Using neural networks to obtain NMR spectra of both small and macromolecules from blood samples in a single experiment

Metabolomics plays a crucial role in understanding metabolic processes within biological systems. Using specific pulse sequences, NMR-based metabolomics detects small and macromolecular metabolites that are altered in blood samples. Here we proposed a method called spectral editing neural network, which can effectively edit and separate the spectral signals of small and macromolecules in 1H NMR spectra of serum and plasma based on the linewidth of the peaks. We applied the model to process the 1H NMR spectra of plasma and serum. The extracted small and macromolecular spectra were then compared with experimentally obtained relaxation-edited and diffusion-edited spectra. Correlation analysis demonstrated the quantitative capability of the model in the extracted small molecule signals from 1H NMR spectra. The principal component analysis showed that the spectra extracted by the model and those obtained by NMR spectral editing methods reveal similar group information, demonstrating the effectiveness of the model in signal extraction.

Preliminary Findings on the Salivary Metabolome of Hookah and Cigarette Smokers

The aim of the study was to evaluate the salivary metabolomic profile of patients who habitually smoke hookah and cigarettes. The groups consisted of 33 regular and exclusive hookah smokers, 26 regular and exclusive cigarette smokers, and 30 nonsmokers. Unstimulated whole saliva was collected for the measurement of salivary metabolites by gas chromatography coupled with tandem mass spectrometry (GC-MS/MS). The MetaboAnalyst software was used for statistical analysis and evaluation of biomarkers. 11 smoking salivary biomarkers were identified using the area under receiving-operator curver criterion and threshold of 0.9. Xylitol and octadecanol were higher in cigarette smokers compared to controls; arabitol and maltose were higher in controls compared to cigarette smokers; octadecanol and tyramine were higher in hookah smokers compared to controls; phenylalanine was higher in controls compared to hookah smokers; and fructose, isocitric acid, glucuronic acid, tryptamine, maltose, tyramine, and 3-hydroxyisolvaleric acid were higher in hookah smokers compared to cigarettes smokers. Conclusions: The evaluation of the salivary metabolome of hookah smokers, showing separation between the groups, especially between the control versus hookah groups and cigarette versus hookah groups, and it seems to demonstrate that the use of hookah tobacco is more damaging to health.

Role of untargeted omics biomarkers of exposure and effect for tobacco research

Tobacco research remains a clear priority to improve individual and population health, and has recently become more complex with emerging combustible and noncombustible tobacco products. The use of omics methods in prevention and cessation studies are intended to identify new biomarkers for risk, compared risks related to other products and never use, and compliance for cessation and reinitation. to assess the relative effects of tobacco products to each other. They are important for the prediction of reinitiation of tobacco use and relapse prevention. In the research setting, both technical and clinical validation is required, which presents a number of complexities in the omics methodologies from biospecimen collection and sample preparation to data collection and analysis. When the results identify differences in omics features, networks or pathways, it is unclear if the results are toxic effects, a healthy response to a toxic exposure or neither. The use of surrogate biospecimens (e.g., urine, blood, sputum or nasal) may or may not reflect target organs such as the lung or bladder. This review describes the approaches for the use of omics in tobacco research and provides examples of prior studies, along with the strengths and limitations of the various methods. To date, there is little consistency in results, likely due to small number of studies, limitations in study size, the variability in the analytic platforms and bioinformatic pipelines, differences in biospecimen collection and/or human subject study design. Given the demonstrated value for the use of omics in clinical medicine, it is anticipated that the use in tobacco research will be similarly productive.

