We demonstrate the potential for machine learning systems to predict three-dimensional (3D)-relevant NMR properties beyond traditional 1H- and 13C-based data, with comparable accuracy to density functional theory (DFT) (but orders of magnitude faster). Predictions of DFT-calculated 15N chemical shifts for 3D molecular structures can be achieved using a machine learning system—IMPRESSION (Intelligent Machine PREdiction of Shift and Scalar information Of Nuclei), with an accuracy of 6.12-ppm mean absolute error (∼1% of the δ15N chemical shift range) and an error of less than 20 ppm for 95% of the chemical shifts. It provides less accurate raw predictions of experimental chemical shifts, due to the limited size and chemical space diversity of the training dataset used in its creation, coupled with the limitations of the underlying DFT methodology in reproducing experiment.
Bibliographical noteFunding Information:
WG thanks the EPSRC National Productivity Investment Fund (NPIF) and AstraZeneca for Doctoral Studentship funding. CY thanks Genentech for Doctoral Studentship funding.
© 2021 The Authors. Magnetic Resonance in Chemistry published by John Wiley & Sons Ltd.
- machine learning
- NMR chemical shifts