Predictive Modeling for Metabolomics Data

Non-destructive detection and identification of plasticizers in PVC objects by means of machine learning-assisted Raman spectroscopy

Vibrational spectroscopic techniques, such as Raman spectroscopy, as a non-destructive method combined with machine learning (ML), were successfully tested as a quick method of plasticizer identification in poly(vinyl chloride) - PVC objects in heritage collection. ML algorithms such as Convolutional Neural Network (CNN), Random Forest (RF), Support Vector Machines (SVM), and Linear Discriminant Analysis (LDA) were applied to the classification and identification of the most common plasticizers used in the case of PVC. The CNN model was able to successfully classify the five plasticizers under study from their Raman spectra with a high accuracy of (98%), whereas the highest accuracy (100%) was observed with the RF algorithm. The finding opens doors for the development of robust and economical tools for conservators and museum professionals for fast identification of materials in heritage collections.

Enhancing supervised analysis of imbalanced untargeted metabolomics datasets using a CWGAN-GP framework for data augmentation

Untargeted metabolomics is an extremely useful approach for the discrimination of biological systems and biomarker identification. However, data analysis workflows are complex and face many challenges. Two of these challenges are the demand of high sample size and the possibility of severe class imbalance, which is particularly common in clinical studies. The latter can make statistical models less generalizable, increase the risk of overfitting and skew the analysis in favour of the majority class. One possible approach to mitigate this problem is data augmentation. However, the use of artificial data requires adequate data augmentation methods and criteria for assessing the quality of the generated data. In this work, we used Conditional Wasserstein Generative Adversarial Networks with Gradient Penalty (CWGAN-GPs) for data augmentation of metabolomics data. Using a set of benchmark datasets, we applied several criteria for the evaluation of the quality of generated data and assessed the performance of supervised predictive models trained with datasets that included such data. CWGAN-GP models generated realistic data with identical characteristics to real samples, mostly avoiding mode collapse. Furthermore, in cases of class imbalance, the performance of predictive models improved by supplementing the minority class with generated samples. This is evident for high quality datasets with well separated classes. Conversely, model improvements were quite modest for high class overlap datasets. This trend was confirmed by using synthetic datasets with different class separation levels. Data augmentation is a viable procedure to alleviate class imbalance problems but is not universally beneficial in metabolomics.

Administration Route Differentiation of Altrenogest via the Metabolomic LC-HRMS Analysis of Equine Urine

Altrenogest, also known as allyltrenbolone, is a synthetic form of progesterone used therapeutically to suppress unwanted symptoms of estrus in female horses. Altrenogest affects the system by decreasing levels of endogenous gonadotrophin and luteinizing and follicle-stimulating hormones, which in turn decreases estrogen and mimics the increase of progesterone production. This results in more manageable mares for training and competition alongside male horses while improving the workplace safety of riders and handlers. However, when altrenogest is administered, prohibited steroid impurities such as trendione, trenbolone, and epitrenbolone can be detected. It has been assumed that greater concentrations of these steroid impurities are present in injectable preparations and, therefore, pose a greater risk of causing anabolic effects when administered. For this reason, and due to the necessity of this therapeutic substance for the safety of thoroughbred racing participants, a metabolomic approach investigating the differentiation of two main administration routes was conducted. Liquid chromatography high-resolution mass spectrometry analysis of equine urine samples found five sulfated compounds, estrone sulfate, testosterone sulfate, 2-methoxyestradiol sulfate, pregnenolone sulfate, and cortisol sulfate, with the potential to differentiate between oral and intramuscularly administered altrenogest using a random forest classification model. The best model results, comparing two horses' administration normalized peak area datasets, gave an AUC score of 0.965 with a confidence level of 95% (between 0.931 and 0.995). Identifications of these compounds were confirmed with assistance from the Shimadzu Insight Explore Assign feature, together with MS/MS spectrum and retention time matching of purchased and synthesized reference standards. This study proposes a new potential application for metabolomic multi-tool workflows and machine learning models in a forensic toxicological context.

Metabolomic profiling of dengue infection: unraveling molecular signatures by LC-MS/MS and machine learning models

BACKGROUND & OBJECTIVE

The progression of dengue fever to severe dengue (SD) is a major public health concern that impairs the capacity of the medical system to predict and treat dengue patients. Hence, the present study used a metabolomic approach integrated with machine models to identify differentially expressed metabolites in patients with SD compared to nonsevere patients and healthy controls.

METHODS

Comprehensively, the plasma was collected at different clinical phases during dengue without warning signs (DWOW, N = 10), dengue with warning signs (DWW, N = 10), and SD (N = 10) at different stages [i.e., day of admission (DOA), day of defervescence (DOD), and day of convalescent (DOC)] in comparison to healthy control (HC). The samples were subjected to LC‒ESI‒MS/MS to identify metabolites. Statistical and machine learning analyses were performed using R and Python language. Further, biomarker, pathway and correlation analysis was performed to identify potential predictors of dengue.