Approaches to Integrating Metabolomics and Multi-Omics Data: A Primer

Metabolomics deals with multiple and complex chemical reactions within living organisms and how these are influenced by external or internal perturbations. It lies at the heart of omics profiling technologies not only as the underlying biochemical layer that reflects information expressed by the genome, the transcriptome and the proteome, but also as the closest layer to the phenome. The combination of metabolomics data with the information available from genomics, transcriptomics, and proteomics offers unprecedented possibilities to enhance current understanding of biological functions, elucidate their underlying mechanisms and uncover hidden associations between omics variables. As a result, a vast array of computational tools have been developed to assist with integrative analysis of metabolomics data with different omics. Here, we review and propose five criteria-hypothesis, data types, strategies, study design and study focus- to classify statistical multi-omics data integration approaches into state-of-the-art classes under which all existing statistical methods fall. The purpose of this review is to look at various aspects that lead the choice of the statistical integrative analysis pipeline in terms of the different classes. We will draw particular attention to metabolomics and genomics data to assist those new to this field in the choice of the integrative analysis pipeline.

Predicting human health from biofluid-based metabolomics using machine learning

Biofluid-based metabolomics has the potential to provide highly accurate, minimally invasive diagnostics. Metabolomics studies using mass spectrometry typically reduce the high-dimensional data to only a small number of statistically significant features, that are often chemically identified—where each feature corresponds to a mass-to-charge ratio, retention time, and intensity. This practice may remove a substantial amount of predictive signal. To test the utility of the complete feature set, we train machine learning models for health state-prediction in 35 human metabolomics studies, representing 148 individual data sets. Models trained with all features outperform those using only significant features and frequently provide high predictive performance across nine health state categories, despite disparate experimental and disease contexts. Using only non-significant features it is still often possible to train models and achieve high predictive performance, suggesting useful predictive signal. This work highlights the potential for health state diagnostics using all metabolomics features with data-driven analysis.

Inhibition of enteric methanogenesis in dairy cows induces changes in plasma metabolome highlighting metabolic shifts and potential markers of emission

There is scarce information on whether inhibition of rumen methanogenesis induces metabolic changes on the host ruminant. Understanding these possible changes is important for the acceptance of methane-reducing practices by producers. In this study we explored the changes in plasma profiles associated with the reduction of methane emissions. Plasma samples were collected from lactating primiparous Holstein cows fed the same diet with (Treated, n = 12) or without (Control, n = 13) an anti-methanogenic feed additive for six weeks. Daily methane emissions (CH₄, g/d) were reduced by 23% in the Treated group with no changes in milk production, feed intake, body weight, and biochemical indicators of health status. Plasma metabolome analyses were performed using untargeted [nuclear magnetic resonance (NMR) and liquid chromatography-mass spectrometry (LC–MS)] and targeted (LC–MS/MS) approaches. We identified 48 discriminant metabolites. Some metabolites mainly of microbial origin such as dimethylsulfone, formic acid and metabolites containing methylated groups like stachydrine, can be related to rumen methanogenesis and can potentially be used as markers. The other discriminant metabolites are produced by the host or have a mixed microbial-host origin. These metabolites, which increased in treated cows, belong to general pathways of amino acids and energy metabolism suggesting a systemic non-negative effect on the animal.

Use of an Exposome Approach to Understand the Effects of Exposures From the Natural, Built, and Social Environments on Cardio-Vascular Disease Onset, Progression, and Outcomes

Obesity, diabetes, and hypertension have increased by epidemic proportions in recent years among African Americans in comparison to Whites resulting in significant adverse cardiovascular disease (CVD) disparities. Today, African Americans are 30% more likely to die of heart disease than Whites and twice as likely to have a stroke. The causes of these disparities are not yet well-understood. Improved methods for identifying underlying risk factors is a critical first step toward reducing Black:White CVD disparities. This article will focus on environmental exposures in the external environment and how they can lead to changes at the cellular, molecular, and organ level to increase the personal risk for CVD and lead to population level CVD racial disparities. The external environment is defined in three broad domains: natural (air, water, land), built (places you live, work, and play) and social (social, demographic, economic, and political). We will describe how environmental exposures in the natural, built, and social environments "get under the skin" to affect gene expression though epigenetic, pan-omics, and related mechanisms that lead to increased risk for adverse CVD health outcomes and population level disparities. We also will examine the important role of metabolomics, proteomics, transcriptomics, genomics, and epigenomics in understanding how exposures in the natural, built, and social environments lead to CVD disparities with implications for clinical, public health, and policy interventions. In this review, we apply an exposome approach to Black:White CVD racial disparities. The exposome is a measure of all the exposures of an individual across the life course and the relationship of those exposures to health effects. The exposome represents the totality of exogenous (external) and endogenous (internal) exposures from conception onwards, simultaneously distinguishing, characterizing, and quantifying etiologic, mediating, moderating, and co-occurring risk and protective factors and their relationship to disease. Specifically, it assesses the biological mechanisms and underlying pathways through which chemical and non-chemical environmental exposures are associated with CVD onset, progression and outcomes. The exposome is a promising approach for understanding the complex relationships among environment, behavior, biology, genetics, and disease phenotypes that underlie population level, Black: White CVD disparities.

