Rapid prediction of full spin systems using uncertainty-aware machine learning

Accurate simulation of solution NMR spectra requires knowledge of all chemical shift and scalar coupling parameters, traditionally accomplished by heuristic-based techniques or ab initio computational chemistry methods. Here we present a novel machine learning technique which combines uncertainty-aware deep learning with rapid estimates of conformational geometries to generate Full Spin System Predictions with UnCertainty (FullSSPrUCe). We improve on previous state of the art in accuracy on chemical shift values, predicting protons to within 0.209 ppm and carbons to within 1.213 ppm. Further, we are able to predict all scalar coupling values, unlike previous GNN models, achieving 3JHH accuracies between 0.838 Hz and 1.392 Hz on small experimental datasets. Our uncertainty quantification shows a strong, useful correlation with accuracy, with the most confident predictions having significantly reduced error, including our top-80% most confident proton shift predictions having an average error of only 0.140 ppm. We also properly handle stereoisomerism and intelligently augment experimental data with ab initio data through disagreement regularization to account for deficiencies in training data. FullSSPrUCe is an uncertainty-aware deep learning system which predicts all spin system parameters from 2D structures through rapid estimates of conformational geometries.

Automated summary

This study presents a novel machine learning technique that combines uncertainty-aware deep learning with rapid estimates of conformational geometries to accurately simulate solution NMR spectra. The method improves the accuracy of chemical shift values and can predict all scalar coupling values. The uncertainty quantification results show a strong correlation with accuracy and the method is able to handle stereoisomerism and disagreement regularization.