RESULTS & CONCLUSION

A total of 423 metabolites were identified in all the study groups. Paired and unpaired t-tests revealed 14 highly differentially expressed metabolites between and across the dengue groups, with four metabolites (shikimic acid, ureidosuccinic acid, propionyl carnitine, and alpha-tocopherol) showing significant differences compared to HC. Furthermore, biomarker (ROC) analysis revealed 11 potential molecules with a significant AUC value of 1 that could serve as potential biomarkers for identifying different dengue clinical stages that are beneficial for predicting dengue disease outcomes. The logistic regression model revealed that S-adenosylhomocysteine, hypotaurine, and shikimic acid metabolites could be beneficial indicators for predicting severe dengue, with an accuracy and AUC of 0.75. The data showed that dengue infection is related to lipid metabolism, oxidative stress, inflammation, metabolomic adaptation, and virus manipulation. Moreover, the biomarkers had a significant correlation with biochemical parameters like platelet count, and hematocrit. These results shed some light on host-derived small-molecule biomarkers that are associated with dengue severity and novel insights into metabolomics mechanisms interlinked with disease severity.

Metabonomic analysis to identify exometabolome changes underlying antifungal and growth promotion mechanisms of endophytic Actinobacterium Streptomyces albidoflavus for sustainable agriculture practice

In recent years, there has been an increasing focus on microbial ecology and its possible impact on agricultural production, owing to its eco-friendly nature and sustainable use. The current study employs metabolomics technologies and bioinformatics approaches to identify changes in the exometabolome of Streptomyces albidoflavus B24. This research aims to shed light on the mechanisms and metabolites responsible for the antifungal and growth promotion strategies, with potential applications in sustainable agriculture. Metabolomic analysis was conducted using Q Exactive UPLC-MS/MS. Our findings indicate that a total of 3,840 metabolites were identified, with 137 metabolites exhibiting significant differences divided into 61 up and 75 downregulated metabolites based on VIP >1, |FC| >1, and p < 0.01. The interaction of S. albidoflavus B24 monoculture with the co-culture demonstrated a stronger correlation coefficient. The Principal Component Analysis (PCA) demonstrates that PCA1 accounted for 23.36%, while PCA2 accounted for 20.28% distinction. OPLS-DA score plots indicate significant separation among different groups representing (t1) 24% as the predicted component (to1) depicts 14% as the orthogonal component. According to the findings of this comprehensive study, crude extracts from S. albidoflavus demonstrated varying abilities to impede phytopathogen growth and enhance root and shoot length in tested plants. Through untargeted metabolomics, we discovered numerous potential molecules with antagonistic activity against fungal phytopathogens among the top 10 significant metabolites with the highest absolute log2FC values. These include Tetrangulol, 4-Hydroxybenzaldehyde, and Cyclohexane. Additionally, we identified plant growth-regulating metabolites such as N-Succinyl-L-glutamate, Nicotinic acid, L-Aspartate, and Indole-3-acetamide. The KEGG pathway analysis has highlighted these compounds as potential sources of antimicrobial properties. The inhibitory effect of S. albidoflavus crude extracts on pathogen growth is primarily attributed to the presence of specific gene clusters responsible for producing cyclic peptides such as ansamycins, porphyrin, alkaloid derivatives, and neomycin. Overall, it is apparent that crude extracts from S. albidoflavus exhibited varying abilities to inhibit the growth of three phytopathogens and enhancement in both root and shoot length of tested plants. This research enhances our understanding of how secondary metabolites contribute to growth promotion and biocontrol, supporting ecosystem sustainability and resilience while boosting productivity in sustainable agriculture.

Finding the best predictive model for hypertensive depression in older adults based on machine learning and metabolomics research

Objective

Depression is a common comorbidity in hypertensive older adults, yet depression is more difficult to diagnose correctly. Our goal is to find predictive models of depression in hypertensive patients using a combination of various machine learning (ML) methods and metabolomics.

Methods

Methods We recruited 379 elderly people aged ≥65 years from the Chinese community. Plasma samples were collected and assayed by gas chromatography/liquid chromatography-mass spectrometry (GC/LC-MS). Orthogonal partial least squares discriminant analysis (OPLS-DA), volcano diagrams and thermograms were used to distinguish metabolites. The attribute discriminators CfsSubsetEval combined with search method BestFirst in WEKA software was used to find the best predicted metabolite combinations, and then 24 classification methods with 10-fold cross-validation were used for prediction.

Results

34 individuals were considered hypertensive combined with depression according to our criteria, and 34 subjects with hypertension only were matched according to age and sex. 19 metabolites by GC-MS and 65 metabolites by LC-MS contributed significantly to the differentiation between the depressed and non-depressed cohorts, with a VIP value of more than 1 and a P value of less than 0.05. There were multiple metabolic pathway alterations. The metabolite combinations screened with WEKA for optimal diagnostic value included 12 metabolites. The machine learning methods with AUC values greater than 0.9 were bayesNet and random forests, and their other evaluation measures are also better.

Conclusion

Altered metabolites and metabolic pathways are present in older adults with hypertension combined with depression. Methods using metabolomics and machine learning performed quite well in predicting depression in hypertensive older adults, contributing to further clinical research.

Metabolic profiling during COVID-19 infection in humans: Identification of potential biomarkers for occurrence, severity and outcomes using machine learning

BACKGROUND

After its emergence in China, the coronavirus SARS-CoV-2 has swept the world, leading to global health crises with millions of deaths. COVID-19 clinical manifestations differ in severity, ranging from mild symptoms to severe disease. Although perturbation of metabolism has been reported as a part of the host response to COVID-19 infection, scarce data exist that describe stage-specific changes in host metabolites during the infection and how this could stratify patients based on severity.

METHODS

Given this knowledge gap, we performed targeted metabolomics profiling and then used machine learning models and biostatistics to characterize the alteration patterns of 50 metabolites and 17 blood parameters measured in a cohort of 295 human subjects. They were categorized into healthy controls, non-severe, severe and critical groups with their outcomes. Subject's demographic and clinical data were also used in the analyses to provide more robust predictive models.