COPD phenotypes and machine learning cluster analysis: A systematic review and future research agenda

Chronic Obstructive Pulmonary Disease (COPD) is a highly heterogeneous condition projected to become the third leading cause of death worldwide by 2030. To better characterize this condition, clinicians have classified patients sharing certain symptomatic characteristics, such as symptom intensity and history of exacerbations, into distinct phenotypes. In recent years, the growing use of machine learning algorithms, and cluster analysis in particular, has promised to advance this classification through the integration of additional patient characteristics, including comorbidities, biomarkers, and genomic information. This combination would allow researchers to more reliably identify new COPD phenotypes, as well as better characterize existing ones, with the aim of improving diagnosis and developing novel treatments. Here, we systematically review the last decade of research progress, which uses cluster analysis to identify COPD phenotypes. Collectively, we provide a systematized account of the extant evidence, describe the strengths and weaknesses of the main methods used, identify gaps in the literature, and suggest recommendations for future research.

Metabolomics Profiles of Smokers from Two Ethnic Groups with Differing Lung Cancer Risk

African American (AA) smokers are at a higher risk of developing lung cancer compared to whites. The variations in the metabolism of nicotine and tobacco-derived carcinogens in these groups were reported previously with the levels of nicotine metabolites and carcinogen-derived metabolites measured using targeted approaches. While useful, these targeted strategies are not able to detect global metabolic changes for use in predicting the detrimental effects of tobacco use and ultimately lung cancer susceptibility among smokers. To address this limitation, we have performed global untargeted metabolomics profiling in urine of AA and white smokers to characterize the pattern of metabolites, identify differentially regulated pathways, and correlate these profiles with the observed variations in lung cancer risk between these two populations. Urine samples from AA (n = 30) and white (n = 30) smokers were used for metabolomics analysis acquired in both positive and negative electrospray ionization modes. LC-MS data were uploaded onto the cloud-based XCMS online (http://xcmsonline.scripps.edu) platform for retention time correction, alignment, feature detection, annotation, statistical analysis, data visualization, and automated systems biology pathway analysis. The latter identified global differences in the metabolic pathways in the two groups including the metabolism of carbohydrates, amino acids, nucleotides, fatty acids, and nicotine. Significant differences in the nicotine degradation pathway (cotinine glucuronidation) in the two groups were observed and confirmed using a targeted LC-MS/MS approach. These results are consistent with previous studies demonstrating AA smokers with lower glucuronidation capacity compared to whites. Furthermore, the d-glucuronate degradation pathway was found to be significantly different between the two populations, with lower amounts of the putative metabolites detected in AA compared to whites. We hypothesize that the differential regulation of the d-glucuronate degradation pathway is a consequence of the variations in the glucuronidation capacity observed in the two groups. Other pathways including the metabolism of amino acids, nucleic acids, and fatty acids were also identified, however, the biological relevance and implications of these differences across ethnic groups need further investigation. Overall, the applied metabolomics approach revealed global differences in the metabolic networks and endogenous metabolites in AA and whites, which could be used and validated as a new potential panel of biomarkers that could be used to predict lung cancer susceptibility among smokers in population-based studies.

