Prediction accuracy of ligand binding kinetics.

Area of Science:

• Computational chemistry
• Pharmacology
• Drug discovery

Background:

• Ligand binding kinetics, including binding (k_on) and dissociation (k_off) rates, are crucial for effective drug design.
• Drug residence time, determined by dissociation rates, often shows a stronger correlation with therapeutic efficacy than simple binding affinities.

Purpose of the Study:

• To collect and analyze computational predictions of absolute ligand binding and dissociation rates.
• To assess the prediction accuracy of these kinetic parameters by comparing them with experimental data.
• To propose quantitative criteria for evaluating the accuracy of ligand binding kinetics predictions.

Main Methods:

• Gathering computational predictions for k_on and k_off rates.
• Comparing predicted rates with experimentally determined values.
• Analyzing the distribution of prediction accuracies.

Main Results:

• The study analyzed the accuracy distributions of computational predictions for ligand binding and dissociation rates.
• Variability in prediction accuracies was observed when compared against experimental data.

Conclusions:

• Quantitative criteria for evaluating the prediction accuracy of ligand binding kinetics are proposed based on the analysis.
• These criteria can aid in selecting reliable computational methods for drug design.