Using Machine Learning to Predict Obesity Based on Genome-Wide and Epigenome-Wide Gene-Gene and Gene-Diet Interactions

Understanding and applying gene-environment interactions: a guide for nutrition professionals with an emphasis on integration in African research settings

Noncommunicable diseases (NCDs) are influenced by the interplay between genetics and environmental exposures, particularly diet. However, many healthcare professionals, including nutritionists and dietitians, have limited genetic background and, therefore, they may lack understanding of gene-environment interactions (GxEs) studies. Even researchers deeply involved in nutrition studies, but with a focus elsewhere, can struggle to interpret, evaluate, and conduct GxE studies. There is an urgent need to study African populations that bear a heavy burden of NCDs, demonstrate unique genetic variability, and have cultural practices resulting in distinctive environmental exposures compared with Europeans or Americans, who are studied more. Although diverse and rapidly changing environments, as well as the high genetic variability of Africans and difference in linkage disequilibrium (ie, certain gene variants are inherited together more often than expected by chance), provide unparalleled potential to investigate the omics fields, only a small percentage of studies come from Africa. Furthermore, research evidence lags behind the practices of companies offering genetic testing for personalized medicine and nutrition. We need to generate more evidence on GxEs that also considers continental African populations to be able to prevent unethical practices and enable tailored treatments. This review aims to introduce nutrition professionals to genetics terms and valid methods to investigate GxEs and their challenges, and proposes ways to improve quality and reproducibility. The review also provides insight into the potential contributions of nutrigenetics and nutrigenomics to the healthcare sphere, addresses direct-to-consumer genetic testing, and concludes by offering insights into the field's future, including advanced technologies like artificial intelligence and machine learning.

Data science using the human epigenome for predicting multifactorial diseases and symptoms

Tweetable abstract This article reviews machine learning models that leverages epigenomic data for predicting multifactorial diseases and symptoms as well as how such models can be utilized to explore new research questions.

As a matter of fat: Emerging roles of lipid-sensitive E3 ubiquitin ligases

The dynamic structure and composition of lipid membranes need to be tightly regulated to control the vast array of cellular processes from cell and organelle morphology to protein-protein interactions and signal transduction pathways. To maintain membrane integrity, sense-and-response systems monitor and adjust membrane lipid composition to the ever-changing cellular environment, but only a relatively small number of control systems have been described. Here, we explore the emerging role of the ubiquitin-proteasome system in monitoring and maintaining membrane lipid composition. We focus on the ER-resident RNF145 E3 ubiquitin ligase, its role in regulating adiponectin receptor 2 (ADIPOR2), its lipid hydrolase substrate, and the broader implications for understanding the homeostatic processes that fine-tune cellular membrane composition.

Rate-Limiting Enzymes in Cardiometabolic Health and Aging in Humans

INTRODUCTION

Rate-limiting enzymes (RLEs) are innate slow points in metabolic pathways, and many function in bio-processes related to nutrient sensing. Many RLEs carry causal mutations relevant to inherited metabolic disorders. Because the activity of RLEs in cardiovascular health is poorly characterized, our objective was to assess their involvement in cardiometabolic health and disease and where altered biophysical and biochemical functions can promote disease.

METHODS

A dataset of 380 human RLEs was compared to protein and gene datasets for factors likely to contribute to cardiometabolic disease, including proteins showing significant age-related altered expression in blood and genetic loci with variants that associate with common cardiometabolic phenotypes. The biochemical reactions catalyzed by RLEs were evaluated for metabolites enriched in RLE subsets associating with various cardiometabolic phenotypes. Most significance tests were based on Z-score enrichment converted to p values with a normal distribution function.

