Comments on prediction of the aqueous solubility using the general solubility equation (GSE) versus a genetic algorithm and a support vector machine model

Abstract

The general solubility equation (GSE) is the state-of-the-art method for estimating the aqueous solubilities of organic compounds. It is an extremely simple equation that expresses aqueous solubility as a function of only two inputs: the octanol–water partition coefficient calculated by readily available softwares like clogP and ACD/logP, and the commonly known melting point of the solute. Recently, Bahadori et al. proposed that their genetic algorithm support vector machine is a “better” predictor. This paper compares the use of the of Bahadori et al. model for the prediction of aqueous solubility to the existing GSE model.

Keywords:

• Aqueous solubility
• support vector machine
• quantitative structure–property relationship
• Genetic algorithm

Disclosure statement

No potential conflict of interest was reported by the authors.