Machine learning of human plasma lipidomes for obesity estimation in a large population cohort

Metabolite profiles of distinct obesity phenotypes integrating impacts of altitude and their association with diet and metabolic disorders in Tibetans

OBJECTIVE

Improved understanding of metabolic obesity phenotypes holds great promise for personalized strategies to combat obesity and its co-morbidities. Such investigation is however lacking in Tibetans with unique living environments and lifestyle in the highlands. Effects of altitude on heterogeneous metabolic obesity phenotypes remain unexplored.

METHODS

We defined metabolic obesity phenotypes i.e., metabolically healthy/unhealthy and obesity/normal weight in Tibetans (n = 1204) living at 2800 m in the suburb or over 4000 m in pastoral areas. 129 lipoprotein parameters and 25 low-molecular-weight metabolites were quantified and their associations with each phenotype were assessed using logistic regression models adjusting for potential confounders. The metabolic BMI (mBMI) was generated using a machine learning strategy and its relationship with prevalence of obesity co-morbidities and dietary exposures were investigated.

RESULTS

Ultrahigh altitude positively associated with the metabolically healthy and non-obese phenotype and had a tendency towards a negative association with metabolically unhealthy phenotype. Phenotype-specific associations were found for 107 metabolites (e.g., lipoprotein subclasses, N-acetyl-glycoproteins, amino acids, fatty acids and lactate, p < 0.05), among which 55 were manipulated by altitude. The mBMI showed consistent yet more pronounced associations with cardiometabolic outcomes than BMI. The ORs for diabetes, prediabetes and hypertriglyceridemia were reduced in individuals residing at ultrahigh altitude compared to those residing at high altitude. The mBMI mediated the negative association between pastoral diet and prevalence of prediabetes, hypertension and hypertriglyceridemia, respectively.

CONCLUSIONS

We found metabolite markers representing distinct obesity phenotypes associated with obesity co-morbidities and the modification effect of altitude, deciphering mechanisms underlying protective effect of ultrahigh altitude and the pastoral diet on metabolic health.

Deep serum lipidomics identifies evaluative and predictive biomarkers for individualized glycemic responses following low-energy diet-induced weight loss: a PREVention of diabetes through lifestyle Intervention and population studies in Europe and around the World (PREVIEW) substudy

BACKGROUND

Weight loss through lifestyle interventions, notably low-energy diets, offers glycemic benefits in populations with overweight-associated prediabetes. However, >50% of these individuals fail to achieve normoglycemia after weight loss. Circulating lipids hold potential for evaluating dietary impacts and predicting diabetes risk.

OBJECTIVES

This study sought to identify serum lipids that could serve as evaluative or predictive biomarkers for individual glycemic changes following diet-induced weight loss.

METHODS

We studied 104 participants with overweight-associated prediabetes, who lost ≥8% weight via a low-energy diet over 8 wk. High-coverage lipidomics was conducted in serum samples before and after the dietary intervention. The lipidomic recalibration was assessed using differential lipid abundance comparisons and partial least squares discriminant analyses. Associations between lipid changes and clinical characteristics were determined by Spearman correlation and Bootstrap Forest of ensemble machine learning model. Baseline lipids, predictive of glycemic parameters changes postweight loss, were assessed using Bootstrap Forest analyses.

RESULTS

We quantified 439 serum lipid species and 9 related organic acids. Dietary intervention significantly reduced diacylglycerols, ceramides, lysophospholipids, and ether-linked phosphatidylethanolamine. In contrast, acylcarnitines, short-chain fatty acids, organic acids, and ether-linked phosphatidylcholine increased significantly. Changes in certain lipid species (e.g., saturated and monounsaturated fatty acid-containing glycerolipids, sphingadienine-based very long-chain sphingolipids, and organic acids) were closely associated with clinical glycemic parameters. Six baseline bioactive sphingolipids primarily predicted changes in fasting plasma glucose. In addition, a number of baseline lipid species, mainly diacylglycerols and triglycerides, were predictive of clinical changes in hemoglobin A1c, insulin and homeostasis model assessment of insulin resistance.

CONCLUSIONS

Newly discovered serum lipidomic alterations and the associated changes in lipid-clinical variables suggest broad metabolic reprogramming related to diet-mediated glycemic control. Novel lipid predictors of glycemic outcomes could facilitate early stratification of individuals with prediabetes who are metabolically less responsive to weight loss, enabling more tailored intervention strategies beyond 1-size-fits-all lifestyle modification advice. The PREVIEW lifestyle intervention study was registered at clinicaltrials.gov as NCT01777893 (https://clinicaltrials.gov/study/NCT01777893).

Liquid-liquid phase separation in subcellular assemblages and signaling pathways: Chromatin modifications induced gene regulation for cellular physiology and functions including carcinogenesis

Liquid-liquid phase separation (LLPS) describes many biochemical processes, including hydrogel formation, in the integrity of macromolecular assemblages and existence of membraneless organelles, including ribosome, nucleolus, nuclear speckles, paraspeckles, promyelocytic leukemia (PML) bodies, Cajal bodies (all exert crucial roles in cellular physiology), and evidence are emerging day by day. Also, phase separation is well documented in generation of plasma membrane subdomains and interplay between membranous and membraneless organelles. Intrinsically disordered regions (IDRs) of biopolymers/proteins are the most critical sticking regions that aggravate the formation of such condensates. Remarkably, phase separated condensates are also involved in epigenetic regulation of gene expression, chromatin remodeling, and heterochromatinization. Epigenetic marks on DNA and histones cooperate with RNA-binding proteins through their IDRs to trigger LLPS for facilitating transcription. How phase separation coalesces mutant oncoproteins, orchestrate tumor suppressor genes expression, and facilitated cancer-associated signaling pathways are unravelling. That autophagosome formation and DYRK3-mediated cancer stem cell modification also depend on phase separation is deciphered in part. In view of this, and to linchpin insight into the subcellular membraneless organelle assembly, gene activation and biological reactions catalyzed by enzymes, and the downstream physiological functions, and how all these events are precisely facilitated by LLPS inducing organelle function, epigenetic modulation of gene expression in this scenario, and how it goes awry in cancer progression are summarized and presented in this article.

Plasma metabolomic profile of adiposity and body composition in childhood: The Genetics of Glucose regulation in Gestation and Growth cohort

OBJECTIVE

This study identified metabolite modules associated with adiposity and body fat distribution in childhood using gold-standard measurements.

METHODS

We used cross-sectional data from 329 children at mid-childhood (age 5.3 ± 0.3 years; BMI 15.7 ± 1.5 kg/m2) from the Genetics of Glucose regulation in Gestation and Growth (Gen3G), a prospective pre-birth cohort. We quantified 1038 plasma metabolites and measured body composition using the gold-standard dual-energy x-ray absorptiometry (DXA), in addition to skinfold, waist circumference, and BMI. We applied weighted-correlation network analysis to identify a network of highly correlated metabolite modules. Spearman's partial correlations were applied to determine the associations of adiposity with metabolite modules and individual metabolites with false discovery rate (FDR) correction.

RESULTS

We identified a 'green' module of 120 metabolites, primarily comprised of lipids (mostly sphingomyelins and phosphatidylcholine), that showed positive correlations (all FDR p < 0.05) with DXA estimates of total and truncal fat (ρadjusted = 0.11-0.19), skinfold measures (ρadjusted = 0.09-0.26), and BMI and waist circumference (ρadjusted = 0.15 and 0.18, respectively). These correlations were similar when stratified by sex. Within this module, sphingomyelin (d18:2/14:0, d18:1/14:1)*, a sphingomyelin sub-specie that is an important component of cell membranes, showed the strongest associations.

CONCLUSIONS

A module of metabolites was associated with adiposity measures in childhood.

iLipidome: enhancing statistical power and interpretability using hidden biosynthetic interdependencies in the lipidome

Numerous biological processes and diseases are influenced by lipid composition. Advances in lipidomics are elucidating their roles, but analyzing and interpreting lipidomics data at the systems level remain challenging. To address this, we present iLipidome, a method for analyzing lipidomics data in the context of the lipid biosynthetic network, thus accounting for the interdependence of measured lipids. iLipidome enhances statistical power, enables reliable clustering and lipid enrichment analysis, and links lipidomic changes to their genetic origins. We applied iLipidome to investigate mechanisms driving changes in cellular lipidomes following supplementation of docosahexaenoic acid (DHA) and successfully identified the genetic causes of alterations. We further demonstrated how iLipidome can disclose enzyme-substrate specificity and pinpoint prospective glioblastoma therapeutic targets. Finally, iLipidome enabled us to explore underlying mechanisms of cardiovascular disease and could guide the discovery of early lipid biomarkers. Thus, iLipidome can assist researchers studying the essence of lipidomic data and advance the field of lipid biology.

