Fast Prediction of the Equivalent Alkane Carbon Number Using Graph Machines and Neural Networks

Abstract

The hydrophobicity of oils is a key parameter to design surfactant/oil/water (SOW) macro-, micro-, or nanodispersed systems with the desired features. This essential physicochemical characteristic is quantitatively expressed by the equivalent alkane carbon number (EACN) whose experimental determination is tedious since it requires knowledge of the phase behavior of the SOW systems at different temperatures and for different surfactant concentrations. In this work, two mathematical models are proposed for the rapid prediction of the EACN of oils. They have been designed using artificial intelligence (machine-learning) methods, namely, neural networks (NN) and graph machines (GM). While the GM model is implemented from the SMILES codes of a 111-molecule training set of known EACN values, the NN model is fed with some σ-moment descriptors computed with the COSMOtherm software for the 111-molecule set. In a preliminary step, the leave-one-out algorithm is used to select, given the available data, the appropriate complexity of the two models. A comparison of the EACNs of liquids of a fresh set of 10 complex cosmetic and perfumery molecules shows that the two approaches provide comparable results in terms of accuracy and reliability. Finally, the NN and GM models are applied to nine series of homologous compounds, for which the GM model results are in better agreement with the experimental EACN trends than the NN model predictions. The results obtained by the GMs and by the NN based on σ-moments can be duplicated with the demonstration tool available for download as detailed in the Supporting Information.

Fast and Accurate Prediction of Refractive Index of Organic Liquids with Graph Machines

“…Interestingly, the size of these datasets is much larger than the size (typically a few hundred compounds) of the datasets used in previous graph machine approaches [29][30][31]; in addition, the number of atom types (up to 16) is larger than the number of atom types present in previous graph machine approaches (up to 6).…”

“…Also based on the 2D structure of molecules, graph machines have the advantage not only of taking the nature of the atoms into account, like a fragment-based approach, but also of preserving their sequence in the molecular structures, thus allowing the estimation of different values for isomers and even diastereomers. We successfully used this tool to predict physico-chemical properties such as surface tension, viscosity, and equivalent alkane carbon number (oil hydrophobicity), showing in particular that it gives complementary results to those obtained with a COSMO-RS approach based on five σ-moment descriptors [29][30][31]. This study thus has two objectives: to test the ability of graph machines to predict the refractive indices of several thousand organic liquids more efficiently than existing models and, from a more methodological standpoint, to verify the ability of graph machines to handle diastereomers.…”

Comparison of Machine Learning Approaches for Prediction of the Equivalent Alkane Carbon Number for Microemulsions Based on Molecular Properties

“…It describes the oil phase hydrophobicity and its ability to penetrate the interfacial film and modify the spontaneous curvature. 26,27,32,33 The EACN represents complex nonpolar molecules as a straight carbon chain. However, beyond those, there is not an intuitive basis for estimating EACN "at a glance".…”