Improved Prediction of Carbonless NMR Spectra by the Machine Learning of Theoretical and Fragment Descriptors for Environmental Mixture Analysis

Abstract

As the first multidimensional NMR approach, 2D Jresolved (2DJ) spectroscopy is distinguished by signal resolution and detection sensitivity with remarkable advantages for the exhaustive evaluation of complex mixtures and environmental samples due to its carbonless feature without the requirement of 13 C connectivity. Generally, the 2DJ signal assignment of metabolic mixtures is problematic in spite of references to experimental NMR databases, owing to the existence of metabolic "dark matter." In this study, a new method to predict 2DJ spectra was developed with a combination of quantum mechanical (QM) computation and machine learning (ML). The predictive accuracy of J-coupling constants was evaluated using validated data. The root-mean-square deviation (RMSD) for QM computation was 3.52 Hz, while the RMSD for QM + ML was 1.21 Hz, indicating a substantial increase in predictive accuracy. The proposed model was applied to predict the 2DJ spectra of 60 standard substances and 55 components of seawater. Furthermore, two practical environmental samples were used to evaluate the robustness of the constructed predictive model. A J-coupling tree and J-split spectra produced from QM + ML of aliphatic moieties had good consistency with the experimental data, as compared with the theoretical data produced by QM computation. The predicted J-coupling tree for the Jcoupling multiplet analysis of freely rotating bonds in the complex mixture, which is traditionally difficult, was interpretable. In addition, in silico identification of the J-split 1 H NMR signals, which was independent of experimental databases, aided in the discovery of new components in a mixture.

Prediction of chemical shift in NMR: A review

“…Shibata Kaneko [65] use RDKit descriptors and decision trees via LightGBM and achieves errors from 0.880 to 0.99 for 1 H, 2 H, and 3 H HH, CH, and NH couplings. Ito et al [66] use a combination of DFT and ML and report an error of 1.21 Hz for HH couplings.…”

Materials informatics approach using domain modelling for exploring structure–property relationships of polymers

“…Traditional design approaches for materials are experimentally driven, facing significant challenges due to the vast design space of materials. Experimental science can be supported by materials informatics 1 – 3 that makes full use of theoretical and computational science such as density functional theory (DFT) 4 , 5 and molecular dynamics (MD) 6 and data science using computers (Artificial Intelligence; AI) 7 , 8 . Computational science solves equations numerically based on theory and physical models.…”

Regression Machine Learning Models Used to Predict DFT-Computed NMR Parameters of Zeolites

“…Furthermore, the symmetry functions are widely used for describing the chemical environment in the neural network representation [1]. Molecular descriptors and fragment descriptors [25] led to predicting with a great accuracy the J-coupling constants in small organic molecules. The ML combination with DFT is therefore a promising tool, and further validations are of high interest.…”