Impact of maternal smoking associated lyso-phosphatidylcholine 20:3 on offspring brain development

Maternal smoking during pregnancy affects fetal neurological development. Metabolomic studies in the general population suggest that smoking is associated with characteristic metabolic alterations. We investigated the association between the maternal smoking status, the fetal metabolome and head circumference at birth, as a surrogate parameter of brain development. 320 mother/newborn pairs of the Berlin Birth Cohort were investigated. Anthropometric parameters, including head circumference, of newborns of smoking mothers, former smoking mothers, and never smoking mothers were compared to assess the impact of maternal smoking behavior. Associations between maternal smoking behavior and 163 cord blood metabolites and associations between newborn head circumference and concentrations of smoking behavior related metabolites were analysed. Male newborns of smoking mothers had a reduced head circumference when compared with newborns from former smoking and never smoking mothers (p < 0.05). Using linear regression models corrected for established confounding factors, maternal smoking during pregnancy showed an independent association with head circumference (95% CI: -0.75～-0.41 cm, p = 2.45×10^-11). In a stepwise linear regression model corrected for known confounding factors of brain growth lyso-phosphatidylcholine 20:3 (95% CI: 6.68～39.88 cm, p = 4.62×10^-4) was associated with head circumference in male offspring only. None of the metabolites were associated with head circumference of female newborns. In conclusion, maternal smoking during pregnancy impacted on male offspring's development including brain development. The smoking related metabolite lyso-phosphatidylcholine 20:3 was associated with head circumference of male offspring.

Multi-omics systems toxicology study of mouse lung assessing the effects of aerosols from two heat-not-burn tobacco products and cigarette smoke

•

Multi-omics systems toxicology study, comprising five omics data modalities.

•

Multi-Omics Factor Analysis and multi-modality functional network interpretation.

•

Cigarettes smoke (CS) induced complex immunoregulatory interactions across molecular layers.

•

Aerosols from two heat-not-burn tobacco products had less impact on lungs than CS.

Cigarette smoke (CS) causes adverse health effects and, for smoker who do not quit, modified risk tobacco products (MRTPs) can be an alternative to reduce the risk of developing smoking-related diseases. Standard toxicological endpoints can lack sensitivity, with systems toxicology approaches yielding broader insights into toxicological mechanisms. In a 6-month systems toxicology study on ApoE^−/− mice, we conducted an integrative multi-omics analysis to assess the effects of aerosols from the Carbon Heated Tobacco Product (CHTP) 1.2 and Tobacco Heating System (THS) 2.2—a potential and a candidate MRTP based on the heat-not-burn (HnB) principle—compared with CS at matched nicotine concentrations. Molecular exposure effects in the lungs were measured by mRNA/microRNA transcriptomics, proteomics, metabolomics, and lipidomics. Integrative data analysis included Multi-Omics Factor Analysis and multi-modality functional network interpretation. Across all five data modalities, CS exposure was associated with an increased inflammatory and oxidative stress response, and lipid/surfactant alterations. Upon HnB aerosol exposure these effects were much more limited or absent, with reversal of CS-induced effects upon cessation and switching to CHTP 1.2. Functional network analysis revealed CS-induced complex immunoregulatory interactions across the investigated molecular layers (e.g., itaconate, quinolinate, and miR-146) and highlighted the engagement of the heme–Hmox–bilirubin oxidative stress axis by CS. This work exemplifies how multi-omics approaches can be leveraged within systems toxicology studies and the generated multi-omics data set can facilitate the development of analysis methods and can yield further insights into the effects of toxicological exposures on the lung of mice.

Pathway Analysis of Global Metabolomic Profiles Identified Enrichment of Caffeine, Energy, and Arginine Metabolism in Smokers but Not Moist Snuff Consumers