RESULTS

The non-severe and severe COVID-19 patients experienced the strongest changes in metabolite repertoire, whereas less intense changes occur during the critical phase. Panels of 15, 14, 2 and 2 key metabolites were identified as predictors for non-severe, severe, critical and dead patients, respectively. Specifically, arginine and malonyl methylmalonyl succinylcarnitine were significant biomarkers for the onset of COVID-19 infection and tauroursodeoxycholic acid were potential biomarkers for disease progression. Measuring blood parameters enhanced the predictive power of metabolic signatures during critical illness.

CONCLUSIONS

Metabolomic signatures are distinctive for each stage of COVID-19 infection. This has great translation potential as it opens new therapeutic and diagnostic prospective based on key metabolites.

Paper spray mass spectrometry combined with machine learning as a rapid diagnostic for chronic kidney disease

A new analytical method for chronic kidney disease (CKD) detection utilizing paper spray mass spectrometry (PS-MS) combined with machine learning is presented. The analytical protocol is rapid and simple, based on metabolic profile alterations in urine. Anonymized raw urine samples were deposited (10 μL each) onto pointed PS-MS sample strips. Without waiting for the sample to dry, 75 μL of acetonitrile and high voltage were applied to the strips, using high resolution mass spectrometry measurement (15 s per sample) with polarity switching to detect a wide range of metabolites. Random forest machine learning was used to classify the resulting data. The diagnostic performance for the potential diagnosis of CKD was evaluated for accuracy, sensitivity, and specificity, achieving results >96% for the training data and >91% for validation and test data sets. Metabolites selected by the classification model as up- or down-regulated in healthy or CKD samples were tentatively identified and in agreement with previously reported literature. The potential utilization of this approach to discriminate albuminuria categories (normo, micro, and macroalbuminuria) was also demonstrated. This study indicates that PS-MS combined with machine learning has the potential to be used as a rapid and simple diagnostic tool for CKD.

Prediction of Clinical Remission with Adalimumab Therapy in Patients with Ulcerative Colitis by Fourier Transform-Infrared Spectroscopy Coupled with Machine Learning Algorithms

We aimed to develop prediction models for clinical remission associated with adalimumab treatment in patients with ulcerative colitis (UC) using Fourier transform-infrared (FT-IR) spectroscopy coupled with machine learning (ML) algorithms. This prospective, observational, multicenter study enrolled 62 UC patients and 30 healthy controls. The patients were treated with adalimumab for 56 weeks, and clinical remission was evaluated using the Mayo score. Baseline fecal samples were collected and analyzed using FT-IR spectroscopy. Various data preprocessing methods were applied, and prediction models were established by 10-fold cross-validation using various ML methods. Orthogonal partial least squares-discriminant analysis (OPLS-DA) showed a clear separation of healthy controls and UC patients, applying area normalization and Pareto scaling. OPLS-DA models predicting short- and long-term remission (8 and 56 weeks) yielded area-under-the-curve values of 0.76 and 0.75, respectively. Logistic regression and a nonlinear support vector machine were selected as the best prediction models for short- and long-term remission, respectively (accuracy of 0.99). In external validation, prediction models for short-term (logistic regression) and long-term (decision tree) remission performed well, with accuracy values of 0.73 and 0.82, respectively. This was the first study to develop prediction models for clinical remission associated with adalimumab treatment in UC patients by fecal analysis using FT-IR spectroscopy coupled with ML algorithms. Logistic regression, nonlinear support vector machines, and decision tree were suggested as the optimal prediction models for remission, and these were noninvasive, simple, inexpensive, and fast analyses that could be applied to personalized treatments.

Machine learning-based analysis of risk factors for chronic total occlusion in an Asian population

OBJECTIVES

Chronic total occlusion (CTO) is a form of coronary artery disease (CAD) requiring percutaneous coronary intervention. There has been minimal research regarding CTO-specific risk factors and predictive models. We developed machine learning predictive models based on clinical characteristics to identify patients with CTO before coronary angiography.

METHODS

Data from 1473 patients with CAD, including 317 patients with and 1156 patients without CTO, were retrospectively analyzed. Partial least squares discriminant analysis (PLS-DA), random forest (RF), and support vector machine (SVM) models were used to identify CTO-specific risk factors and predict CTO development. Receiver operating characteristic (ROC) curve analysis was performed for model validation.

RESULTS

For CTO prediction, the PLS-DA model included 10 variables; the ROC value was 0.706. The RF model included 42 variables; the ROC value was 0.702. The SVM model included 20 variables; the ROC value was 0.696. DeLong's test showed no difference among the three models. Four variables were present in all models: sex, neutrophil percentage, creatinine, and brain natriuretic peptide (BNP).

CONCLUSIONS

Validation of machine learning prediction models for CTO revealed that the PLS-DA model had the best prediction performance. Sex, neutrophil percentage, creatinine, and BNP may be important risk factors for CTO development.