A combined observational and Mendelian randomization investigation reveals NMR-measured analytes to be risk factors of major cardiovascular diseases

Dyslipidaemias is the leading risk factor of several major cardiovascular diseases (CVDs), but there is still a lack of sufficient evidence supporting a causal role of lipoprotein subspecies in CVDs. In this study, we comprehensively investigated several lipoproteins and their subspecies, as well as other metabolites, in relation to coronary heart disease (CHD), heart failure (HF) and ischemic stroke (IS) longitudinally and by Mendelian randomization (MR) leveraging NMR-measured metabolomic data from 118,012 UK Biobank participants. We found that 123, 110 and 36 analytes were longitudinally associated with myocardial infarction, HF and IS (FDR < 0.05), respectively, and 25 of those were associated with all three outcomes. MR analysis suggested that genetically predicted levels of 70, 58 and 7 analytes were associated with CHD, HF and IS (FDR < 0.05), respectively. Two analytes, ApoB/ApoA1 and M-HDL-C were associated with all three CVD outcomes in the MR analyses, and the results for M-HDL-C were concordant in both observational and MR analyses. Our results implied that the apoB/apoA1 ratio and cholesterol in medium size HDL were particularly of importance to understand the shared pathophysiology of CHD, HF and IS and thus should be further investigated for the prevention of all three CVDs.

Plasma Lipidomic Profiling Using Mass Spectrometry for Multiple Sclerosis Diagnosis and Disease Activity Stratification (LipidMS)

This investigation explores the potential of plasma lipidomic signatures for aiding in the diagnosis of Multiple Sclerosis (MS) and evaluating the clinical course and disease activity of diseased patients. Plasma samples from 60 patients with MS (PwMS) were clinically stratified to either a relapsing-remitting (RRMS) or a chronic progressive MS course and 60 age-matched controls were analyzed using state-of-the-art direct infusion quantitative shotgun lipidomics. To account for potential confounders, data were filtered for age and BMI correlations. The statistical analysis employed supervised and unsupervised multivariate data analysis techniques, including a principal component analysis (PCA), a partial least squares discriminant analysis (oPLS-DA) and a random forest (RF). To determine whether the significant absolute differences in the lipid subspecies have a relevant effect on the overall composition of the respective lipid classes, we introduce a class composition visualization (CCV). We identified 670 lipids across 16 classes. PwMS showed a significant increase in diacylglycerols (DAG), with DAG 16:0;0_18:1;0 being proven to be the lipid with the highest predictive ability for MS as determined by RF. The alterations in the phosphatidylethanolamines (PE) were mainly linked to RRMS while the alterations in the ether-bound PEs (PE O-) were found in chronic progressive MS. The amount of CE species was reduced in the CPMS cohort whereas TAG species were reduced in the RRMS patients, both lipid classes being relevant in lipid storage. Combining the above mentioned data analyses, distinct lipidomic signatures were isolated and shown to be correlated with clinical phenotypes. Our study suggests that specific plasma lipid profiles are not merely associated with the diagnosis of MS but instead point toward distinct clinical features in the individual patient paving the way for personalized therapy and an enhanced understanding of MS pathology.

Identifying subgroups of childhood obesity by using multiplatform metabotyping

Introduction: Obesity results from an interplay between genetic predisposition and environmental factors such as diet, physical activity, culture, and socioeconomic status. Personalized treatments for obesity would be optimal, thus necessitating the identification of individual characteristics to improve the effectiveness of therapies. For example, genetic impairment of the leptin-melanocortin pathway can result in rare cases of severe early-onset obesity. Metabolomics has the potential to distinguish between a healthy and obese status; however, differentiating subsets of individuals within the obesity spectrum remains challenging. Factor analysis can integrate patient features from diverse sources, allowing an accurate subclassification of individuals. Methods: This study presents a workflow to identify metabotypes, particularly when routine clinical studies fail in patient categorization. 110 children with obesity (BMI > +2 SDS) genotyped for nine genes involved in the leptin-melanocortin pathway (CPE, MC3R, MC4R, MRAP2, NCOA1, PCSK1, POMC, SH2B1, and SIM1) and two glutamate receptor genes (GRM7 and GRIK1) were studied; 55 harboring heterozygous rare sequence variants and 55 with no variants. Anthropometric and routine clinical laboratory data were collected, and serum samples processed for untargeted metabolomic analysis using GC-q-MS and CE-TOF-MS and reversed-phase U(H)PLC-QTOF-MS/MS in positive and negative ionization modes. Following signal processing and multialignment, multivariate and univariate statistical analyses were applied to evaluate the genetic trait association with metabolomics data and clinical and routine laboratory features. Results and Discussion: Neither the presence of a heterozygous rare sequence variant nor clinical/routine laboratory features determined subgroups in the metabolomics data. To identify metabolomic subtypes, we applied Factor Analysis, by constructing a composite matrix from the five analytical platforms. Six factors were discovered and three different metabotypes. Subtle but neat differences in the circulating lipids, as well as in insulin sensitivity could be established, which opens the possibility to personalize the treatment according to the patients categorization into such obesity subtypes. Metabotyping in clinical contexts poses challenges due to the influence of various uncontrolled variables on metabolic phenotypes. However, this strategy reveals the potential to identify subsets of patients with similar clinical diagnoses but different metabolic conditions. This approach underscores the broader applicability of Factor Analysis in metabotyping across diverse clinical scenarios.

Genome-wide association analysis of plasma lipidome identifies 495 genetic associations

The human plasma lipidome captures risk for cardiometabolic diseases. To discover new lipid-associated variants and understand the link between lipid species and cardiometabolic disorders, we perform univariate and multivariate genome-wide analyses of 179 lipid species in 7174 Finnish individuals. We fine-map the associated loci, prioritize genes, and examine their disease links in 377,277 FinnGen participants. We identify 495 genome-trait associations in 56 genetic loci including 8 novel loci, with a considerable boost provided by the multivariate analysis. For 26 loci, fine-mapping identifies variants with a high causal probability, including 14 coding variants indicating likely causal genes. A phenome-wide analysis across 953 disease endpoints reveals disease associations for 40 lipid loci. For 11 coronary artery disease risk variants, we detect strong associations with lipid species. Our study demonstrates the power of multivariate genetic analysis in correlated lipidomics data and reveals genetic links between diseases and lipid species beyond the standard lipids.

Understanding cross-data dynamics of individual and social/environmental factors through a public health lens: explainable machine learning approaches

Introduction

The rising prevalence of obesity has become a public health concern, requiring efficient and comprehensive prevention strategies.

Methods

This study innovatively investigated the combined influence of individual and social/environmental factors on obesity within the urban landscape of Seoul, by employing advanced machine learning approaches. We collected 'Community Health Surveys' and credit card usage data to represent individual factors. In parallel, we utilized 'Seoul Open Data' to encapsulate social/environmental factors contributing to obesity. A Random Forest model was used to predict obesity based on individual factors. The model was further subjected to Shapley Additive Explanations (SHAP) algorithms to determine each factor's relative importance in obesity prediction. For social/environmental factors, we used the Geographically Weighted Least Absolute Shrinkage and Selection Operator (GWLASSO) to calculate the regression coefficients.

Results

The Random Forest model predicted obesity with an accuracy of >90%. The SHAP revealed diverse influential individual obesity-related factors in each Gu district, although 'self-awareness of obesity', 'weight control experience', and 'high blood pressure experience' were among the top five influential factors across all Gu districts. The GWLASSO indicated variations in regression coefficients between social/environmental factors across different districts.

Conclusion

Our findings provide valuable insights for designing targeted obesity prevention programs that integrate different individual and social/environmental factors within the context of urban design, even within the same city. This study enhances the efficient development and application of explainable machine learning in devising urban health strategies. We recommend that each autonomous district consider these differential influential factors in designing their budget plans to tackle obesity effectively.

Metabolic phenotyping of BMI to characterize cardiometabolic risk: evidence from large population-based cohorts

Obesity is a risk factor for type 2 diabetes and cardiovascular disease. However, a substantial proportion of patients with these conditions have a seemingly normal body mass index (BMI). Conversely, not all obese individuals present with metabolic disorders giving rise to the concept of "metabolically healthy obese". We use lipidomic-based models for BMI to calculate a metabolic BMI score (mBMI) as a measure of metabolic dysregulation associated with obesity. Using the difference between mBMI and BMI (mBMIΔ), we identify individuals with a similar BMI but differing in their metabolic health and disease risk profiles. Exercise and diet associate with mBMIΔ suggesting the ability to modify mBMI with lifestyle intervention. Our findings show that, the mBMI score captures information on metabolic dysregulation that is independent of the measured BMI and so provides an opportunity to assess metabolic health to identify "at risk" individuals for targeted intervention and monitoring.

