Machine learning of plasma metabolome identifies biomarker panels for metabolic syndrome: findings from the China Suboptimal Health Cohort

Artificial Intelligence in Metabolomics: A Current Review

Metabolomics and artificial intelligence (AI) form a synergistic partnership. Metabolomics generates large datasets comprising hundreds to thousands of metabolites with complex relationships. AI, aiming to mimic human intelligence through computational modeling, possesses extraordinary capabilities for big data analysis. In this review, we provide a recent overview of the methodologies and applications of AI in metabolomics studies in the context of systems biology and human health. We first introduce the AI concept, history, and key algorithms for machine learning and deep learning, summarizing their strengths and weaknesses. We then discuss studies that have successfully used AI across different aspects of metabolomic analysis, including analytical detection, data preprocessing, biomarker discovery, predictive modeling, and multi-omics data integration. Lastly, we discuss the existing challenges and future perspectives in this rapidly evolving field. Despite limitations and challenges, the combination of metabolomics and AI holds great promises for revolutionary advancements in enhancing human health.

Novel Metabolites Associated With Blood Pressure After Dietary Interventions

BACKGROUND

The blood pressure (BP) etiologic study is complex due to multifactorial influences, including genetic, environmental, lifestyle, and their intricate interplays. We used a metabolomics approach to capture internal pathways and external exposures and to study BP regulation mechanisms after well-controlled dietary interventions.

METHODS

In the ProBP trail (Protein and Blood Pressure), a double-blinded crossover randomized controlled trial, participants underwent dietary interventions of carbohydrate, soy protein, and milk protein, receiving 40 g daily for 8 weeks, with 3-week washout periods. We measured plasma samples collected at baseline and at the end of each dietary intervention. Multivariate linear models were used to evaluate the association between metabolites and systolic/diastolic BP. Nominally significant metabolites were examined for enriching biological pathways. Significant ProBP findings were evaluated for replication among 1311 participants of the BHS (Bogalusa Heart Study), a population-based study conducted in the same area as ProBP.

RESULTS

After Bonferroni correction for 77 independent metabolite clusters (α=6.49×10-4), 18 metabolites were significantly associated with BP at baseline or the end of a dietary intervention, of which 11 were replicated in BHS. Seven emerged as novel discoveries, which are as follows: 1-linoleoyl-GPE (18:2), 1-oleoyl-GPE (18:1), 1-stearoyl-2-linoleoyl-GPC (18:0/18:2), 1-palmitoyl-2-oleoyl-GPE (16:0/18:1), maltose, N-stearoyl-sphinganine (d18:0/18:0), and N6-carbamoylthreonyladenosine. Pathway enrichment analyses suggested dietary protein intervention might reduce BP through pathways related to G protein-coupled receptors, incretin function, selenium micronutrient network, and mitochondrial biogenesis.

CONCLUSIONS

Seven novel metabolites were identified to be associated with BP at the end of different dietary interventions. The beneficial effects of protein interventions might be mediated through specific metabolic pathways.

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.

Strategy for Identifying a Robust Metabolomic Signature Reveals the Altered Lipid Metabolism in Pituitary Adenoma

Despite the well-established connection between systematic metabolic abnormalities and the pathophysiology of pituitary adenoma (PA), current metabolomic studies have reported an extremely limited number of metabolites associated with PA. Moreover, there was very little consistency in the identified metabolite signatures, resulting in a lack of robust metabolic biomarkers for the diagnosis and treatment of PA. Herein, we performed a global untargeted plasma metabolomic profiling on PA and identified a highly robust metabolomic signature based on a strategy. Specifically, this strategy is unique in (1) integrating repeated random sampling and a consensus evaluation-based feature selection algorithm and (2) evaluating the consistency of metabolomic signatures among different sample groups. This strategy demonstrated superior robustness and stronger discriminative ability compared with that of other feature selection methods including Student's t-test, partial least-squares-discriminant analysis, support vector machine recursive feature elimination, and random forest recursive feature elimination. More importantly, a highly robust metabolomic signature comprising 45 PA-specific differential metabolites was identified. Moreover, metabolite set enrichment analysis of these potential metabolic biomarkers revealed altered lipid metabolism in PA. In conclusion, our findings contribute to a better understanding of the metabolic changes in PA and may have implications for the development of diagnostic and therapeutic approaches targeting lipid metabolism in PA. We believe that the proposed strategy serves as a valuable tool for screening robust, discriminating metabolic features in the field of metabolomics.

Blood-Derived Lipid and Metabolite Biomarkers in Cardiovascular Research from Clinical Studies: A Recent Update

The primary prevention, early detection, and treatment of cardiovascular disease (CVD) have been long-standing scientific research goals worldwide. In the past decades, traditional blood lipid profiles have been routinely used in clinical practice to estimate the risk of CVDs such as atherosclerotic cardiovascular disease (ASCVD) and as treatment targets for the primary prevention of adverse cardiac events. These blood lipid panel tests often fail to fully predict all CVD risks and thus need to be improved. A comprehensive analysis of molecular species of lipids and metabolites (defined as lipidomics and metabolomics, respectively) can provide molecular insights into the pathophysiology of the disease and could serve as diagnostic and prognostic indicators of disease. Mass spectrometry (MS) and nuclear magnetic resonance (NMR)-based lipidomics and metabolomics analysis have been increasingly used to study the metabolic changes that occur during CVD pathogenesis. In this review, we provide an overview of various MS-based platforms and approaches that are commonly used in lipidomics and metabolomics workflows. This review summarizes the lipids and metabolites in human plasma/serum that have recently (from 2018 to December 2022) been identified as promising CVD biomarkers. In addition, this review describes the potential pathophysiological mechanisms associated with candidate CVD biomarkers. Future studies focused on these potential biomarkers and pathways will provide mechanistic clues of CVD pathogenesis and thus help with the risk assessment, diagnosis, and treatment of CVD.