Anti-Inflammatory Activity of Lauraceae Plant Species and Prediction Models Based on Their Metabolomics Profiling Data

The Lauraceae is a botanical family known for its anti-inflammatory potential. However, several species have not yet been studied. Thus, this work aimed to screen the anti-inflammatory activity of this plant family and to build statistical prediction models. The methodology was based on the statistical analysis of high-resolution liquid chromatography coupled with mass spectrometry data and the ex vivo anti-inflammatory activity of plant extracts. The ex vivo results demonstrated significant anti-inflammatory activity for several of these plants for the first time. The sample data were applied to build anti-inflammatory activity prediction models, including the partial least square acquired, artificial neural network, and stochastic gradient descent, which showed adequate fitting and predictive performance. Key anti-inflammatory markers, such as aporphine and benzylisoquinoline alkaloids were annotated with confidence level 2. Additionally, the validated prediction models proved to be useful for predicting active extracts using metabolomics data and studying their most bioactive metabolites.

Risk Factor Analysis and Multiple Predictive Machine Learning Models for Mortality in COVID-19: A Multicenter and Multi-Ethnic Cohort Study

BACKGROUND

The COVID-19 pandemic presents a significant challenge to the global health care system. Implementing timely, accurate, and cost-effective screening approaches is crucial in preventing infections and saving lives by guiding disease management.

OBJECTIVES

The study aimed to use machine learning algorithms to analyze clinical features from routine clinical data to identify risk factors and predict the mortality of COVID-19.

METHODS

The data used in this research were originally collected for the study titled "Neurologic Syndromes Predict Higher In-Hospital Mortality in COVID-19." A total of 4711 patients with confirmed COVID-19 were enrolled consecutively from four hospitals. Three machine learning models, including random forest (RF), partial least squares discriminant analysis (PLS-DA), and support vector machine (SVM), were used to find risk factors and predict COVID-19 mortality.

RESULTS

The predictive models were developed based on three machine learning algorithms. The RF model was trained with 20 variables and had a receiver operating characteristic (ROC) value of 0.859 (95% confidence interval [CI] 0.804-0.920). The PLS-DA model was trained with 20 variables and had a ROC value of 0.775 (95% CI 0.694-0.833). The SVM model was trained with 10 variables and had a ROC value of 0.828 (95% CI 0.785-0.865). The nine variables that were present in all three models were age, procalcitonin, ferritin, C-reactive protein, troponin, blood urea nitrogen, mean arterial pressure, aspartate transaminase, and alanine transaminase.

CONCLUSION

This study developed and validated three machine learning prediction models for COVID-19 mortality based on accessible clinical features. The RF model showed the best performance among the three models. The nine variables identified in the models may warrant further investigation as potential prognostic indicators of severe COVID-19.

Using Multivariate Data Analysis to Project Performance in Biathletes and Cross-Country Skiers

PURPOSE

To determine whether competitive performance, as defined by International Biathlon Union (IBU) and International Ski Federation (FIS) points in biathlon and cross-country (XC) skiing, respectively, can be projected using a combination of anthropometric and physiological metrics. Shooting accuracy was also included in the biathlon models.

METHODS

Data were analyzed using multivariate methods from 45 (23 female and 22 male) biathletes and 202 (86 female and 116 male) XC skiers who were all members of senior national teams, national development teams, or ski-university or high school invite-only programs (age range: 16-36 y). Anthropometric and physiological characteristics were assessed via dual-energy X-ray absorptiometry and incremental roller-ski treadmill tests, respectively. Shooting accuracy was assessed via an outdoor standardized testing protocol.

RESULTS

Valid projective models were identified for female biathletes' IBU points (R2 = .80/Q2 = .65) and female XC skiers' FIS distance (R2 = .81/Q2 = .74) and sprint (R2 = .81/Q2 = .70) points. No valid models were identified for the men. The most important variables for the projection of IBU points were shooting accuracy, speeds at blood lactate concentrations of 4 and 2 mmol·L-1, peak aerobic power, and lean mass. The most important variables for the projection of FIS distance and sprint points were speeds at blood lactate concentrations of 4 and 2 mmol·L-1 and peak aerobic power.

CONCLUSIONS

This study highlights the relative importance of specific anthropometric, physiological, and shooting-accuracy metrics in female biathletes and XC skiers. The data can help to identify the specific metrics that should be targeted when monitoring athletes' progression and designing training plans.

Metabolomics in viral hepatitis: advances and review

Viral hepatitis is a major worldwide public health issue, affecting hundreds of millions of people and causing substantial morbidity and mortality. The majority of the worldwide burden of viral hepatitis is caused by five biologically unrelated hepatotropic viruses: hepatitis A virus (HAV), hepatitis B virus (HBV), hepatitis C virus (HCV), hepatitis D virus (HDV), and hepatitis E virus (HEV). Metabolomics is an emerging technology that uses qualitative and quantitative analysis of easily accessible samples to provide information of the metabolic levels of biological systems and changes in metabolic and related regulatory pathways. Alterations in glucose, lipid, and amino acid levels are involved in glycolysis, the tricarboxylic acid cycle, the pentose phosphate pathway, and amino acid metabolism. These changes in metabolites and metabolic pathways are associated with the pathogenesis and medication mechanism of viral hepatitis and related diseases. Additionally, differential metabolites can be utilized as biomarkers for diagnosis, prognosis, and therapeutic responses. In this review, we present a thorough overview of developments in metabolomics for viral hepatitis.

New Diagnostic Modality Combining Mass Spectrometry and Machine Learning for the Discrimination of Malignant Intraductal Papillary Mucinous Neoplasms

BACKGROUND

An intraductal papillary mucinous neoplasm (IPMN) is a pancreatic tumor with malignant potential. Although we anticipate a sensitive method to diagnose the malignant conversion of IPMN, an effective strategy has not yet been established. The combination of probe electrospray ionization-mass spectrometry (PESI-MS) and machine learning provides a promising solution for this purpose.