Evaluation of Lipid Extraction Protocols for Untargeted Analysis of Mouse Tissue Lipidome

Lipidomics refers to the full characterization of lipids present within a cell, tissue, organism, or biological system. One of the bottlenecks affecting reliable lipidomic analysis is the extraction of lipids from biological samples. An ideal extraction method should have a maximum lipid recovery and the ability to extract a broad range of lipid classes with acceptable reproducibility. The most common lipid extraction relies on either protein precipitation (monophasic methods) or liquid-liquid partitioning (bi- or triphasic methods). In this study, three monophasic extraction systems, isopropanol (IPA), MeOH/MTBE/CHCl3 (MMC), and EtOAc/EtOH (EE), alongside three biphasic extraction methods, Folch, butanol/MeOH/heptane/EtOAc (BUME), and MeOH/MTBE (MTBE), were evaluated for their performance in characterization of the mouse lipidome of six different tissue types, including pancreas, spleen, liver, brain, small intestine, and plasma. Sixteen lipid classes were investigated in this study using reversed-phase liquid chromatography/mass spectrometry. Results showed that all extraction methods had comparable recoveries for all tested lipid classes except lysophosphatidylcholines, lysophosphatidylethanolamines, acyl carnitines, sphingomyelines, and sphingosines. The recoveries of these classes were significantly lower with the MTBE method, which could be compensated by the addition of stable isotope-labeled internal standards prior to lipid extraction. Moreover, IPA and EE methods showed poor reproducibility in extracting lipids from most tested tissues. In general, Folch is the optimum method in terms of efficacy and reproducibility for extracting mouse pancreas, spleen, brain, and plasma. However, MMC and BUME methods are more favored when extracting mouse liver or intestine.

The metabolomic profiling of total fat and fat distribution in a multi-cohort study of women and men

Currently studies aiming for the comprehensive metabolomics profiling of measured total fat (%) as well as fat distribution in both sexes are lacking. In this work, bioimpedance analysis was applied to measure total fat (%) and fat distribution (trunk to leg ratio). Liquid chromatography-mass spectrometry-based untargeted metabolomics was employed to profile the metabolic signatures of total fat (%) and fat distribution in 3447 participants from three Swedish cohorts (EpiHealth, POEM and PIVUS) using a discovery-replication cross-sectional study design. Total fat (%) and fat distribution were associated with 387 and 120 metabolites in the replication cohort, respectively. Enriched metabolic pathways for both total fat (%) and fat distribution included protein synthesis, branched-chain amino acids biosynthesis and metabolism, glycerophospholipid metabolism and sphingolipid metabolism. Four metabolites were mainly related to fat distribution: glutarylcarnitine (C5-DC), 6-bromotryptophan, 1-stearoyl-2-oleoyl-GPI (18:0/18:1) and pseudouridine. Five metabolites showed different associations with fat distribution in men and women: quinolinate, (12Z)-9,10-dihydroxyoctadec-12-enoate (9,10-DiHOME), two sphingomyelins and metabolonic lactone sulfate. To conclude, total fat (%) and fat distribution were associated with a large number of metabolites, but only a few were exclusively associated with fat distribution and of those metabolites some were associated with sex*fat distribution. Whether these metabolites mediate the undesirable effects of obesity on health outcomes remains to be further investigated.

Advanced omics techniques shed light on CD1d-mediated lipid antigen presentation to iNKT cells

Invariant natural killer T cells (iNKT cells) can be activated through binding antigenic lipid/CD1d complexes to their TCR. Antigenic lipids are processed, loaded, and displayed in complex with CD1d by lipid antigen presenting cells (LAPCs). The mechanism of lipid antigen presentation via CD1d is highly conserved with recent work showing adipocytes are LAPCs that, besides having a role in lipid storage, can activate iNKT cells and play an important role in systemic metabolic disease. Recent studies shed light on parameters potentially dictating cytokine output and how obesity-associated metabolic disease may affect such parameters. By following a lipid antigen's journey, we identify five key areas which may dictate cytokine skew: co-stimulation, structural properties of the lipid antigen, stability of lipid antigen/CD1d complexes, intracellular and extracellular pH, and intracellular and extracellular lipid environment. Recent publications indicate that the combination of advanced omics-type approaches and machine learning may be a fruitful way to interconnect these 5 areas, with the ultimate goal to provide new insights for therapeutic exploration.

Clinical lipidomics in the era of the big data

Lipidomics as a branch of metabolomics provides unique information on the complex lipid profile in biological materials. In clinically focused studies, hundreds of lipids together with available clinical information proved to be an effective tool in the discovery of biomarkers and understanding of pathobiochemistry. However, despite the introduction of lipidomics nearly twenty years ago, only dozens of big data studies using clinical lipidomics have been published to date. In this review, we discuss the lipidomics workflow, statistical tools, and the challenges of standartisation. The consequent summary divided into major clinical areas of cardiovascular disease, cancer, diabetes mellitus, neurodegenerative and liver diseases is demonstrating the importance of clinical lipidomics. In these publications, the potential of lipidomics for prediction, diagnosis or finding new targets for the treatment of selected diseases can be seen. The first of these results have already been implemented in clinical practice in the field of cardiovascular diseases, while in other areas we can expect the application of the results summarized in this review in the near future.

Artificial intelligence and body composition

AIMS

Although obesity is associated with chronic disease, a large section of the population with high BMI does not have an increased risk of metabolic disease. Increased visceral adiposity and sarcopenia are also risk factors for metabolic disease in people with normal BMI. Artificial Intelligence (AI) techniques can help assess and analyze body composition parameters for predicting cardiometabolic health. The purpose of the study was to systematically explore literature involving AI techniques for body composition assessment and observe general trends.

METHODS

We searched the following databases: Embase, Web of Science, and PubMed. There was a total of 354 search results. After removing duplicates, irrelevant studies, and reviews(a total of 303), 51 studies were included in the systematic review.

RESULTS

AI techniques have been studied for body composition analysis in the context of diabetes mellitus, hypertension, cancer and many specialized diseases. Imaging techniques employed for AI methods include CT (Computerized Tomography), MRI (Magnetic Resonance Imaging), ultrasonography, plethysmography, and EKG(Electrocardiogram). Automatic segmentation of body composition by deep learning with convolutional networks has helped determine and quantify muscle mass. Limitations include heterogeneity of study populations, inherent bias in sampling, and lack of generalizability. Different bias mitigation strategies should be evaluated to address these problems and improve the applicability of AI to body composition analysis.

CONCLUSIONS

AI assisted measurement of body composition might assist in improved cardiovascular risk stratification when applied in the appropriate clinical context.

The metabolic profile of waist to hip ratio-A multi-cohort study

BACKGROUND

The genetic background of general obesity and fat distribution is different, pointing to separate underlying physiology. Here, we searched for metabolites and lipoprotein particles associated with fat distribution, measured as waist/hip ratio adjusted for fat mass (WHRadjfatmass), and general adiposity measured as percentage fat mass.

METHOD

The sex-stratified association of 791 metabolites detected by liquid chromatography-mass spectrometry (LC-MS) and 91 lipoprotein particles measured by nuclear magnetic spectroscopy (NMR) with WHRadjfatmass and fat mass were assessed using three population-based cohorts: EpiHealth (n = 2350) as discovery cohort, with PIVUS (n = 603) and POEM (n = 502) as replication cohorts.

RESULTS

Of the 193 LC-MS-metabolites being associated with WHRadjfatmass in EpiHealth (false discovery rate (FDR) <5%), 52 were replicated in a meta-analysis of PIVUS and POEM. Nine metabolites, including ceramides, sphingomyelins or glycerophosphatidylcholines, were inversely associated with WHRadjfatmass in both sexes. Two of the sphingomyelins (d18:2/24:1, d18:1/24:2 and d18:2/24:2) were not associated with fat mass (p>0.50). Out of 91, 82 lipoprotein particles were associated with WHRadjfatmass in EpiHealth and 42 were replicated. Fourteen of those were associated in both sexes and belonged to very-large or large HDL particles, all being inversely associated with both WHRadjfatmass and fat mass.

CONCLUSION

Two sphingomyelins were inversely linked to body fat distribution in both men and women without being associated with fat mass, while very-large and large HDL particles were inversely associated with both fat distribution and fat mass. If these metabolites represent a link between an impaired fat distribution and cardiometabolic diseases remains to be established.

Applications of Artificial Intelligence to Obesity Research: Scoping Review of Methodologies

BACKGROUND

Obesity is a leading cause of preventable death worldwide. Artificial intelligence (AI), characterized by machine learning (ML) and deep learning (DL), has become an indispensable tool in obesity research.

OBJECTIVE

This scoping review aimed to provide researchers and practitioners with an overview of the AI applications to obesity research, familiarize them with popular ML and DL models, and facilitate the adoption of AI applications.

