Application of quantitative artificial neural network analysis to 2D NMR spectra of hydrocarbon mixtures

Artificial neural networks for quantitative online NMR spectroscopy

Industry 4.0 is all about interconnectivity, sensor-enhanced process control, and data-driven systems. Process analytical technology (PAT) such as online nuclear magnetic resonance (NMR) spectroscopy is gaining in importance, as it increasingly contributes to automation and digitalization in production. In many cases up to now, however, a classical evaluation of process data and their transformation into knowledge is not possible or not economical due to the insufficiently large datasets available. When developing an automated method applicable in process control, sometimes only the basic data of a limited number of batch tests from typical product and process development campaigns are available. However, these datasets are not large enough for training machine-supported procedures. In this work, to overcome this limitation, a new procedure was developed, which allows physically motivated multiplication of the available reference data in order to obtain a sufficiently large dataset for training machine learning algorithms. The underlying example chemical synthesis was measured and analyzed with both application-relevant low-field NMR and high-field NMR spectroscopy as reference method. Artificial neural networks (ANNs) have the potential to infer valuable process information already from relatively limited input data. However, in order to predict the concentration at complex conditions (many reactants and wide concentration ranges), larger ANNs and, therefore, a larger training dataset are required. We demonstrate that a moderately complex problem with four reactants can be addressed using ANNs in combination with the presented PAT method (low-field NMR) and with the proposed approach to generate meaningful training data. Graphical abstract.

Stability and robustness of human metabolic phenotypes in response to sequential food challenges

High-resolution spectroscopic profiles of biofluids can define metabolic phenotypes, providing a window onto the impact of diet on health to reflect gene-environment interactions. (1)H NMR spectroscopic profiling was used to characterize the effect of nutritional intervention on the stability of the metabolic phenotype of 7 individuals following a controlled 7 day dietary protocol. Inter-individual metabolic differences influenced proportionally more of the spectrum than dietary modulation, with certain individuals displaying a greater stability of metabolic phenotypes than others. Correlation structures between urinary metabolites were identified and used to map inter-individual pathway differences. Choline degradation was the pathway most affected by the individual, suggesting that the gut microbiota influence host metabolic phenotypes. This influence was further emphasized by the highly correlated excretion of the microbial-mammalian co-metabolites phenylacetylglutamine, 4-cresylsulfate (r = 0.87), and indoxylsulfate (r = 0.67) across all individuals. Above the background of inter-individual differences, clear biochemical effects of single type dietary interventions, animal protein, fruit and wine intake, were observed; for example, the spectral variance introduced by fruit ingestion was attributed to the metabolites tartrate, proline betaine, hippurate, and 4-hydroxyhippurate. This differential metabolic baseline and response to selected dietary challenges highlights the importance of understanding individual differences in metabolism and provides a rationale for evaluating dietary interventions and stratification of individuals with respect to guiding nutrition and health programmes.

Estimation of VLDL, IDL, LDL, HDL2, apoA-I, and apoB from the Friedewald inputs--apoB and IDL, but not LDL, are associated with mortality in type 1 diabetes

BACKGROUND

There is an unmet need for a straightforward and cost-effective assessment of multiple lipoprotein risk factors for vascular diseases.

AIMS

1) To study the relation of various lipoprotein lipid and apolipoprotein (apo) measures on the Friedewald inputs, i.e. plasma triglycerides (TG), cholesterol (TC), and high-density lipoprotein cholesterol (HDL-C). 2) To build up regression models for the appropriate measures based solely on the Friedewald inputs.

METHODS

Data were available for 1,775 plasma samples, from which very-low-density lipoprotein (VLDL), intermediate-density lipoprotein (IDL), low-density lipoprotein (LDL), and HDL were also isolated by ultracentrifugation. For HDL(2)-C and apolipoproteins, 343 and 247 samples were available, respectively.

RESULTS

Accurate models were obtained for VLDL-TG (cross-validation r=0.98), LDL-C (r=0.91), HDL(2)-C (r=0.92), apoA-I (r=0.92), and apoB (r=0.95). A semi-quantitative model was obtained for IDL-C (r=0.78). Due to the anticipated role of IDL-C in atherosclerosis, it was still kept within the accepted models and pursued further. The associations of the estimates with premature deaths were studied in 4,084 patients with type 1 diabetes. The associations of IDL-C and LDL-C were markedly different, the best predictors of mortality being apoB, apoB to apoA-I ratio, and IDL-C.

CONCLUSIONS

The new models allow identification of clinically relevant lipoprotein profiles with no added cost to the conventional Friedewald formula.

1H NMR metabonomics of plasma lipoprotein subclasses: elucidation of metabolic clustering by self-organising maps

(1)H NMR spectra of plasma are known to provide specific information on lipoprotein subclasses in the form of complex overlapping resonances. A combination of (1)H NMR and self-organising map (SOM) analysis was applied to investigate if automated characterisation of subclass-related metabolic interactions can be achieved. To reliably assess the intrinsic capability of (1)H NMR for resolving lipoprotein subclass profiles, sum spectra representing the pure lipoprotein subclass part of actual plasma were simulated with the aid of experimentally derived model signals for 11 distinct lipoprotein subclasses. Two biochemically characteristic categories of spectra, representing normolipidaemic and metabolic syndrome status, were generated with corresponding lipoprotein subclass profiles. A set of spectra representing a metabolic pathway between the two categories was also generated. The SOM analysis, based solely on the aliphatic resonances of these simulated spectra, clearly revealed the lipoprotein subclass profiles and their changes. Comparable SOM analysis in a group of 69 experimental (1)H NMR spectra of serum samples, which according to biochemical analyses represented a wide range of lipoprotein lipid concentrations, corroborated the findings based on the simulated data. Interestingly, the choline-N(CH(3))(3) region seems to provide more resolved clustering of lipoprotein subclasses in the SOM analyses than the methyl-CH(3) region commonly used for subclass quantification. The results illustrate the inherent suitability of (1)H NMR metabonomics for automated studies of lipoprotein subclass-related metabolism and demonstrate the power of SOM analysis in an extensive and representative case of (1)H NMR metabonomics.

Experimental study of non-linear second-order analytical data with focus on the second-order advantage

Three different experimental systems have been studied regarding the determination of analytes in complex samples, using non-linear second-order instrumental data, which are intrinsically able to provide the second-order advantage. This permits the quantitation of calibrated analytes in the presence of unexpected sample components, although a suitable algorithm is required. The recently described combination of artificial neural networks with post-training residual bilinearization has been applied to the three data sets, with successful results concerning prediction accuracy and precision, as well as profile recovery for the potential interferents in test samples. The studies involve: (1) the determination of two pharmaceuticals in the presence of an unexpected excipient by absorbance-pH matrix measurements, (2) the quantitation of iron(II) by its catalytic effect on the kinetics of the bromate oxidation of a colorant in the presence of a second interfering organic dye, and (3) the analysis of the antibiotic amoxicillin by fluorescence excitation-emission matrices in the presence of a fluorescent anti-inflammatory. The prediction results were compared and shown to be significantly better than those yielded by the unfolded partial least-squares/residual bilinearization model, due to the non-linear nature of the studied data.