METHODS

We prospectively analyzed 42 serum samples obtained from IPMN patients who underwent pancreatic resection between 2020 and 2021. Based on the postoperative pathological diagnosis, patients were classified into two groups: IPMN-low grade dysplasia (n = 17) and advanced-IPMN (n = 25). Serum samples were analyzed by PESI-MS, and the obtained mass spectral data were converted into continuous variables. These variables were used to discriminate advanced-IPMN from IPMN-low grade dysplasia by partial least square regression or support vector machine analysis. The areas under receiver operating characteristics curves were obtained to visualize the difference between the two groups.

RESULTS

Partial least square regression successfully discriminated the two disease classes. From another standpoint, we selected 130 parameters from the entire dataset by PESI-MS, which were fed into the support vector machine. The diagnostic accuracy was 88.1%, and the area under the receiver operating characteristics curve was 0.924 by this method. Approximately 10 min were required to perform each method.

CONCLUSION

PESI-MS combined with machine learning is an easy-to-use tool with the advantage of rapid on-site analysis. Here, we show the great potential of our system to diagnose the malignant conversion of IPMN, which would be a promising diagnostic tool in clinical settings.

Advances in Microbial NMR Metabolomics

Confidently, nuclear magnetic resonance (NMR) is the most informative technique in analytical chemistry and its use as an analytical platform in metabolomics is well proven. This chapter aims to present NMR as a viable tool for microbial metabolomics discussing its fundamental aspects and applications in metabolomics using some chosen examples.

Discovering New Natural Products Using Metabolomics-Based Approaches

The incessant search for new natural molecules with biological activities has forced researchers in the field of chemistry of natural products to seek different approaches for their prospection studies. In particular, researchers around the world are turning to approaches in metabolomics to avoid high rates of re-isolation of certain compounds, something recurrent in this branch of science. Thanks to the development of new technologies in the analytical instrumentation of spectroscopic and spectrometric techniques, as well as the advance in the computational processing modes of the results, metabolomics has been gaining more and more space in studies that involve the prospection of natural products. Thus, this chapter summarizes the precepts and good practices in the metabolomics of microbial natural products using mass spectrometry and nuclear magnetic resonance spectroscopy, and also summarizes several examples where this approach has been applied in the discovery of bioactive molecules.

Toxicology assessment of manganese oxide nanomaterials with enhanced electrochemical properties using human in vitro models representing different exposure routes

In the present study, a comparative human toxicity assessment between newly developed Mn₃O₄ nanoparticles with enhanced electrochemical properties (GNA35) and their precursor material (Mn₃O₄) was performed, employing different in vitro cellular models representing main exposure routes (inhalation, intestinal and dermal contact), namely the human alveolar carcinoma epithelial cell line (A549), the human colorectal adenocarcinoma cell line (HT29), and the reconstructed 3D human epidermal model EpiDerm. The obtained results showed that Mn₃O₄ and GNA35 harbour similar morphological characteristics, whereas differences were observed in relation to their surface area and electrochemical properties. In regard to their toxicological properties, both nanomaterials induced ROS in the A549 and HT29 cell lines, while cell viability reduction was only observed in the A549 cells. Concerning their skin irritation potential, the studied nanomaterials did not cause a reduction of the skin tissue viability in the test conditions nor interleukin 1 alpha (IL- 1 α) release. Therefore, they can be considered as not irritant nanomaterials according to EU and Globally Harmonized System of Classification and Labelling Chemicals. Our findings provide new insights about the potential harmful effects of Mn₃O₄ nanomaterials with different properties, demonstrating that the hazard assessment using different human in vitro models is a critical aspect to increase the knowledge on their potential impact upon different exposure routes.

Label compliance for ingredient verification: regulations, approaches, and trends for testing botanical products marketed for "immune health" in the United States

During the COVID-19 pandemic, the botanical product market saw a consumer interest increase in immune health supplements. While data are currently insufficient to support public health guidance for using foods and dietary supplements to prevent or treat COVID-19 and other immune disorders, consumer surveys indicate that immune support is the second-most cited reason for supplement use in the United States. Meanwhile, consumers showed increased attention to dietary supplement ingredient labels, especially concerning authenticity and ingredient claims. Top-selling botanical ingredients such as elderberry, turmeric, and functional mushrooms have been increasingly marketed toward consumers to promote immune health, but these popular products succumb to adulteration with inaccurate labeling due to the intentional or unintentional addition of lower grade ingredients, non-target plants, and synthetic compounds, partially due to pandemic-related supply chain issues. This review highlights the regulatory requirements and recommendations for analytical approaches, including chromatography, spectroscopy, and DNA approaches for ingredient claim verification. Demonstrating elderberry, turmeric, and functional mushrooms as examples, this review aims to provide industrial professionals and scientists an overview of current United States regulations, testing approaches, and trends for label compliance verification to ensure the safety of botanical products marketed for "immune health."

Metabolic, proteomic and microbial changes postmortem and during beef aging

The purpose of this review is to provide an overview of the current knowledge about proteomic and metabolic changes in beef, the microbiological alteration postmortem and during aging, and observe the influence on beef quality parameters, such as tenderness, taste and flavor. This review will also focus on the different aging types (wet- and dry-aging), the aging or postmortem time of beef and their effect on the proteome and metabolome of beef. The Ca2+ homeostasis and adenosine 5'-triphosphate breakdown are the main reactions in the pre-rigor phase. After rigor mortis, the enzymatic degradation of connective tissues and breakdown of energy metabolism dominate molecular changes in beef. Important metabolic processes leading to the formation of saccharides, nucleotides, organic acids (e.g. lactic acid), creatine and fatty acids are considered in this context as possible flavor precursors or formers of beef flavor and taste. Flavor precursors are substrates for lipid oxidation, Strecker degradation and Maillard reaction during cooking or roasting. The findings presented should serve as a basis for a better understanding of beef aging and its molecular effects and are intended to contribute to meeting the challenges of improving beef quality.