METHODS

We conducted a scoping review in PubMed and Web of Science on the applications of AI to measure, predict, and treat obesity. We summarized and categorized the AI methodologies used in the hope of identifying synergies, patterns, and trends to inform future investigations. We also provided a high-level, beginner-friendly introduction to the core methodologies to facilitate the dissemination and adoption of various AI techniques.

RESULTS

We identified 46 studies that used diverse ML and DL models to assess obesity-related outcomes. The studies found AI models helpful in detecting clinically meaningful patterns of obesity or relationships between specific covariates and weight outcomes. The majority (18/22, 82%) of the studies comparing AI models with conventional statistical approaches found that the AI models achieved higher prediction accuracy on test data. Some (5/46, 11%) of the studies comparing the performances of different AI models revealed mixed results, indicating the high contingency of model performance on the data set and task it was applied to. An accelerating trend of adopting state-of-the-art DL models over standard ML models was observed to address challenging computer vision and natural language processing tasks. We concisely introduced the popular ML and DL models and summarized their specific applications in the studies included in the review.

CONCLUSIONS

This study reviewed AI-related methodologies adopted in the obesity literature, particularly ML and DL models applied to tabular, image, and text data. The review also discussed emerging trends such as multimodal or multitask AI models, synthetic data generation, and human-in-the-loop that may witness increasing applications in obesity research.

Interplay between dietary intake, gut microbiota, and metabolic profile in obese adolescents: Sex-dependent differential patterns

BACKGROUND & AIMS

The interplay among dietary intake, gut microbiota, gut metabolites and circulating metabolites in adolescents is barely known, not to mention sex-dependent pattern. We aimed to explore unique profiles of gut bacterial, gut metabolites and circulating metabolites from both genders of adolescents due to BMI and eating pattern.

METHODS

Clinical indices, fecal gut microbiota, fecal and plasma metabolites, and diet intake information were collected in case-control sample matched for normal and obesity in girls (normal = 12, obesity = 12) and boys (normal = 20, obesity = 20), respectively. 16S rRNA gene sequencing and untargeted metabolomics was performed to analysis the signature of gut microbiota and metabolites. Unique profiles of girls associated with BMI and eating pattern was revealed by Spearman's correlations analysis, co-occurrence network analysis, Kruskal-Wallis test, and Wilcoxon rank-sum test.

RESULTS

Gender difference was found between normal and obese adolescents in gut microbiota, fecal metabolites, and plasma metabolites. The Parabacteroides were only decreased in obese girls. And the characteristic of obese girls' and boys' cases in fecal and plasma was xanthine and glutamine, ornithine and LCA, respectively. Soy products intake was negatively associated with Parabacteroides. The predicted model has a higher accuracy based on the combined markers in obesity boys (AUC = 0.97) and girls (AUC = 0.97), respectively.

CONCLUSIONS

Reduced abundance of Phascolarctobacterium and Parabacteroides, as well as the increased fecal xanthine and ornithine, may provide a novel biomarker signature in obesity girls and boys. Soy products intake was positively and negatively associated with Romboutsia and Parabacteroides abundance, respectively. And the combined markers facilitate the accuracy of predicting obesity in girls and boys in advance.

Trip13 Depletion in Liver Cancer Induces a Lipogenic Response Contributing to Plin2-Dependent Mitotic Cell Death

Aberrant energy metabolism and cell cycle regulation both critically contribute to malignant cell growth and both processes represent targets for anticancer therapy. It is shown here that depletion of the AAA+-ATPase thyroid hormone receptor interacting protein 13 (Trip13) results in mitotic cell death through a combined mechanism linking lipid metabolism to aberrant mitosis. Diminished Trip13 levels in hepatocellular carcinoma cells result in insulin-receptor-/Akt-pathway-dependent accumulation of lipid droplets, which act as functional acentriolar microtubule organizing centers disturbing mitotic spindle polarity. Specifically, the lipid-droplet-coating protein perilipin 2 (Plin2) is required for multipolar spindle formation, induction of DNA damage, and mitotic cell death. Plin2 expression in different tumor cells confers susceptibility to cell death induced by Trip13 depletion as well as treatment with paclitaxel, a spindle-interfering drug commonly used against different cancers. Thus, assessment of Plin2 levels enables the stratification of tumor responsiveness to mitosis-targeting drugs, including clinically approved paclitaxel and Trip13 inhibitors currently under development.

IoT Framework for a Decision-Making System of Obesity and Overweight Extrapolation among Children, Youths, and Adults

Approximately 30% of the global population is suffering from obesity and being overweight, which is approximately 2.1 billion people worldwide. The ratio is expected to surpass 40% by 2030 if the current balance continues to grow. The global pandemic due to COVID-19 will also impact the predicted obesity rates. It will cause a significant increase in morbidity and mortality worldwide. Multiple chronic diseases are associated with obesity and several threat elements are associated with obesity. Various challenges are involved in the understanding of risk factors and the ratio of obesity. Therefore, diagnosing obesity in its initial stages might significantly increase the patient’s chances of effective treatment. The Internet of Things (IoT) has attained an evolving stage in the development of the contemporary environment of healthcare thanks to advancements in information and communication technologies. Therefore, in this paper, we thoroughly investigated machine learning techniques for making an IoT-enabled system. In the first phase, the proposed system analyzed the performances of random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM), decision tree (DT), logistic regression (LR), and naïve Bayes (NB) algorithms on the obesity dataset. The second phase, on the other hand, introduced an IoT-based framework that adopts a multi-user request system by uploading the data to the cloud for the early diagnosis of obesity. The IoT framework makes the system available to anyone (and everywhere) for precise obesity categorization. This research will help the reader understand the relationships among risk factors with weight changes and their visualizations. Furthermore, it also focuses on how existing datasets can help one study the obesity nature and which classification and regression models perform well in correspondence to others.

Dysregulated Metabolic Pathways in Subjects with Obesity and Metabolic Syndrome

Background: Obesity coexists with variable features of metabolic syndrome, which is associated with dysregulated metabolic pathways. We assessed potential associations between serum metabolites and features of metabolic syndrome in Arabic subjects with obesity. Methods: We analyzed a dataset of 39 subjects with obesity only (OBO, n = 18) age-matched to subjects with obesity and metabolic syndrome (OBM, n = 21). We measured 1069 serum metabolites and correlated them to clinical features. Results: A total of 83 metabolites, mostly lipids, were significantly different (p < 0.05) between the two groups. Among lipids, 22 sphingomyelins were decreased in OBM compared to OBO. Among non-lipids, quinolinate, kynurenine, and tryptophan were also decreased in OBM compared to OBO. Sphingomyelin is negatively correlated with glucose, HbA1C, insulin, and triglycerides but positively correlated with HDL, LDL, and cholesterol. Differentially enriched pathways include lysine degradation, amino sugar and nucleotide sugar metabolism, arginine and proline metabolism, fructose and mannose metabolism, and galactose metabolism. Conclusions: Metabolites and pathways associated with chronic inflammation are differentially expressed in subjects with obesity and metabolic syndrome compared to subjects with obesity but without the clinical features of metabolic syndrome.

Human Milk Metabolomics Are Related to Maternal Adiposity, Infant Growth Rate and Allergies: The Chinese Human Milk Project

The metabolomic profiles of Chinese human milk have been poorly documented. The objective of the study was to explore associations between human milk metabotypes, maternal adiposity, infant growth patterns, and risk of allergies. Two hundred mother−infant dyads from seven cities were randomly selected from the Chinese Human Milk Project (CHMP). Untargeted human milk metabolomic profiles were determined using HPLC-MS/MS. Two human milk metabotypes were identified using principal component analysis. Principal component (PC) 1 was characterized by high linoleic acid metabolites with low purine nucleosides and metabolites of glutamate and glutathione metabolism. PC 2 was characterized by high glycerophospholipids and sphingomyelins content. Higher PC1 scores were associated with slower infant growth rate and higher ambient temperature (p < 0.05). Higher PC 2 scores were related to higher maternal BMI and increased risk of infant allergies (p < 0.05). Future work is needed to understand the biologic mechanisms of these human milk metabotypes.

Metabolic Profiling of Obesity With and Without the Metabolic Syndrome: A Multisample Evaluation

CONTEXT

There is a dispute whether obesity without major metabolic derangements may represent a benign condition or not.

OBJECTIVE

We aimed to compare the plasma metabolome in obese subjects without metabolic syndrome (MetS) with normal-weight subjects without MetS and with obese subjects with MetS.

METHODS

This was a cross-sectional study at 2 academic centers in Sweden. Individuals from 3 population-based samples (EpiHealth, n = 2342, SCAPIS-Uppsala, n = 4985, and SCAPIS-Malmö, n = 3978) were divided into groups according to their body mass index (BMI) and presence/absence of MetS (National Cholesterol Education Program [NCEP]/consensus criteria). In total, 791 annotated endogenous metabolites were measured by ultra-performance liquid chromatography-tandem mass spectrometry.