Existing US epidemiological data demonstrate that consumption of smokeless tobacco, particularly moist snuff, is less harmful than cigarette smoking. However, the molecular and biochemical changes due to moist snuff consumption relative to smoking remain incompletely understood. We previously reported that smokers (SMK) exhibit elevated oxidative stress and inflammation relative to moist snuff consumers (MSC) and non-tobacco consumers (NTC), based on metabolomic profiling data of saliva, plasma, and urine from MSC, SMK, and NTC. In this study, we investigated the effects of tobacco consumption on additional metabolic pathways using pathway-based analysis tools. To this end, metabolic pathway enrichment analysis and topology analysis were performed through pair-wise comparisons of global metabolomic profiles of SMK, MSC, and NTC. The analyses identified >8 significantly perturbed metabolic pathways in SMK compared with NTC and MSC in all 3 matrices. Among these differentially enriched pathways, perturbations of caffeine metabolism, energy metabolism, and arginine metabolism were mostly observed. In comparison, fewer enriched metabolic pathways were identified in MSC compared with NTC (5 in plasma, none in urine and saliva). This is consistent with our transcriptomics profiling results that show no significant differences in peripheral blood mononuclear cell gene expression between MSC and NTC. These findings, taken together with our previous biochemical, metabolomic, and transcriptomic analysis results, provide a better understanding of the relative changes in healthy tobacco consumers, and demonstrate that chronic cigarette smoking, relative to the use of smokeless tobacco, results in more pronounced biological changes, which could culminate in smoking-related diseases.

Improving sample classification by harnessing the potential of 1H-NMR signal chemical shifts

NMR spectroscopy is a technology that is widely used in metabolomic studies. The information that these studies most commonly use from NMR spectra is the metabolite concentration. However, as well as concentration, pH and ionic strength information are also made available by the chemical shift of metabolite signals. This information is typically not used even though it can enhance sample discrimination, since many conditions show pH or ionic imbalance. Here, we demonstrate how chemical shift information can be used to improve the quality of the discrimination between case and control samples in three public datasets of different human matrices. In two of these datasets, chemical shift information helped to provide an AUROC value higher than 0.9 during sample classification. In the other dataset, the chemical shift also showed discriminant potential (AUROC 0.831). These results are consistent with the pH imbalance characteristic of the condition studied in the datasets. In addition, we show that this signal misalignment dependent on sample class can alter the results of fingerprinting approaches in the three datasets. Our results show that it is possible to use chemical shift information to enhance the diagnostic and predictive properties of NMR.

Evaluation and comparison of bioinformatic tools for the enrichment analysis of metabolomics data

BACKGROUND

Bioinformatic tools for the enrichment of 'omics' datasets facilitate interpretation and understanding of data. To date few are suitable for metabolomics datasets. The main objective of this work is to give a critical overview, for the first time, of the performance of these tools. To that aim, datasets from metabolomic repositories were selected and enriched data were created. Both types of data were analysed with these tools and outputs were thoroughly examined.

RESULTS

An exploratory multivariate analysis of the most used tools for the enrichment of metabolite sets, based on a non-metric multidimensional scaling (NMDS) of Jaccard's distances, was performed and mirrored their diversity. Codes (identifiers) of the metabolites of the datasets were searched in different metabolite databases (HMDB, KEGG, PubChem, ChEBI, BioCyc/HumanCyc, LipidMAPS, ChemSpider, METLIN and Recon2). The databases that presented more identifiers of the metabolites of the dataset were PubChem, followed by METLIN and ChEBI. However, these databases had duplicated entries and might present false positives. The performance of over-representation analysis (ORA) tools, including BioCyc/HumanCyc, ConsensusPathDB, IMPaLA, MBRole, MetaboAnalyst, Metabox, MetExplore, MPEA, PathVisio and Reactome and the mapping tool KEGGREST, was examined. Results were mostly consistent among tools and between real and enriched data despite the variability of the tools. Nevertheless, a few controversial results such as differences in the total number of metabolites were also found. Disease-based enrichment analyses were also assessed, but they were not found to be accurate probably due to the fact that metabolite disease sets are not up-to-date and the difficulty of predicting diseases from a list of metabolites.

CONCLUSIONS

We have extensively reviewed the state-of-the-art of the available range of tools for metabolomic datasets, the completeness of metabolite databases, the performance of ORA methods and disease-based analyses. Despite the variability of the tools, they provided consistent results independent of their analytic approach. However, more work on the completeness of metabolite and pathway databases is required, which strongly affects the accuracy of enrichment analyses. Improvements will be translated into more accurate and global insights of the metabolome.