Approaches for assessing performance of high-resolution mass spectrometry-based non-targeted analysis methods

Non-targeted analysis (NTA) using high-resolution mass spectrometry has enabled the detection and identification of unknown and unexpected compounds of interest in a wide range of sample matrices. Despite these benefits of NTA methods, standardized procedures do not yet exist for assessing performance, limiting stakeholders’ abilities to suitably interpret and utilize NTA results. Herein, we first summarize existing performance assessment metrics for targeted analyses to provide context and clarify terminology that may be shared between targeted and NTA methods (e.g., terms such as accuracy, precision, sensitivity, and selectivity). We then discuss promising approaches for assessing NTA method performance, listing strengths and key caveats for each approach, and highlighting areas in need of further development. To structure the discussion, we define three types of NTA study objectives: sample classification, chemical identification, and chemical quantitation. Qualitative study performance (i.e., focusing on sample classification and/or chemical identification) can be assessed using the traditional confusion matrix, with some challenges and limitations. Quantitative study performance can be assessed using estimation procedures developed for targeted methods with consideration for additional sources of uncontrolled experimental error. This article is intended to stimulate discussion and further efforts to develop and improve procedures for assessing NTA method performance. Ultimately, improved performance assessments will enable accurate communication and effective utilization of NTA results by stakeholders.

Unraveling the Rewired Metabolism in Lung Cancer Using Quantitative NMR Metabolomics

Lung cancer cells are well documented to rewire their metabolism and energy production networks to enable proliferation and survival in a nutrient-poor and hypoxic environment. Although metabolite profiling of blood plasma and tissue is still emerging in omics approaches, several techniques have shown potential in cancer diagnosis. In this paper, the authors describe the alterations in the metabolic phenotype of lung cancer patients. In addition, we focus on the metabolic cooperation between tumor cells and healthy tissue. Furthermore, the authors discuss how metabolomics could improve the management of lung cancer patients.

NMR in Metabolomics: From Conventional Statistics to Machine Learning and Neural Network Approaches

NMR measurements combined with chemometrics allow achieving a great amount of information for the identification of potential biomarkers responsible for a precise metabolic pathway. These kinds of data are useful in different fields, ranging from food to biomedical fields, including health science. The investigation of the whole set of metabolites in a sample, representing its fingerprint in the considered condition, is known as metabolomics and may take advantage of different statistical tools. The new frontier is to adopt self-learning techniques to enhance clustering or classification actions that can improve the predictive power over large amounts of data. Although machine learning is already employed in metabolomics, deep learning and artificial neural networks approaches were only recently successfully applied. In this work, we give an overview of the statistical approaches underlying the wide range of opportunities that machine learning and neural networks allow to perform with accurate metabolites assignment and quantification.Various actual challenges are discussed, such as proper metabolomics, deep learning architectures and model accuracy.

Biomarker selection and a prospective metabolite-based machine learning diagnostic for lyme disease

We provide a pipeline for data preprocessing, biomarker selection, and classification of liquid chromatography–mass spectrometry (LCMS) serum samples to generate a prospective diagnostic test for Lyme disease. We utilize tools of machine learning (ML), e.g., sparse support vector machines (SSVM), iterative feature removal (IFR), and k-fold feature ranking to select several biomarkers and build a discriminant model for Lyme disease. We report a 98.13% test balanced success rate (BSR) of our model based on a sequestered test set of LCMS serum samples. The methodology employed is general and can be readily adapted to other LCMS, or metabolomics, data sets.

Current State of Metabolomics Research in Meat Quality Analysis and Authentication

In the past decades, as an emerging omic, metabolomics has been widely used in meat science research, showing promise in meat quality analysis and meat authentication. This review first provides a brief overview of the concept, analytical techniques, and analysis workflow of metabolomics. Additionally, the metabolomics research in quality analysis and authentication of meat is comprehensively described. Finally, the limitations, challenges, and future trends of metabolomics application in meat quality analysis and meat authentication are critically discussed. We hope to provide valuable insights for further research in meat quality.

Global metabolome profiling of exhaled breath condensates in male smokers with asthma COPD overlap and prediction of the disease

Asthma-chronic obstructive pulmonary disease (COPD) overlap, termed as ACO, is a complex heterogeneous disease characterised by persistent airflow limitation, which manifests features of both asthma and COPD. These patients have a worse prognosis, in terms of more frequent and severe exacerbations, more frequent symptoms, worse quality of life, increased comorbidities and a faster lung function decline. In absence of clear diagnostic or therapeutic guidelines, ACO presents as a challenge to clinicians. The present study aims to investigate whether ACO patients have a distinct exhaled breath condensate (EBC) metabolic profile in comparison to asthma and COPD. A total of 132 age and BMI matched male smokers were recruited in the exploratory phase which consisted of (i) controls = 33 (ii) asthma = 34 (iii) COPD = 30 and (iv) ACO = 35. Using nuclear magnetic resonance (NMR) metabolomics, 8 metabolites (fatty acid, propionate, isopropanol, lactate, acetone, valine, methanol and formate) were identified to be significantly dysregulated in ACO subjects when compared to both, asthma and COPD. The expression of these dysregulated metabolites were further validated in a fresh patient cohort consisting of (i) asthma = 32 (ii) COPD = 32 and (iii) ACO = 40, which exhibited a similar expression pattern. Multivariate receiver operating characteristic (ROC) curves generated using these metabolites provided a robust ACO classification model. The findings were also integrated with previously identified serum metabolites and inflammatory markers to develop a robust predictive model for differentiation of ACO. Our findings suggest that NMR metabolomics of EBC holds potential as a platform to identify robust, non-invasive biomarkers for differentiating ACO from asthma and COPD.