RESULTS

We observed major differences in metabolite profiles (427 metabolites) between obese (BMI ≥ 30 kg/m2) and normal-weight (BMI < 25 kg/m2) subjects without MetS after adjustment for major lifestyle factors. Pathway enrichment analysis highlighted branch-chained and aromatic amino acid synthesis/metabolism, aminoacyl-tRNA biosynthesis, and sphingolipid metabolism. The same pathways, and similar metabolites, were also highlighted when obese subjects with and without MetS were compared despite adjustment for BMI and waist circumference, or when the metabolites were related to BMI and number of MetS components in a continuous fashion. Similar metabolites and pathways were also related to insulin sensitivity (Matsuda index) in a separate study (POEM, n = 501).

CONCLUSION

Our data suggest a graded derangement of the circulating metabolite profile from lean to obese to MetS, in particular for metabolites involved in amino acid synthesis/metabolism and sphingolipid metabolism. Insulin resistance is a plausible mediator of this gradual metabolic deterioration.

Lipidomic risk scores are independent of polygenic risk scores and can predict incidence of diabetes and cardiovascular disease in a large population cohort

Type 2 diabetes (T2D) and cardiovascular disease (CVD) represent significant disease burdens for most societies and susceptibility to these diseases is strongly influenced by diet and lifestyle. Physiological changes associated with T2D or CVD, such has high blood pressure and cholesterol and glucose levels in the blood, are often apparent prior to disease incidence. Here we integrated genetics, lipidomics, and standard clinical diagnostics to assess future T2D and CVD risk for 4,067 participants from a large prospective population-based cohort, the Malmö Diet and Cancer-Cardiovascular Cohort. By training Ridge regression-based machine learning models on the measurements obtained at baseline when the individuals were healthy, we computed several risk scores for T2D and CVD incidence during up to 23 years of follow-up. We used these scores to stratify the participants into risk groups and found that a lipidomics risk score based on the quantification of 184 plasma lipid concentrations resulted in a 168% and 84% increase of the incidence rate in the highest risk group and a 77% and 53% decrease of the incidence rate in lowest risk group for T2D and CVD, respectively, compared to the average case rates of 13.8% and 22.0%. Notably, lipidomic risk correlated only marginally with polygenic risk, indicating that the lipidome and genetic variants may constitute largely independent risk factors for T2D and CVD. Risk stratification was further improved by adding standard clinical variables to the model, resulting in a case rate of 51.0% and 53.3% in the highest risk group for T2D and CVD, respectively. The participants in the highest risk group showed significantly altered lipidome compositions affecting 167 and 157 lipid species for T2D and CVD, respectively. Our results demonstrated that a subset of individuals at high risk for developing T2D or CVD can be identified years before disease incidence. The lipidomic risk, which is derived from only one single mass spectrometric measurement that is cheap and fast, is informative and could extend traditional risk assessment based on clinical assays.

Untargeted Metabolomics Analysis of the Serum Metabolic Signature of Childhood Obesity

Obesity rates among children are growing rapidly worldwide, placing massive pressure on healthcare systems. Untargeted metabolomics can expand our understanding of the pathogenesis of obesity and elucidate mechanisms related to its symptoms. However, the metabolic signatures of obesity in children have not been thoroughly investigated. Herein, we explored metabolites associated with obesity development in childhood. Untargeted metabolomic profiling was performed on fasting serum samples from 27 obese Caucasian children and adolescents and 15 sex- and age-matched normal-weight children. Three metabolomic assays were combined and yielded 726 unique identified metabolites: gas chromatography–mass spectrometry (GC–MS), hydrophilic interaction liquid chromatography coupled to mass spectrometry (HILIC LC–MS/MS), and lipidomics. Univariate and multivariate analyses showed clear discrimination between the untargeted metabolomes of obese and normal-weight children, with 162 significantly differentially expressed metabolites between groups. Children with obesity had higher concentrations of branch-chained amino acids and various lipid metabolites, including phosphatidylcholines, cholesteryl esters, triglycerides. Thus, an early manifestation of obesity pathogenesis and its metabolic consequences in the serum metabolome are correlated with altered lipid metabolism. Obesity metabolite patterns in the adult population were very similar to the metabolic signature of childhood obesity. Identified metabolites could be potential biomarkers and used to study obesity pathomechanisms.

Circulating Lipid- and Inflammation-Based Risk (CLIR) Score: A Promising New Model for Predicting Outcomes in Complete Colorectal Liver Metastases Resection

BACKGROUND

Colorectal cancer liver metastasis (CRLM) is a determining factor affecting the survival of colorectal cancer (CRC) patients. This study aims at developing a novel prognostic stratification tool for CRLM resection.

METHODS

In this retrospective study, 666 CRC patients who underwent complete CRLM resection from two Chinese medical institutions between 2001 and 2016 were classified into the training (341 patients) and validation (325 patients) cohorts. The primary endpoint was overall survival (OS). Associations between clinicopathological variables, circulating lipid and inflammation biomarkers, and OS were explored. The five most significant prognostic factors were incorporated into the Circulating Lipid- and Inflammation-based Risk (CLIR) score. The predictive ability of the CLIR score and Fong's Clinical Risk Score (CRS) was compared by time-dependent receiver operating characteristic (ROC) analysis.

RESULTS

Five independent predictors associated with worse OS were identified in the training cohort: number of CRLMs >4, maximum diameter of CRLM >4.4 cm, primary lymph node-positive, serum lactate dehydrogenase (LDH) level >250.5 U/L, and serum low-density lipoprotein-cholesterol (LDL-C)/high-density lipoprotein-cholesterol (HDL-C) ratio >2.9. These predictors were included in the CLIR score and each factor was assigned one point. Median OS for the low (score 0-1)-, intermediate (score 2-3)-, and high (score 4-5)-risk groups was 134.0 months, 39.9 months, and 18.7 months in the pooled cohort. The CLIR score outperformed the Fong score with superior discriminatory capacities for OS and RFS, both in the training and validation cohorts.

CONCLUSIONS

The CLIR score demonstrated a promising ability to predict the long-term survival of CRC patients after complete hepatic resection.

Lipid–Protein Interactions in Plasma Membrane Organization and Function

Lipid–protein interactions in cells are involved in various biological processes, including metabolism, trafficking, signaling, host–pathogen interactions, and transmembrane transport. At the plasma membrane, lipid–protein interactions play major roles in membrane organization and function. Several membrane proteins have motifs for specific lipid binding, which modulate protein conformation and consequent function. In addition to such specific lipid–protein interactions, protein function can be regulated by the dynamic, collective behavior of lipids in membranes. Emerging analytical, biochemical, and computational technologies allow us to study the influence of specific lipid–protein interactions, as well as the collective behavior of membranes on protein function. In this article, we review the recent literature on lipid–protein interactions with a specific focus on the current state-of-the-art technologies that enable novel insights into these interactions.

Expected final online publication date for the Annual Review of Biophysics, Volume 51 is May 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Adverse Effects of Refeeding on the Plasma Lipidome in Young Individuals With Anorexia Nervosa?

OBJECTIVE

Refeeding is the cornerstone of anorexia nervosa (AN) treatment, but little is known regarding the optimal pace and dietary composition or possible adverse effects of current clinical practices. Plasma lipids may be a moderating factor underlying unfavorable refeeding effects in AN, such as an abnormal central body fat distribution. The objective of this study was to analyze the plasma lipidome in the acutely underweight state of AN before and after refeeding.

METHOD

Using high-throughput quantitative mass spectrometry-based shotgun lipidomics, we measured 13 lipid classes and 204 lipid species or subspecies in the plasma of young female patients with acute AN, before (n = 39) and after (n = 23) short-term weight restoration during an intensive inpatient refeeding program (median body mass index [BMI] increase = 26.4%), in comparison to those in healthy control participants (n = 37).

RESULTS

Before inpatient treatment, patients with AN exhibited increased concentrations of cholesterol and several other lipid classes. After refeeding, multiple lipid classes including cholesterol and ceramides, as well as certain ceramide species previously associated with obesity or overfeeding, showed increased concentrations, and a pattern of shorter and more saturated triacylgycerides emerged. A machine learning model trained to predict BMI based on the lipidomic profiles revealed a sizable overprediction in patients with AN after weight restoration.

CONCLUSION

The results point toward a profound lipid dysregulation with similarities to obesity and other features of the metabolic syndrome after short-term weight restoration. Thus, this study provides evidence for possible short-term adverse effects of current refeeding practices on the metabolic state and should inspire more research on nutritional interventions in AN.

Predicting obesity and smoking using medication data: A machine-learning approach

PURPOSE

Administrative health datasets are widely used in public health research but often lack information about common confounders. We aimed to develop and validate machine learning (ML)-based models using medication data from Australia's Pharmaceutical Benefits Scheme (PBS) database to predict obesity and smoking.

