Leveraging graph neural networks to predict Hammett’s constants for benzoic acid derivatives

The Hammett constants, σ m and σ p , reflect the electron-withdrawing and electron-donating abilities of substituents on aromatic compounds, and have been successfully used in various structure-activity relationship studies. However, determining these constants experimentally is both resource-intensive and time-consuming approach. In this study, we explore the use of graph neural networks (GNNs) to predict Hammett constant parameters using graph-based features. This innovative approach aims to provide rapid and efficient predictions of σ m and σ p values, eliminating the need for extensive computational and experimental setups. By leveraging the power of GNNs, we hope to streamline the process of obtaining these critical parameters, thereby facilitating more efficient reaction design and enhancing the applicability of linear free energy relationship studies in chemical research. This study employs graph neural networks (GNNs) to predict Hammett’s constants, aiming for rapid, efficient predictions without extensive experimental setups, enhancing reaction design and chemical research. • Utilization of graph-based molecular encoding derived from SMILES notations for organic molecules. • Pioneering study using a large-scale dataset to predict Hammett's constant parameters. • Dataset is publicly available, supporting reproducibility and further research. • The Attentive FP algorithm demonstrated high predictive accuracy, achieving an R² score of 0.93 on the test set. • This method offers a rapid and efficient solution for predicting Hammett's constants.