waters, is critical for successful docking.
• Errors in dataset preparation can significantly impact docking success rates (over 20% fluctuation for a 0.5 kcal/mol error).