Machine Learning of Molecular Classification and Quantum Mechanical Calculations

Boosting Graph Neural Networks with Molecular Mechanics: A Case Study of Sigma Profile Prediction

“…As such, the development of predictive methodologies able to estimate sigma profiles without the use of quantum-chemistry-based calculations would greatly en- hance the applicability of these molecular descriptors in largescale ML models. Efforts in this direction are scarce in the literature and are mostly focused on complex language-based ML architectures such as those deployed by Lin and coworkers, 9,10 Kang et al, 11 and Chen et al 12 The work of Zhang et al 13 stands out as the exception, where message passing graph neural networks (GNNs) were explored. GNNs are a class of ML models that have a similar architecture to artificial neural networks but operate on graphs.…”