NMR-TS: de novo molecule identification from NMR spectra

Nuclear magnetic resonance (NMR) spectroscopy is an effective tool for identifying molecules in a sample. Although many previously observed NMR spectra are accumulated in public databases, they cover only a tiny fraction of the chemical space, and molecule identification is typically accomplished manually based on expert knowledge. Herein, we propose NMR-TS, a machine-learning-based python library, to automatically identify a molecule from its NMR spectrum. NMR-TS discovers candidate molecules whose NMR spectra match the target spectrum by using deep learning and density functional theory (DFT)-computed spectra. As a proof-of-concept, we identify prototypical metabolites from their computed spectra. After an average 5451 DFT runs for each spectrum, six of the nine molecules are identified correctly, and proximal molecules are obtained in the other cases. This encouraging result implies that de novo molecule generation can contribute to the fully automated identification of chemical structures. NMR-TS is available at https://github.com/tsudalab/NMR-TS.

InterSpin: Integrated Supportive Webtools for Low- and High-Field NMR Analyses Toward Molecular Complexity

InterSpin (http://dmar.riken.jp/interspin/) comprises integrated, supportive, and freely accessible preprocessing webtools and a database to advance signal assignment in low- and high-field NMR analyses of molecular complexities ranging from small molecules to macromolecules for food, material, and environmental applications. To support handling of the broad spectra obtained from solid-state NMR or low-field benchtop NMR, we have developed and evaluated two preprocessing tools: sensitivity improvement with spectral integration, which enhances the signal-to-noise ratio by spectral integration, and peaks separation, which separates overlapping peaks by several algorithms, such as non-negative sparse coding. In addition, the InterSpin Laboratory Information Management System (SpinLIMS) database stores numerous standard spectra ranging from small molecules to macromolecules in solid and solution states (dissolved in polar/nonpolar solvents), and can be searched under various conditions using the following molecular assignment tools. SpinMacro supports easy assignment of macromolecules in natural mixtures via solid-state ¹³C peaks and dimethyl sulfoxide-dissolved ¹H-¹³C correlation peaks. InterAnalysis improves the accuracy of molecular assignment by integrated analysis of ¹H-¹³C correlation peaks and ¹H-J correlation peaks of small molecules dissolved in D₂O or deuterated methanol, which supports easy narrowing down of metabolite candidates. Finally, by enabling database interoperability, SpinLIMS's client software will ultimately support scientific discovery by facilitating sharing and reusing of NMR data.

Exploratory machine-learned theoretical chemical shifts can closely predict metabolic mixture signals

Various chemical shift predictive methodologies have been studied and developed, but there remains the problem of prediction accuracy. Assigning the NMR signals of metabolic mixtures requires high predictive performance owing to the complexity of the signals. Here we propose a new predictive tool that combines quantum chemistry and machine learning. A scaling factor as the objective variable to correct the errors of 2355 theoretical chemical shifts was optimized by exploring 91 machine learning algorithms and using the partial structure of 150 compounds as explanatory variables. The optimal predictive model gave RMSDs between experimental and predicted chemical shifts of 0.2177 ppm for δ 1H and 3.3261 ppm for δ 13C in the test data; thus, better accuracy was achieved compared with existing empirical and quantum chemical methods. The utility of the predictive model was demonstrated by applying it to assignments of experimental NMR signals of a complex metabolic mixture.

Systemic Homeostasis in Metabolome, Ionome, and Microbiome of Wild Yellowfin Goby in Estuarine Ecosystem

Data-driven approaches were applied to investigate the temporal and spatial changes of 1,022 individuals of wild yellowfin goby and its potential interaction with the estuarine environment in Japan. Nuclear magnetic resonance (NMR)-based metabolomics revealed that growth stage is a primary factor affecting muscle metabolism. Then, the metabolic, elemental and microbial profiles of the pooled samples generated according to either the same habitat or sampling season as well as the river water and sediment samples from their habitats were measured using NMR spectra, inductively coupled plasma optical emission spectrometry and next-generation 16 S rRNA gene sequencing. Hidden interactions in the integrated datasets such as the potential role of intestinal bacteria in the control of spawning migration, essential amino acids and fatty acids synthesis in wild yellowfin goby were further extracted using correlation clustering and market basket analysis-generated networks. Importantly, our systematic analysis of both the seasonal and latitudinal variations in metabolome, ionome and microbiome of wild yellowfin goby pointed out that the environmental factors such as the temperature play important roles in regulating the body homeostasis of wild fish.