RESULTS

Of 380 RLEs analyzed, 112 function in mitochondria, and 53 are assigned to inherited metabolic disorders. There was a depletion of RLE proteins known as aging biomarkers. At the gene level, RLEs were assessed for common genetic variants that associated with important cardiometabolic traits of LDL-cholesterol or any of the five outcomes pertinent to metabolic syndrome. This revealed several RLEs with links to cardiometabolic traits, from a minimum of 26 for HDL-cholesterol to a maximum of 45 for plasma glucose. Analysis of these GWAS-linked RLEs for enrichment of the molecular constituents of the catalyzed reactions disclosed a number of significant phenotype-metabolite links. These included blood pressure with acetate (p = 2.2 × 10-4) and NADP+ (p = 0.0091), plasma HDL-cholesterol and triglyceride with diacylglycerol (p = 2.6 × 10-5, 6.4 × 10-5, respectively) and diolein (p = 2.2 × 10-6, 5.9 × 10-6), and waist circumference with d-glucosamine-6-phosphate (p = 1.8 × 10-4).

CONCLUSION

In the context of cardiometabolic health, aging, and disease, these results highlight key diet-derived metabolites that are central to specific rate-limited processes that are linked to cardiometabolic health. These metabolites include acetate and diacylglycerol, pertinent to blood pressure and triglycerides, respectively, as well as diacylglycerol and HDL-cholesterol.

The impact of alcoholic drinks and dietary factors on epigenetic markers associated with triglyceride levels

Background: Many epigenetic loci have been associated with plasma triglyceride (TG) levels, but epigenetic connections between those loci and dietary exposures are largely unknown. This study aimed to characterize the epigenetic links between diet, lifestyle, and TG. Methods: We first conducted an epigenome-wide association study (EWAS) for TG in the Framingham Heart Study Offspring population (FHS, n = 2,264). We then examined relationships between dietary and lifestyle-related variables, collected four times in 13 years, and differential DNA methylation sites (DMSs) associated with the last TG measures. Third, we conducted a mediation analysis to evaluate the causal relationships between diet-related variables and TG. Finally, we replicated three steps to validate identified DMSs associated with alcohol and carbohydrate intake in the Genetics of Lipid-Lowering Drugs and Diet Network (GOLDN) study (n = 993). Results: In the FHS, the EWAS revealed 28 TG-associated DMSs at 19 gene regions. We identified 102 unique associations between these DMSs and one or more dietary and lifestyle-related variables. Alcohol and carbohydrate intake showed the most significant and consistent associations with 11 TG-associated DMSs. Mediation analyses demonstrated that alcohol and carbohydrate intake independently affect TG via DMSs as mediators. Higher alcohol intake was associated with lower methylation at seven DMSs and higher TG. In contrast, increased carbohydrate intake was associated with higher DNA methylation at two DMSs (CPT1A and SLC7A11) and lower TG. Validation in the GOLDN further supports the findings. Conclusion: Our findings imply that TG-associated DMSs reflect dietary intakes, particularly alcoholic drinks, which could affect the current cardiometabolic risk via epigenetic changes. This study illustrates a new method to map epigenetic signatures of environmental factors for disease risk. Identification of epigenetic markers of dietary intake can provide insight into an individual's risk of cardiovascular disease and support the application of precision nutrition. Clinical Trial Registration: www.ClinicalTrials.gov, the Framingham Heart Study (FHS), NCT00005121; the Genetics of Lipid Lowering Drugs and Diet Network (GOLDN), NCT01023750.

Precision Nutrition: Recent Advances in Obesity

"Precision nutrition" is an emerging area of nutrition research that focuses on understanding metabolic variability within and between individuals and helps develop customized dietary plans and interventions to maintain optimal individual health. It encompasses nutritional genomic (gene-nutrient interactions), epigenetic, microbiome, and environmental factors. Obesity is a complex disease that is affected by genetic and environmental factors and thus a relevant target of precision nutrition-based approaches. Recent studies have shown significant associations between obesity phenotypes (body weight, body mass index, waist circumference, and central and regional adiposity) and genetic variants, epigenetic factors (DNA methylation and noncoding RNA), microbial species, and environment (sociodemographics and physical activity). Additionally, studies have also shown that the interactions between genetic variants, microbial metabolites, and epigenetic factors affect energy balance and adiposity. These include variants in FTO, MC4R, PPAR, APOA, and FADS genes, DNA methylation in CpG island regions, and specific miRNAs and microbial species such as Firmicutes, Bacteriodes, Clostridiales, etc. Similarly, studies have shown that microbial metabolites, folate, B-vitamins, and short-chain fatty acids interact with miRNAs to influence obesity phenotypes. With the advent of next-generation sequencing and analytical approaches, the advances in precision nutrition have the potential to lead to new paradigms, which can further lead to interventions or customized treatments specific to individuals or susceptible groups of individuals. This review highlights the recent advances in precision nutrition as applied to obesity and projects the importance of precision nutrition in obesity and weight management.

