Solubility is a widely measured physicochemical property and a highly sought-after descriptor in the evaluation of molecules for drug design and development, medicinal chemistry, and agrochemical sciences. Consequently, a wide range of in-silico models based on deductive (QM, MM, MD) and inductive (QSAR/QSPR and ML) reasoning have been created. However, the General Solubility Equation (GSE) stands out as a simple, thermodynamically derived equation to predict intrinsic solubility (log S₀, in mol/L) of molecules from their melting point and n-octanol/water partition coefficient(log S₀ = 0.5 - 0.01(mp - 25) - log P_o/w). Despite its widespread use, there has been no comprehensive evaluation to identify the molecular characteristics that lead to accurate predictions with this equation. Therefore, a large-scale evaluation of the GSE was conducted using an extensive database of fully experimental log S₀, melting point, and log P_o/w data for 1354 mostly small molecules. A data analysis was performed using chemoinformatic descriptors to determine the most important factors for the GSE to yield a fit within ±1 log S units. This information has been used to create GSESolver: a supervised Machine Learning classification model that, based on the selected descriptors, can predict whether the GSE will perform accurately solely from its SMILES string. GSESolver was developed under the FAIR principles (Findable, Accessible, Interoperable, and Reusable) in a Google Colab notebook to ensure ease of use. Finally, the utility of the Colab was demonstrated in medicinal chemistry applications as a tool for the in-silico prediction of the Maximum Absorbable Dose (MAD) of potential drugs, and in a rational agrochemical design application where solubility is a key factor. This tool can be used by anyone in the scientific community to quickly assess the solubility of their molecules.

Jose Medina-Franco

Universidad Nacional Autonoma de Mexico

Appearances