Application of kernel principal component analysis and computational machine learning to exploration of metabolites strongly associated with diet

Computer-based technological innovation provides advancements in sophisticated and diverse analytical instruments, enabling massive amounts of data collection with relative ease. This is accompanied by a fast-growing demand for technological progress in data mining methods for analysis of big data derived from chemical and biological systems. From this perspective, use of a general “linear” multivariate analysis alone limits interpretations due to “non-linear” variations in metabolic data from living organisms. Here we describe a kernel principal component analysis (KPCA)-incorporated analytical approach for extracting useful information from metabolic profiling data. To overcome the limitation of important variable (metabolite) determinations, we incorporated a random forest conditional variable importance measure into our KPCA-based analytical approach to demonstrate the relative importance of metabolites. Using a market basket analysis, hippurate, the most important variable detected in the importance measure, was associated with high levels of some vitamins and minerals present in foods eaten the previous day, suggesting a relationship between increased hippurate and intake of a wide variety of vegetables and fruits. Therefore, the KPCA-incorporated analytical approach described herein enabled us to capture input–output responses, and should be useful not only for metabolic profiling but also for profiling in other areas of biological and environmental systems.

Environmental metabolomics with data science for investigating ecosystem homeostasis

A natural ecosystem can be viewed as the interconnections between complex metabolic reactions and environments. Humans, a part of these ecosystems, and their activities strongly affect the environments. To account for human effects within ecosystems, understanding what benefits humans receive by facilitating the maintenance of environmental homeostasis is important. This review describes recent applications of several NMR approaches to the evaluation of environmental homeostasis by metabolic profiling and data science. The basic NMR strategy used to evaluate homeostasis using big data collection is similar to that used in human health studies. Sophisticated metabolomic approaches (metabolic profiling) are widely reported in the literature. Further challenges include the analysis of complex macromolecular structures, and of the compositions and interactions of plant biomass, soil humic substances, and aqueous particulate organic matter. To support the study of these topics, we also discuss sample preparation techniques and solid-state NMR approaches. Because NMR approaches can produce a number of data with high reproducibility and inter-institution compatibility, further analysis of such data using machine learning approaches is often worthwhile. We also describe methods for data pretreatment in solid-state NMR and for environmental feature extraction from heterogeneously-measured spectroscopic data by machine learning approaches.

NMR window of molecular complexity showing homeostasis in superorganisms

NMR offers tremendous advantages in the analyses of molecular complexity. The “big-data” are produced during the acquisition of fingerprints that must be stored and shared for posterior analysis and verifications.

Structure and Metabolic-Flow Analysis of Molecular Complexity in a (13) C-Labeled Tree by 2D and 3D NMR

Improved signal identification for biological small molecules (BSMs) in a mixture was demonstrated by using multidimensional NMR on samples from (13) C-enriched Rhododendron japonicum (59.5 atom%) cultivated in air containing (13) C-labeled carbon dioxide for 14 weeks. The resonance assignment of 386 carbon atoms and 380 hydrogen atoms in the mixture was achieved. 42 BSMs, including eight that were unlisted in the spectral databases, were identified. Comparisons between the experimental values and the (13) C chemical shift values calculated by density functional theory supported the identifications of unlisted BSMs. Tracing the (13) C/(12) C ratio by multidimensional NMR spectra revealed faster and slower turnover ratios of BSMs involved in central metabolism and those categorized as secondary metabolites, respectively. The identification of BSMs and subsequent flow analysis provided insight into the metabolic systems of the plant.

The Effect of Molecular Conformation on the Accuracy of Theoretical (1)H and (13)C Chemical Shifts Calculated by Ab Initio Methods for Metabolic Mixture Analysis

NMR spectroscopy is a powerful method for analyzing metabolic mixtures. The information obtained from an NMR spectrum is in the form of physical parameters, such as chemical shifts, and construction of databases for many metabolites will be useful for data interpretation. To increase the accuracy of theoretical chemical shifts for development of a database for a variety of metabolites, the effects of sets of conformations (structural ensembles) and the levels of theory on computations of theoretical chemical shifts were systematically investigated for a set of 29 small molecules in the present study. For each of the 29 compounds, 101 structures were generated by classical molecular dynamics at 298.15 K, and then theoretical chemical shifts for 164 (1)H and 123 (13)C atoms were calculated by ab initio quantum chemical methods. Six levels of theory were used by pairing Hartree-Fock, B3LYP (density functional theory), or second order Møller-Plesset perturbation with 6-31G or aug-cc-pVDZ basis set. The six average fluctuations in the (1)H chemical shift were ±0.63, ± 0.59, ± 0.70, ± 0.62, ± 0.75, and ±0.66 ppm for the structural ensembles, and the six average errors were ±0.34, ± 0.27, ± 0.32, ± 0.25, ± 0.32, and ±0.25 ppm. The results showed that chemical shift fluctuations with changes in the conformation because of molecular motion were larger than the differences between computed and experimental chemical shifts for all six levels of theory. In conclusion, selection of an appropriate structural ensemble should be performed before theoretical chemical shift calculations for development of an accurate database for a variety of metabolites.