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

Abstract

The refractive index (RI) of liquids is a key physical property of molecular compounds and materials. In addition to its ubiquitous role in physics, it is also exploited to impart specific optical properties (transparency, opacity, and gloss) to materials and various end-use products. Since few methods exist to accurately estimate this property, we have designed a graph machine model (GMM) capable of predicting the RI of liquid organic compounds containing up to 16 different types of atoms and effective in discriminating between stereoisomers. Using 8267 carefully checked RI values from the literature and the corresponding 2D organic structures, the GMM provides a training root mean square relative error of less than 0.5%, i.e., an RMSE of 0.004 for the estimation of the refractive index of the 8267 compounds. The GMM predictive ability is also compared to that obtained by several fragment-based approaches. Finally, a Docker-based tool is proposed to predict the RI of organic compounds solely from their SMILES code. The GMM developed is easy to apply, as shown by the video tutorials provided on YouTube.

Can Graph Machines Accurately Estimate ¹³C NMR Chemical Shifts of Benzenic Compounds?

“…The design of graph machine models requires a dataset of measured experimental values, presently a set of 13 C chemical shift values in the present case. An important difference as compared to the estimation of surface tension, viscosity or refraction index [81][82][83], is that the property under study is an atomic property instead of a molecular one. Consequently, the carbons of the benzenic molecules must be annotated with their experimental chemical shifts.…”

“…In graph machine-based models, molecules are described as graphs derived from their 2Dstructure, and the parameterized functions (called graph machines) that compute the estimation of the property or activity of interest reflect the compound molecular structures. The procedure for graph machine construction has been described in details elsewhere [83,85,86]. In this case, the main difference with previous descriptions of graph machine design, is that the property is computed for each benzenic carbon atom of all molecular structures, i.e.…”