Assessing modified risk tobacco and nicotine products: Description of the scientific framework and assessment of a closed modular electronic cigarette

Cigarette smoking causes many human diseases including cardiovascular disease, lung disease and cancer. Novel tobacco products with reduced yields of toxicants compared to cigarettes, such as tobacco-heating products, snus and electronic cigarettes, hold great potential for reducing the harms associated with tobacco use. In the UK several public health agencies have advocated a potential role for novel products in tobacco harm reduction. Public Health England has stated that "The current best estimate is that e-cigarettes are around 95% less harmful than smoking" and the Royal College of Physicians has urged public health to "Promote e-cigarettes widely as substitute for smoking". Health related claims on novel products such as 'reduced exposure' and 'reduced risk' should be substantiated using a weight of evidence approach based on a comprehensive scientific assessment. The US FDA, has provided draft guidance outlining a framework to assess novel products as Modified Risk Tobacco Products (MRTP). Based on this, we now propose a framework comprising pre-clinical, clinical, and population studies to assess the risk profile of novel tobacco products. Additionally, the utility of this framework is assessed through the pre-clinical and part of the clinical comparison of a commercial e-cigarette (Vype ePen) with a scientific reference cigarette (3R4F) and the results of these studies suggest that ePen has the potential to be a reduced risk product.

A Review of Pulmonary Toxicity of Electronic Cigarettes in the Context of Smoking: A Focus on Inflammation

The use of electronic cigarettes (e-cigs) is increasing rapidly, but their effects on lung toxicity are largely unknown. Smoking is a well-established cause of lung cancer and respiratory disease, in part through inflammation. It is plausible that e-cig use might affect similar inflammatory pathways. E-cigs are used by some smokers as an aid for quitting or smoking reduction, and by never smokers (e.g., adolescents and young adults). The relative effects for impacting disease risk may differ for these groups. Cell culture and experimental animal data indicate that e-cigs have the potential for inducing inflammation, albeit much less than smoking. Human studies show that e-cig use in smokers is associated with substantial reductions in blood or urinary biomarkers of tobacco toxicants when completely switching and somewhat for dual use. However, the extent to which these biomarkers are surrogates for potential lung toxicity remains unclear. The FDA now has regulatory authority over e-cigs and can regulate product and e-liquid design features, such as nicotine content and delivery, voltage, e-liquid formulations, and flavors. All of these factors may impact pulmonary toxicity. This review summarizes current data on pulmonary inflammation related to both smoking and e-cig use, with a focus on human lung biomarkers. Cancer Epidemiol Biomarkers Prev; 26(8); 1175-91. ©2017 AACR.

A novel hybrid tobacco product that delivers a tobacco flavour note with vapour aerosol (Part 2): In vitro biological assessment and comparison with different tobacco-heating products

This study assessed the toxicological and biological responses of aerosols from a novel hybrid tobacco product. Toxicological responses from the hybrid tobacco product were compared to those from a commercially available Tobacco Heating Product (c-THP), a prototype THP (p-THP) and a 3R4F reference cigarette, using in vitro test methods which were outlined as part of a framework to substantiate the risk reduction potential of novel tobacco and nicotine products. Exposure matrices used included total particulate matter (TPM), whole aerosol (WA), and aqueous aerosol extracts (AqE) obtained after machine-puffing the test products under the Health Canada Intense smoking regime. Levels of carbonyls and nicotine in these matrices were measured to understand the aerosol dosimetry of the products. The hybrid tobacco product tested negative across the in vitro assays including mutagenicity, genotoxicity, cytotoxicity, tumour promotion, oxidative stress and endothelial dysfunction. All the THPs tested demonstrated significantly reduced responses in these in vitro assays when compared to 3R4F. The findings suggest these products have the potential for reduced health risks. Further pre-clinical and clinical assessments are required to substantiate the risk reduction of these novel products at individual and population levels.