DeepSSPred: A Deep Learning Based Sulfenylation Site Predictor Via a Novel nSegmented Optimize Federated Feature Encoder

BACKGROUND

S-sulfenylation (S-sulphenylation, or sulfenic acid) proteins, are special kinds of post-translation modification, which plays an important role in various physiological and pathological processes such as cytokine signaling, transcriptional regulation, and apoptosis. Despite these aforementioned significances, and by complementing existing wet methods, several computational models have been developed for sulfenylation cysteine sites prediction. However, the performance of these models was not satisfactory due to inefficient feature schemes, severe imbalance issues, and lack of an intelligent learning engine.

OBJECTIVE

In this study, our motivation is to establish a strong and novel computational predictor for discrimination of sulfenylation and non-sulfenylation sites.

METHODS

In this study, we report an innovative bioinformatics feature encoding tool, named DeepSSPred, in which, resulting encoded features is obtained via nSegmented hybrid feature, and then the resampling technique called synthetic minority oversampling was employed to cope with the severe imbalance issue between SC-sites (minority class) and non-SC sites (majority class). State of the art 2D-Convolutional Neural Network was employed over rigorous 10-fold jackknife cross-validation technique for model validation and authentication.

RESULTS

Following the proposed framework, with a strong discrete presentation of feature space, machine learning engine, and unbiased presentation of the underline training data yielded into an excellent model that outperforms with all existing established studies. The proposed approach is 6% higher in terms of MCC from the first best. On an independent dataset, the existing first best study failed to provide sufficient details. The model obtained an increase of 7.5% in accuracy, 1.22% in Sn, 12.91% in Sp and 13.12% in MCC on the training data and12.13% of ACC, 27.25% in Sn, 2.25% in Sp, and 30.37% in MCC on an independent dataset in comparison with 2nd best method. These empirical analyses show the superlative performance of the proposed model over both training and Independent dataset in comparison with existing literature studies.

CONCLUSION

In this research, we have developed a novel sequence-based automated predictor for SC-sites, called DeepSSPred. The empirical simulations outcomes with a training dataset and independent validation dataset have revealed the efficacy of the proposed theoretical model. The good performance of DeepSSPred is due to several reasons, such as novel discriminative feature encoding schemes, SMOTE technique, and careful construction of the prediction model through the tuned 2D-CNN classifier. We believe that our research work will provide a potential insight into a further prediction of S-sulfenylation characteristics and functionalities. Thus, we hope that our developed predictor will significantly helpful for large scale discrimination of unknown SC-sites in particular and designing new pharmaceutical drugs in general.

Early Breast Cancer Detection Using Untargeted and Targeted Metabolomics

Breast cancer (BC) is a common cause of morbidity and mortality, particularly in women. Moreover, the discovery of diagnostic biomarkers for early BC remains a challenging task. Previously, we [Jasbi et al. J. Chromatogr. B. 2019, 1105, 26-37] demonstrated a targeted metabolic profiling approach capable of identifying metabolite marker candidates that could enable highly sensitive and specific detection of BC. However, the coverage of this targeted method was limited and exhibited suboptimal classification of early BC (EBC). To expand the metabolome coverage and articulate a better panel of metabolites or mass spectral features for classification of EBC, we evaluated untargeted liquid chromatography quadrupole time-of-flight mass spectrometry (LC-QTOF-MS) data, both individually as well as in conjunction with previously published targeted LC-triple quadruple (QQQ)-MS data. Variable importance in projection scores were used to refine the biomarker panel, whereas orthogonal partial least squares-discriminant analysis was used to operationalize the enhanced biomarker panel for early diagnosis. In this approach, 33 altered metabolites/features were detected by LC-QTOF-MS from 124 BC patients and 86 healthy controls. For EBC diagnosis, significance testing and analysis of the area under receiver operating characteristic (AUROC) curve identified six metabolites/features [ethyl (R)-3-hydroxyhexanoate; caprylic acid; hypoxanthine; and m/z 358.0018, 354.0053, and 356.0037] with p < 0.05 and AUROC > 0.7. These metabolites informed the construction of EBC diagnostic models; evaluation of model performance for the prediction of EBC showed an AUROC = 0.938 (95% CI: 0.895-0.975), with sensitivity = 0.90 when specificity = 0.90. Using the combined untargeted and targeted data set, eight metabolic pathways of potential biological relevance were indicated to be significantly altered as a result of EBC. Metabolic pathway analysis showed fatty acid and aminoacyl-tRNA biosynthesis as well as inositol phosphate metabolism to be most impacted in response to the disease. The combination of untargeted and targeted metabolomics platforms has provided a highly predictive and accurate method for BC and EBC diagnosis from plasma samples. Furthermore, such a complementary approach yielded critical information regarding potential pathogenic mechanisms underlying EBC that, although critical to improved prognosis and enhanced survival, are understudied in the current literature. All mass spectrometry data and deidentified subject metadata analyzed in this study have been deposited to Mendeley Data and are publicly available (DOI: 10.17632/kcjg8ybk45.1).

