Progress in computational methods for the prediction of ADMET properties

Area of Science:

• Pharmacokinetics and Drug Metabolism
• Computational Chemistry and Cheminformatics
• Toxicology and Safety Assessment

Background:

• Computational techniques are increasingly vital for predicting Absorption, Distribution, Metabolism, Elimination, and Toxicity (ADMET) properties in drug discovery.
• Existing models show success on current datasets but face limitations due to data quality and quantity.

Purpose of the Study:

• To review recent advancements in computational methods for predicting ADMET properties.
• To highlight the application of these techniques in areas like intestinal permeability and blood-brain barrier penetration.
• To identify key challenges and future directions in the field.

Main Methods:

• Literature review of computational techniques for ADMET prediction.
• Analysis of methods for predicting properties such as solubility, metabolism, and toxicity.
• Examination of models for transport processes like active transport/efflux and barrier penetration.

Main Results:

• Significant progress has been made in developing computational models for various ADMET endpoints.
• Current models demonstrate utility on existing experimental datasets.
• The field faces a critical need for larger, high-quality experimental datasets.

Conclusions:

• Further development of computational ADMET prediction requires substantial investment in generating high-quality experimental data.
• Improved datasets will form a sound basis for building more robust and accurate predictive models.
• Collaboration between computational scientists and experimentalists is essential for advancing drug safety assessment.