Prediction of ADME Properties
Aixia Yan · 2018
This chapter reviews the current state-of-the-art computational models for the absorption, distribution, metabolism, and excretion (ADME) property prediction. The challenge of predicting a variety of all-important ADME properties has been adopted by several groups of researchers. The methodology and status will be illustrated in more detail with three properties, including aqueous solubility, blood-brain barrier (BBB), and human intestinal absorption (HIA). The acid dissociation constant (pKa) measures the strength of an acid, which influences solubility, permeability, and some other ADME properties such as plasma protein binding (PPB), cardiotoxicity, and metabolism of a compound. To build a structure-property relationship (SPR)/quantitative structure-property relationship (QSPR) prediction model of an ADME property consists of four steps: collection of a dataset, calculation and selection of molecular descriptors, establishment of a mathematical model between the selected descriptors and the ADME property, and model validation.