METHODS

We used data from the D-Health Trial (N = 18 000) and the QSkin Study (N = 43 794). Smoking history, and height and weight were self-reported at study entry. Linkage to the PBS dataset captured 5 years of medication data after cohort entry. We used age, sex, and medication use, classified using anatomical therapeutic classification codes, as potential predictors of smoking (current or quit <10 years ago; never or quit ≥10 years ago) and obesity (obese; non-obese). We trained gradient-boosted machine learning models using data for the first 80% of participants enrolled; models were validated using the remaining 20%. We assessed model performance overall and by sex and age, and compared models generated using 3 and 5 years of PBS data.

RESULTS

Based on the validation dataset using 3 years of PBS data, the area under the receiver operating characteristic curve was 0.70 (95% confidence interval [CI] 0.68-0.71) for predicting obesity and 0.71 (95% CI 0.70-0.72) for predicting smoking. Models performed better in women than in men. Using 5 years of PBS data resulted in marginal improvement.

CONCLUSIONS

Medication data in combination with age and sex can be used to predict obesity and smoking. These models may be of value to researchers using data collected for administrative purposes.

Mouse lipidomics reveals inherent flexibility of a mammalian lipidome

Lipidomics has become an indispensable method for the quantitative assessment of lipid metabolism in basic, clinical, and pharmaceutical research. It allows for the generation of information-dense datasets in a large variety of experimental setups and model organisms. Previous studies, mostly conducted in mice (Mus musculus), have shown a remarkable specificity of the lipid compositions of different cell types, tissues, and organs. However, a systematic analysis of the overall variation of the mouse lipidome is lacking. To fill this gap, in the present study, the effect of diet, sex, and genotype on the lipidomes of mouse tissues, organs, and bodily fluids has been investigated. Baseline quantitative lipidomes consisting of 796 individual lipid molecules belonging to 24 lipid classes are provided for 10 different sample types. Furthermore, the susceptibility of lipidomes to the tested parameters is assessed, providing insights into the organ-specific lipidomic plasticity and flexibility. This dataset provides a valuable resource for basic and pharmaceutical researchers working with murine models and complements existing proteomic and transcriptomic datasets. It will inform experimental design and facilitate interpretation of lipidomic datasets.

Clinical lipidomics: realizing the potential of lipid profiling

Dysregulation of lipid metabolism plays a major role in the etiology and sequelae of inflammatory disorders, cardiometabolic and neurological diseases, and several forms of cancer. Recent advances in lipidomic methodology allow comprehensive lipidomic profiling of clinically relevant biological samples, enabling researchers to associate lipid species and metabolic pathways with disease onset and progression. The resulting data serve not only to advance our fundamental knowledge of the underlying disease process but also to develop risk assessment models to assist in the diagnosis and management of disease. Currently, clinical applications of in-depth lipidomic profiling are largely limited to the use of research-based protocols in the analysis of population or clinical sample sets. However, we foresee the development of purpose-built clinical platforms designed for continuous operation and clinical integration-assisting health care providers with disease risk assessment, diagnosis, and monitoring. Herein, we review the current state of clinical lipidomics, including the use of research-based techniques and platforms in the analysis of clinical samples as well as assays already available to clinicians. With a primary focus on MS-based strategies, we examine instrumentation, analysis techniques, statistical models, prospective design of clinical platforms, and the possible pathways toward implementation of clinical lipidomics.

Use of human PBMC to analyse the impact of obesity on lipid metabolism and metabolic status: a proof-of-concept pilot study

Peripheral blood mononuclear cells (PBMC) are widely used as a biomarker source in nutrition/obesity studies because they reflect gene expression profiles of internal tissues. In this pilot proof-of-concept study we analysed in humans if, as we previously suggested in rodents, PBMC could be a surrogate tissue to study overweight/obesity impact on lipid metabolism. Pre-selected key lipid metabolism genes based in our previous preclinical studies were analysed in PBMC of normoglycemic normal-weight (NW), and overweight-obese (OW-OB) subjects before and after a 6-month weight-loss plan. PBMC mRNA levels of CPT1A, FASN and SREBP-1c increased in the OW-OB group, according with what described in liver and adipose tissue of humans with obesity. This altered expression pattern was related to increased adiposity and early signs of metabolic impairment. Greater weight loss and/or metabolic improvement as result of the intervention was related to lower CPT1A, FASN and SREBP-1c gene expression in an adjusted linear mixed-effects regression analysis, although no gene expression recovery was observed when considering mean comparisons. Thus, human PBMC reflect lipid metabolism expression profile of energy homeostatic tissues, and early obesity-related alterations in metabolic at-risk subjects. Further studies are needed to understand PBMC usefulness for analysis of metabolic recovery in weigh management programs.

A systematic literature review on obesity: Understanding the causes & consequences of obesity and reviewing various machine learning approaches used to predict obesity

Obesity is considered a principal public health concern and ranked as the fifth foremost reason for death globally. Overweight and obesity are one of the main lifestyle illnesses that leads to further health concerns and contributes to numerous chronic diseases, including cancers, diabetes, metabolic syndrome, and cardiovascular diseases. The World Health Organization also predicted that 30% of death in the world will be initiated with lifestyle diseases in 2030 and can be stopped through the suitable identification and addressing of associated risk factors and behavioral involvement policies. Thus, detecting and diagnosing obesity as early as possible is crucial. Therefore, the machine learning approach is a promising solution to early predictions of obesity and the risk of overweight because it can offer quick, immediate, and accurate identification of risk factors and condition likelihoods. The present study conducted a systematic literature review to examine obesity research and machine learning techniques for the prevention and treatment of obesity from 2010 to 2020. Accordingly, 93 papers are identified from the review articles as primary studies from an initial pool of over 700 papers addressing obesity. Consequently, this study initially recognized the significant potential factors that influence and cause adult obesity. Next, the main diseases and health consequences of obesity and overweight are investigated. Ultimately, this study recognized the machine learning methods that can be used for the prediction of obesity. Finally, this study seeks to support decision-makers looking to understand the impact of obesity on health in the general population and identify outcomes that can be used to guide health authorities and public health to further mitigate threats and effectively guide obese people globally.

LipidSig: a web-based tool for lipidomic data analysis

With the continuing rise of lipidomic studies, there is an urgent need for a useful and comprehensive tool to facilitate lipidomic data analysis. The most important features making lipids different from general metabolites are their various characteristics, including their lipid classes, double bonds, chain lengths, etc. Based on these characteristics, lipid species can be classified into different categories and, more interestingly, exert specific biological functions in a group. In an effort to simplify lipidomic analysis workflows and enhance the exploration of lipid characteristics, we have developed a highly flexible and user-friendly web server called LipidSig. It consists of five sections, namely, Profiling, Differential Expression, Correlation, Network and Machine Learning, and evaluates lipid effects on cellular or disease phenotypes. One of the specialties of LipidSig is the conversion between lipid species and characteristics according to a user-defined characteristics table. This function allows for efficient data mining for both individual lipids and subgroups of characteristics. To expand the server's practical utility, we also provide analyses focusing on fatty acid properties and multiple characteristics. In summary, LipidSig is expected to help users identify significant lipid-related features and to advance the field of lipid biology. The LipidSig webserver is freely available at http://chenglab.cmu.edu.tw/lipidsig

Machine learning applied to serum and cerebrospinal fluid metabolomes revealed altered arginine metabolism in neonatal sepsis with meningoencephalitis

Background

Neonatal sepsis with meningoencephalitis is a common complication of sepsis, which is a leading cause of neonatal death and neurological dysfunction. Early identification of neonatal sepsis with meningoencephalitis is particularly important for reducing brain damage. We recruited 70 patients with neonatal sepsis, 42 of which were diagnosed as meningoencephalitis, and collected cerebrospinal fluid (CSF) and serum samples. The purpose of this study was to find neonatal sepsis with meningoencephalitis-related markers using unbiased metabolomics technology and artificial intelligence analysis based on machine learning methods.

Results

We found that the characteristics of neonatal sepsis with meningoencephalitis were manifested mainly as significant decreases in the concentrations of homo-l-arginine, creatinine, and other arginine metabolites in serum and CSF, suggesting possible changes in nitric oxide synthesis. The antioxidants taurine and proline in the serum of the neonatal sepsis with meningoencephalitis increased significantly, suggesting abnormal oxidative stress. Potentially harmful bile salts and aromatic compounds were significantly increased in the serum of the group with meningoencephalitis. We compared different machine learning methods and found that the lasso algorithm performed best. Combining the lasso and XGBoost algorithms was successful in predicting the concentration of homo-l-arginine in CSF per the concentrations of metabolite markers in the serum.

