Peculiar Reaction Products and Mechanisms Revisited with Machine Learning-Augmented Computational NMR

J Org Chem. 2022 Jul 1;87(13):8589-8598. doi: 10.1021/acs.joc.2c00749. Epub 2022 Jun 18.

Abstract

DU8ML, a fast and accurate machine learning-augmented density functional theory (DFT) method for computing nuclear magnetic resonance (NMR) spectra, proved effective for high-throughput revision of misassigned natural products. In this paper, we disclose another important aspect of its application: correction of unusual reaction mechanisms originally proposed because of incorrect product structures.

Publication types

  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Biological Products* / chemistry
  • Machine Learning
  • Magnetic Resonance Imaging
  • Magnetic Resonance Spectroscopy / methods

Substances

  • Biological Products