Metabolomics analysis of maternal serum exposed to high air pollution during pregnancy and risk of autism spectrum disorder in offspring

BACKGROUND

Previously, numerous epidemiologic studies reported an association between autism spectrum disorder (ASD) and exposure to air pollution during pregnancy. However, there have been no metabolomics studies investigating the impact of pregnancy pollution exposure to ASD risk in offspring.

OBJECTIVES

To identify differences in maternal metabolism that may reflect a biological response to exposure to high air pollution in pregnancies of offspring who later did or did not develop ASD.

METHODS

We obtained stored mid-pregnancy serum from 214 mothers who lived in California's Central Valley and experienced the highest levels of air pollution during early pregnancy. We estimated each woman's average traffic-related air pollution exposure (carbon monoxide, nitric oxides, and particulate matter <2.5 μm) during the first trimester using the California Line Source Dispersion Model, version 4 (CALINE4). By utilizing liquid chromatography-high resolution mass spectrometry, we identified the metabolic profiles of maternal serum for 116 mothers with offspring who later developed ASD and 98 control mothers. Partial least squares discriminant analysis (PLS-DA) was employed to select metabolic features associated with air pollution exposure or autism risk in offspring. We also conducted extensive pathway enrichment analysis to elucidate potential ASD-related changes in the metabolome of pregnant women.

RESULTS

We extracted 4022 and 4945 metabolic features from maternal serum samples in hydrophilic interaction (HILIC) chromatography (positive ion mode) and C18 (negative ion mode) columns, respectively. After controlling for potential confounders, we identified 167 and 222 discriminative features (HILIC and C18, respectively). Pathway enrichment analysis to discriminate metabolic features associated with ASD risk indicated various metabolic pathway perturbations linked to the tricarboxylic acid (TCA) cycle and mitochondrial function, including carnitine shuttle, amino acid metabolism, bile acid metabolism, and vitamin A metabolism.

CONCLUSION

Using high resolution metabolomics, we identified several metabolic pathways disturbed in mothers with ASD offspring among women experiencing high exposure to traffic-related air pollution during pregnancy that were associated with mitochondrial dysfunction. These findings provide us with a better understanding of metabolic disturbances involved in the development of ASD under adverse environmental conditions.

Lilikoi V2.0: a deep learning-enabled, personalized pathway-based R package for diagnosis and prognosis predictions using metabolomics data

BACKGROUND

previously we developed Lilikoi, a personalized pathway-based method to classify diseases using metabolomics data. Given the new trends of computation in the metabolomics field, it is important to update Lilikoi software.

RESULTS

here we report the next version of Lilikoi as a significant upgrade. The new Lilikoi v2.0 R package has implemented a deep learning method for classification, in addition to popular machine learning methods. It also has several new modules, including the most significant addition of prognosis prediction, implemented by Cox-proportional hazards model and the deep learning-based Cox-nnet model. Additionally, Lilikoi v2.0 supports data preprocessing, exploratory analysis, pathway visualization, and metabolite pathway regression.

CONCULSION

Lilikoi v2.0 is a modern, comprehensive package to enable metabolomics analysis in R programming environment.

Persistent metabolomic alterations characterize chronic critical illness after severe trauma

BACKGROUND

Following trauma, persistent inflammation, immunosuppression, and catabolism may characterize delayed recovery or failure to recover. Understanding the metabolic response associated with these adverse outcomes may facilitate earlier identification and intervention. We characterized the metabolic profiles of trauma victims who died or developed chronic critical illness (CCI) and hypothesized that differences would be evident within 1-week postinjury.

METHODS

Venous blood samples from trauma victims with shock who survived at least 7 days were analyzed using mass spectrometry. Subjects who died or developed CCI (intensive care unit length of stay of ≥14 days with persistent organ dysfunction) were compared with subjects who recovered rapidly (intensive care unit length of stay, ≤7 days) and uninjured controls. We used partial least squares discriminant analysis, t tests, linear mixed effects regression, and pathway enrichment analyses to make broad comparisons and identify differences in metabolite concentrations and pathways.

RESULTS

We identified 27 patients who died or developed CCI and 33 who recovered rapidly. Subjects were predominantly male (65%) with a median age of 53 years and Injury Severity Score of 36. Healthy controls (n = 48) had similar age and sex distributions. Overall, from the 163 metabolites detected in the samples, 56 metabolites and 21 pathways differed between injury outcome groups, and partial least squares discriminant analysis models distinguished injury outcome groups as early as 1-day postinjury. Differences were observed in tryptophan, phenylalanine, and tyrosine metabolism; metabolites associated with oxidative stress via methionine metabolism; inflammatory mediators including kynurenine, arachidonate, and glucuronic acid; and products of the gut microbiome including indole-3-propionate.

CONCLUSIONS

The metabolic profiles in subjects who ultimately die or develop CCI differ from those who have recovered. In particular, we have identified differences in markers of inflammation, oxidative stress, amino acid metabolism, and alterations in the gut microbiome. Targeted metabolomics has the potential to identify important metabolic changes postinjury to improve early diagnosis and targeted intervention.

LEVEL OF EVIDENCE

Prognostic/epidemiologic, level III.