Conclusions

On the basis of machine learning combined with analysis of the serum and CSF metabolomes, we found metabolite markers related to neonatal sepsis with meningoencephalitis. The characteristics of neonatal sepsis with meningoencephalitis were manifested mainly by changes in arginine metabolism and related changes in creatinine metabolism.

Metabolic View on Human Healthspan: A Lipidome-Wide Association Study

As ageing is a major risk factor for the development of non-communicable diseases, extending healthspan has become a medical and societal necessity. Precise lipid phenotyping that captures metabolic individuality could support healthspan extension strategies. This study applied ‘omic-scale lipid profiling to characterise sex-specific age-related differences in the serum lipidome composition of healthy humans. A subset of the COmPLETE-Health study, composed of 73 young (25.2 ± 2.6 years, 43% female) and 77 aged (73.5 ± 2.3 years, 48% female) clinically healthy individuals, was investigated, using an untargeted liquid chromatography high-resolution mass spectrometry approach. Compared to their younger counterparts, aged females and males exhibited significant higher levels in 138 and 107 lipid species representing 15 and 13 distinct subclasses, respectively. Percentage of difference ranged from 5.8% to 61.7% (females) and from 5.3% to 46.0% (males), with sphingolipid and glycerophophospholipid species displaying the greatest amplitudes. Remarkably, specific sphingolipid and glycerophospholipid species, previously described as cardiometabolically favourable, were found elevated in aged individuals. Furthermore, specific ether-glycerophospholipid and lyso-glycerophosphocholine species displayed higher levels in aged females only, revealing a more favourable lipidome evolution in females. Altogether, age determined the circulating lipidome composition, while lipid species analysis revealed additional findings that were not observed at the subclass level.

MethylDetectR: a software for methylation-based health profiling

DNA methylation is an important biological process that involves the reversible addition of chemical tags called methyl groups to DNA and affects whether genes are active or inactive. Individual methylation profiles are determined by both genetic and environmental influences. Inter-individual variation in DNA methylation profiles can be exploited to estimate or predict a wide variety of human characteristics and disease risk profiles. Indeed, a number of methylation-based predictors of human traits have been developed and linked to important health outcomes. However, there is an unmet need to communicate the applicability and limitations of state-of-the-art methylation-based predictors to the wider community. To address this need, we have created a secure, web-based interactive platform called 'MethylDetectR' which automates the calculation of estimated values or scores for a variety of human traits using blood methylation data. These traits include age, lifestyle traits and high-density lipoprotein cholesterol. Methylation-based predictors often return scores on arbitrary scales. To provide meaning to these scores, users can interactively view how estimated trait scores for a given individual compare against other individuals in the sample. Users can optionally upload binary phenotypes and investigate how estimated traits vary according to case vs. control status for these phenotypes. Users can also view how different methylation-based predictors correlate with one another, and with phenotypic values for corresponding traits in a large reference sample (n = 4,450; Generation Scotland). The 'MethylDetectR' platform allows for the fast and secure calculation of DNA methylation-derived estimates for several human traits. This platform also helps to show the correlations between methylation-based scores and corresponding traits at the level of a sample, report estimated health profiles at an individual level, demonstrate how scores relate to important binary outcomes of interest and highlight the current limitations of molecular health predictors.

Identifying Key Determinants of Childhood Obesity: A Narrative Review of Machine Learning Studies

Machine learning is a class of algorithms able to handle a large number of predictors with potentially nonlinear relationships. By applying machine learning to obesity, researchers can examine how risk factors across multiple settings (e.g., school and home) interact to best predict childhood obesity risk. In this narrative review, we provide an overview of studies that have applied machine learning to predict childhood obesity using a combination of sociodemographic and behavioral risk factors. The objective is to summarize the key determinants of obesity identified in existing machine learning studies and highlight opportunities for future machine learning applications in the field. Of 15 peer-reviewed studies, approximately half examined early childhood (0-24 months of age) determinants. These studies identified child's weight history (e.g., history of overweight/obesity or large increases in weight-related measures between birth and 24 months of age) and parental overweight/obesity (current or prior) as key risk factors, whereas the remaining studies indicated that social factors and physical inactivity were important in middle childhood and late childhood/adolescence. Across age groups, findings suggested that race/ethnic-specific models may be needed to accurately predict obesity from middle childhood onward. Future studies should consider using existing large data sets to take advantage of the benefits of machine learning and should collect a wider range of novel risk factors (e.g., psychosocial and sociocultural determinants of health) to better predict childhood obesity. Ultimately, such research can aid in the development of effective obesity prevention interventions, particularly ones that address the disproportionate burden of obesity experienced by racial/ethnic minorities.

A Pilot Study for Metabolic Profiling of Obesity-Associated Microbial Gut Dysbiosis in Male Wistar Rats

Obesity is one of the most incident and concerning disease worldwide. Definite strategies to prevent obesity and related complications remain elusive. Among the risk factors of the onset of obesity, gut microbiota might play an important role in the pathogenesis of the disease, and it has received extensive attention because it affects the host metabolism. In this study, we aimed to define a metabolic profile of the segregated obesity-associated gut dysbiosis risk factor. The study of the metabolome, in an obesity-associated gut dysbiosis model, provides a relevant way for the discrimination on the different biomarkers in the obesity onset. Thus, we developed a model of this obesity risk factors through the transference of gut microbiota from obese to non-obese male Wistar rats and performed a subsequent metabolic analysis in the receptor rats. Our results showed alterations in the lipid metabolism in plasma and in the phenylalanine metabolism in urine. In consequence, we have identified metabolic changes characterized by: (1) an increase in DG:34:2 in plasma, a decrease in hippurate, (2) an increase in 3-HPPA, and (3) an increase in o-coumaric acid. Hereby, we propose these metabolites as a metabolic profile associated to a segregated dysbiosis state related to obesity disease.

A plasma lipid signature predicts incident coronary artery disease

BACKGROUND

Dyslipidemia is a hallmark of cardiovascular disease but is characterized by crude measurements of triglycerides, HDL- and LDL cholesterol. Lipidomics enables more detailed measurements of plasma lipids, which may help improve risk stratification and understand the pathophysiology of cardiovascular disease.

METHODS

Lipidomics was used to measure 184 lipids in plasma samples from the Malmö Diet and Cancer - Cardiovascular Cohort (N = 3865), taken at baseline examination. During an average follow-up time of 20.3 years, 536 participants developed coronary artery disease (CAD). Least absolute shrinkage and selection operator (LASSO) were applied to Cox proportional hazards models in order to identify plasma lipids that predict CAD.

RESULTS

Eight plasma lipids improved prediction of future CAD on top of traditional cardiovascular risk factors. Principal component analysis of CAD-associated lipids revealed one principal component (PC2) that was associated with risk of future CAD (HR per SD increment =1.46, C·I = 1.35-1.48, P < 0.001). The risk increase for being in the highest quartile of PC2 (HR = 2.33, P < 0.001) was higher than being in the top quartile of systolic blood pressure. Addition of PC2 to traditional risk factors achieved an improvement (2%) in the area under the ROC-curve for CAD events occurring within 10 (P = 0.03), 15 (P = 0.003) and 20 (P = 0.001) years of follow-up respectively.

CONCLUSIONS

A lipid pattern improve CAD prediction above traditional risk factors, highlighting that conventional lipid-measures insufficiently describe dyslipidemia that is present years before CAD. Identifying this hidden dyslipidemia may help motivate lifestyle and pharmacological interventions early enough to reach a substantial reduction in absolute risk.

Proteomic and Metabolomic Characterization of Metabolically Healthy Obesity: A Descriptive Study from a Swedish Cohort

METHOD

Associations between different biomarkers (proteomics, lipidomics, and metabolomics) coupled to either MHO or metabolically unhealthy obese (MUO) individuals were analyzed through principal component analysis (PCA). Subjects were identified from a subsample of 416 obese individuals, selected from the Malmö Diet and Cancer study-Cardiovascular arm (MDCS-CV, n = 3,443). They were further divided into MHO (n = 143) and MUO (n = 273) defined by a history of hospitalization, or not, at baseline inclusion, and nonobese subjects (NOC, n = 3,027). Two distinctive principle components (PL2, PP5) were discovered with a significant difference and thus further investigated through their main loadings.

RESULTS

MHO individuals had a more metabolically favorable lipid and glucose profile than MUO subjects, that is, lower levels of traditional blood glucose and triglycerides, as well as a trend of lower metabolically unfavorable lipid biomarkers. PL2 (lipidomics, p=0.02) showed stronger associations of triacylglycerides with MUO, whereas phospholipids correlated with MHO. PP5 (proteomics, p=0.01) included interleukin-1 receptor antagonist (IL-1ra) and leptin with positive relations to MUO and galanin that correlated positively to MHO. The group differences in metabolite profiles were to a large extent explained by factors included in the metabolic syndrome.