Identifying Metabolic Syndrome Easily and Cost Effectively Using Non-Invasive Methods with Machine Learning Models

Purpose

The objective of this study was to employ machine learning (ML) models utilizing non-invasive factors to achieve early and low-cost identification of MetS in a large physical examination population.

Patients and Methods

The study enrolled 9171 participants who underwent physical examinations at Northern Jiangsu People's Hospital in 2009 and 2019, to determine MetS based on criteria established by the Chinese Diabetes Society. Non-invasive characteristics such as gender, age, body mass index (BMI), systolic blood pressure (SBP), and diastolic blood pressure (DBP) were collected and used as input variables to train and evaluate ML models for MetS identification. Several ML models were used for MetS identification, including logistic regression (LR), k-nearest neighbors algorithm (k-NN), naive bayesian (NB), decision tree (DT), random forest (RF), artificial neural network (ANN), and support vector machine (SVM).

Results

Our ML models all showed good performance in the 10-fold cross-validation except for the SVM model. In the external validation, the NB model exhibited the best performance with an AUC of 0.976, accuracy of 0.923, sensitivity of 98.32%, and specificity of 91.32%.

Conclusion

This study proposed a new non-invasive method for early and low-cost identification of MetS by using ML models. This approach has the potential to serve as a highly sensitive, convenient, and cost-effective tool for large-scale MetS screening.

Identification of candidate metabolite biomarkers for metabolic syndrome and its five components in population-based human cohorts

BACKGROUND

Metabolic Syndrome (MetS) is characterized by risk factors such as abdominal obesity, hypertriglyceridemia, low high-density lipoprotein cholesterol (HDL-C), hypertension, and hyperglycemia, which contribute to the development of cardiovascular disease and type 2 diabetes. Here, we aim to identify candidate metabolite biomarkers of MetS and its associated risk factors to better understand the complex interplay of underlying signaling pathways.

METHODS

We quantified serum samples of the KORA F4 study participants (N = 2815) and analyzed 121 metabolites. Multiple regression models adjusted for clinical and lifestyle covariates were used to identify metabolites that were Bonferroni significantly associated with MetS. These findings were replicated in the SHIP-TREND-0 study (N = 988) and further analyzed for the association of replicated metabolites with the five components of MetS. Database-driven networks of the identified metabolites and their interacting enzymes were also constructed.

RESULTS

We identified and replicated 56 MetS-specific metabolites: 13 were positively associated (e.g., Val, Leu/Ile, Phe, and Tyr), and 43 were negatively associated (e.g., Gly, Ser, and 40 lipids). Moreover, the majority (89%) and minority (23%) of MetS-specific metabolites were associated with low HDL-C and hypertension, respectively. One lipid, lysoPC a C18:2, was negatively associated with MetS and all of its five components, indicating that individuals with MetS and each of the risk factors had lower concentrations of lysoPC a C18:2 compared to corresponding controls. Our metabolic networks elucidated these observations by revealing impaired catabolism of branched-chain and aromatic amino acids, as well as accelerated Gly catabolism.

CONCLUSION

Our identified candidate metabolite biomarkers are associated with the pathophysiology of MetS and its risk factors. They could facilitate the development of therapeutic strategies to prevent type 2 diabetes and cardiovascular disease. For instance, elevated levels of lysoPC a C18:2 may protect MetS and its five risk components. More in-depth studies are necessary to determine the mechanism of key metabolites in the MetS pathophysiology.

Urinary 1H NMR Metabolomic Analysis of Prenatal Maternal Stress Due to a Natural Disaster Reveals Metabolic Risk Factors for Non-Communicable Diseases: The QF2011 Queensland Flood Study

Prenatal stress alters fetal programming, potentially predisposing the ensuing offspring to long-term adverse health outcomes. To gain insight into environmental influences on fetal development, this QF2011 study evaluated the urinary metabolomes of 4-year-old children (n = 89) who were exposed to the 2011 Queensland flood in utero. Proton nuclear magnetic resonance spectroscopy was used to analyze urinary metabolic fingerprints based on maternal levels of objective hardship and subjective distress resulting from the natural disaster. In both males and females, differences were observed between high and low levels of maternal objective hardship and maternal subjective distress groups. Greater prenatal stress exposure was associated with alterations in metabolites associated with protein synthesis, energy metabolism, and carbohydrate metabolism. These alterations suggest profound changes in oxidative and antioxidative pathways that may indicate a higher risk for chronic non-communicable diseases such obesity, insulin resistance, and diabetes, as well as mental illnesses, including depression and schizophrenia. Thus, prenatal stress-associated metabolic biomarkers may provide early predictors of lifetime health trajectories, and potentially serve as prognostic markers for therapeutic strategies in mitigating adverse health outcomes.