Chemical shift prediction in C NMR spectroscopy using ensembles of message passing neural networks (MPNNs).

This study reports a deep learning approach that utilises message passing neural networks (MPNNs) for predicting chemical shifts in C NMR spectra of small molecules. MPNNs were trained on two distinct datasets: one with approximately 4000 labelled structures and another with over 40,000. To reduce stochastic variation, an ensemble framework was implemented, which is simple to deploy on multiple nodes of a High-Performance Computing facility.

Graphical exploration of 600- and 60-MHz proton NMR spectral datasets from ground roast coffee extracts.

This article uses a variety of graphical and mathematical approaches to analyse 600- and 60-MHz ('benchtop') proton NMR spectra acquired from lipophilic and hydrophilic extracts of roasted coffee beans. The collection of 40 authenticated samples comprised various coffee species, cultivars and hybrids. The spectral datasets were analysed by a combination of metabolomics approaches, cross-correlation and whole spectrum methods, assisted by visualisation and mathematical techniques not conventionally employed to treat NMR data.

Benchtop (60 MHz) proton NMR spectroscopy for quantification of 16-O-methylcafestol in lipophilic extracts of ground roast coffee.

We present a method for analysing the lipophilic fraction extracted from ground coffee beans using 60 MHz proton (H) NMR spectroscopy. In addition to the triglycerides from coffee oil, spectral features are seen from a range of secondary metabolites, such as various diterpenes. We demonstrate quantitation of a peak attributed to one such compound, 16-O-methylcafestol (16-OMC), which is of interest as a coffee species marker.

Authentication of saffron using 60 MHz H NMR spectroscopy.

60 MHz proton NMR spectroscopy was used to analyse extracts from saffron spice and a range of potential adulterants and mixtures. Using a simple extraction procedure, good quality spectra were obtained which contain peaks from the characteristic metabolites picrocrocin and crocins, fatty acids and kaempferol. The spectra of samples from trusted suppliers were used to train one-class classification models by SIMCA, nearest neighbour and isolation forest methods.