CONCLUSION

Compared to MUO individuals, corresponding MHO individuals present with a more favorable lipid metabolic profile, accompanied by a downregulation of potentially harmful proteomic biomarkers. This unique and extensive biomarker profiling presents novel data on potentially differentiating traits between these two obese phenotypes.

MethylDetectR: a software for methylation-based health profiling

DNA methylation is an important biological process which involves the reversible addition of chemical tags called methyl groups to DNA and affects whether genes are active or inactive. Individual methylation profiles are determined by both genetic and environmental influences. Inter-individual variation in DNA methylation profiles can be exploited to estimate or predict a wide variety of human characteristics and disease risk profiles. Indeed, a number of methylation-based predictors of human traits have been developed and linked to important health outcomes. However, there is an unmet need to communicate the applicability and limitations of state-of-the-art methylation-based predictors to the wider community. To address this, we created a secure, web-based interactive platform called 'MethylDetectR' which calculates estimated values or scores for a variety of human traits using blood methylation data. These traits include age, lifestyle traits, high-density lipoprotein cholesterol and the levels of 27 blood proteins related to inflammatory and neurological processes and disease. Methylation-based predictors often return scores on arbitrary scales. To provide meaning to these scores, users can interactively view how estimated trait scores for a given individual compare against other individuals in the sample. Users can optionally upload binary phenotypes and investigate how estimated traits vary according to case vs. control status for these phenotypes. Users can also view how different methylation-based predictors correlate with one another, and with phenotypic values for corresponding traits in a large reference sample (n = 4,450; Generation Scotland). The 'MethylDetectR' platform allows for the fast and secure calculation of DNA methylation-derived estimates for many human traits. This platform also helps to show the correlations between methylation-based scores and corresponding traits at the level of a sample, report estimated health profiles at an individual level, demonstrate how scores relate to important binary outcomes of interest and highlight the current limitations of molecular health predictors.

Systems biology approaches to study lipidomes in health and disease

Lipids have many important biological roles, such as energy storage sources, structural components of plasma membranes and as intermediates in metabolic and signaling pathways. Lipid metabolism is under tight homeostatic control, exhibiting spatial and dynamic complexity at multiple levels. Consequently, lipid-related disturbances play important roles in the pathogenesis of most of the common diseases. Lipidomics, defined as the study of lipidomes in biological systems, has emerged as a rapidly-growing field. Due to the chemical and functional diversity of lipids, the application of a systems biology approach is essential if one is to address lipid functionality at different physiological levels. In parallel with analytical advances to measure lipids in biological matrices, the field of computational lipidomics has been rapidly advancing, enabling modeling of lipidomes in their pathway, spatial and dynamic contexts. This review focuses on recent progress in systems biology approaches to study lipids in health and disease, with specific emphasis on methodological advances and biomedical applications.

Plasma lipidomics of monozygotic twins discordant for multiple sclerosis

Blood biomarkers of multiple sclerosis (MS) can provide a better understanding of pathophysiology and enable disease monitoring. Here, we performed quantitative shotgun lipidomics on the plasma of a unique cohort of 73 monozygotic twins discordant for MS. We analyzed 243 lipid species, evaluated lipid features such as fatty acyl chain length and number of acyl chain double bonds, and detected phospholipids that were significantly altered in the plasma of co‐twins with MS compared to their non‐affected siblings. Strikingly, changes were most prominent in ether phosphatidylethanolamines and ether phosphatidylcholines, suggesting a role for altered lipid signaling in the disease.

Smith AA, Giles C, Huynh K, Cinel M, Mellett NA, Cadby G, Hung J, Hui J, Beilby J, Watts GF, Shaw JS, Moses EK, Magliano DJ, Meikle PJ. High-coverage plasma lipidomics reveals novel sex-specific lipidomic fingerprints of age and BMI: Evidence from two large population cohort studies

Obesity and related metabolic diseases show clear sex-related differences. The growing burden of these diseases calls for better understanding of the age- and sex-related metabolic consequences. High-throughput lipidomic analyses of population-based cohorts offer an opportunity to identify disease-risk-associated biomarkers and to improve our understanding of lipid metabolism and biology at a population level. Here, we comprehensively examined the relationship between lipid classes/subclasses and molecular species with age, sex, and body mass index (BMI). Furthermore, we evaluated sex specificity in the association of the plasma lipidome with age and BMI. Some 747 targeted lipid measures, representing 706 molecular lipid species across 36 classes/subclasses, were measured using a high-performance liquid chromatography coupled mass spectrometer on a total of 10,339 participants from the Australian Diabetes, Obesity and Lifestyle Study (AusDiab), with 563 lipid species being validated externally on 4,207 participants of the Busselton Health Study (BHS). Heat maps were constructed to visualise the relative differences in lipidomic profile between men and women. Multivariable linear regression analyses, including sex-interaction terms, were performed to assess the associations of lipid species with cardiometabolic phenotypes. Associations with age and sex were found for 472 (66.9%) and 583 (82.6%) lipid species, respectively. We further demonstrated that age-associated lipidomic fingerprints differed by sex. Specific classes of ether-phospholipids and lysophospholipids (calculated as the sum composition of the species within the class) were inversely associated with age in men only. In analyses with women alone, higher triacylglycerol and lower lysoalkylphosphatidylcholine species were observed among postmenopausal women compared with premenopausal women. We also identified sex-specific associations of lipid species with obesity. Lysophospholipids were negatively associated with BMI in both sexes (with a larger effect size in men), whilst acylcarnitine species showed opposing associations based on sex (positive association in women and negative association in men). Finally, by utilising specific lipid ratios as a proxy for enzymatic activity, we identified stearoyl CoA desaturase (SCD-1), fatty acid desaturase 3 (FADS3), and plasmanylethanolamine Δ1-desaturase activities, as well as the sphingolipid metabolic pathway, as constituent perturbations of cardiometabolic phenotypes. Our analyses elucidate the effect of age and sex on lipid metabolism by offering a comprehensive view of the lipidomic profiles associated with common cardiometabolic risk factors. These findings have implications for age- and sex-dependent lipid metabolism in health and disease and suggest the need for sex stratification during lipid biomarker discovery, establishing biological reference intervals for assessment of disease risk.

Pilot Multi-Omic Analysis of Human Bile from Benign and Malignant Biliary Strictures: A Machine-Learning Approach

Cholangiocarcinoma (CCA) and pancreatic adenocarcinoma (PDAC) may lead to the development of extrahepatic obstructive cholestasis. However, biliary stenoses can also be caused by benign conditions, and the identification of their etiology still remains a clinical challenge. We performed metabolomic and proteomic analyses of bile from patients with benign (n = 36) and malignant conditions, CCA (n = 36) or PDAC (n = 57), undergoing endoscopic retrograde cholangiopancreatography with the aim of characterizing bile composition in biliopancreatic disease and identifying biomarkers for the differential diagnosis of biliary strictures. Comprehensive analyses of lipids, bile acids and small molecules were carried out using mass spectrometry (MS) and nuclear magnetic resonance spectroscopy (¹H-NMR) in all patients. MS analysis of bile proteome was performed in five patients per group. We implemented artificial intelligence tools for the selection of biomarkers and algorithms with predictive capacity. Our machine-learning pipeline included the generation of synthetic data with properties of real data, the selection of potential biomarkers (metabolites or proteins) and their analysis with neural networks (NN). Selected biomarkers were then validated with real data. We identified panels of lipids (n = 10) and proteins (n = 5) that when analyzed with NN algorithms discriminated between patients with and without cancer with an unprecedented accuracy.

Toward a Standardized Strategy of Clinical Metabolomics for the Advancement of Precision Medicine

Despite the tremendous success, pitfalls have been observed in every step of a clinical metabolomics workflow, which impedes the internal validity of the study. Furthermore, the demand for logistics, instrumentations, and computational resources for metabolic phenotyping studies has far exceeded our expectations. In this conceptual review, we will cover inclusive barriers of a metabolomics-based clinical study and suggest potential solutions in the hope of enhancing study robustness, usability, and transferability. The importance of quality assurance and quality control procedures is discussed, followed by a practical rule containing five phases, including two additional "pre-pre-" and "post-post-" analytical steps. Besides, we will elucidate the potential involvement of machine learning and demonstrate that the need for automated data mining algorithms to improve the quality of future research is undeniable. Consequently, we propose a comprehensive metabolomics framework, along with an appropriate checklist refined from current guidelines and our previously published assessment, in the attempt to accurately translate achievements in metabolomics into clinical and epidemiological research. Furthermore, the integration of multifaceted multi-omics approaches with metabolomics as the pillar member is in urgent need. When combining with other social or nutritional factors, we can gather complete omics profiles for a particular disease. Our discussion reflects the current obstacles and potential solutions toward the progressing trend of utilizing metabolomics in clinical research to create the next-generation healthcare system.