Predicting Overweight and Obesity Status Among Malaysian Working Adults With Machine Learning or Logistic Regression: Retrospective Comparison Study

BACKGROUND

Overweight or obesity is a primary health concern that leads to a significant burden of noncommunicable disease and threatens national productivity and economic growth. Given the complexity of the etiology of overweight or obesity, machine learning (ML) algorithms offer a promising alternative approach in disentangling interdependent factors for predicting overweight or obesity status.

OBJECTIVE

This study examined the performance of 3 ML algorithms in comparison with logistic regression (LR) to predict overweight or obesity status among working adults in Malaysia.

METHODS

Using data from 16,860 participants (mean age 34.2, SD 9.0 years; n=6904, 41% male; n=7048, 41.8% with overweight or obesity) in the Malaysia's Healthiest Workplace by AIA Vitality 2019 survey, predictor variables, including sociodemographic characteristics, job characteristics, health and weight perceptions, and lifestyle-related factors, were modeled using the extreme gradient boosting (XGBoost), random forest (RF), and support vector machine (SVM) algorithms, as well as LR, to predict overweight or obesity status based on a BMI cutoff of 25 kg/m².

RESULTS

The area under the receiver operating characteristic curve was 0.81 (95% CI 0.79-0.82), 0.80 (95% CI 0.79-0.81), 0.80 (95% CI 0.78-0.81), and 0.78 (95% CI 0.77-0.80) for the XGBoost, RF, SVM, and LR models, respectively. Weight satisfaction was the top predictor, and ethnicity, age, and gender were also consistent predictor variables of overweight or obesity status in all models.

CONCLUSIONS

Based on multi-domain online workplace survey data, this study produced predictive models that identified overweight or obesity status with moderate to high accuracy. The performance of both ML-based and logistic regression models were comparable when predicting obesity among working adults in Malaysia.

Machine learning insights concerning inflammatory and liver-related risk comorbidities in non-communicable and viral diseases

The liver is a key organ involved in a wide range of functions, whose damage can lead to chronic liver disease (CLD). CLD accounts for more than two million deaths worldwide, becoming a social and economic burden for most countries. Among the different factors that can cause CLD, alcohol abuse, viruses, drug treatments, and unhealthy dietary patterns top the list. These conditions prompt and perpetuate an inflammatory environment and oxidative stress imbalance that favor the development of hepatic fibrogenesis. High stages of fibrosis can eventually lead to cirrhosis or hepatocellular carcinoma (HCC). Despite the advances achieved in this field, new approaches are needed for the prevention, diagnosis, treatment, and prognosis of CLD. In this context, the scientific com-munity is using machine learning (ML) algorithms to integrate and process vast amounts of data with unprecedented performance. ML techniques allow the integration of anthropometric, genetic, clinical, biochemical, dietary, lifestyle and omics data, giving new insights to tackle CLD and bringing personalized medicine a step closer. This review summarizes the investigations where ML techniques have been applied to study new approaches that could be used in inflammatory-related, hepatitis viruses-induced, and coronavirus disease 2019-induced liver damage and enlighten the factors involved in CLD development.

Disease classification for whole-blood DNA methylation: Meta-analysis, missing values imputation, and XAI

BACKGROUND

DNA methylation has a significant effect on gene expression and can be associated with various diseases. Meta-analysis of available DNA methylation datasets requires development of a specific workflow for joint data processing.

RESULTS

We propose a comprehensive approach of combined DNA methylation datasets to classify controls and patients. The solution includes data harmonization, construction of machine learning classification models, dimensionality reduction of models, imputation of missing values, and explanation of model predictions by explainable artificial intelligence (XAI) algorithms. We show that harmonization can improve classification accuracy by up to 20% when preprocessing methods of the training and test datasets are different. The best accuracy results were obtained with tree ensembles, reaching above 95% for Parkinson's disease. Dimensionality reduction can substantially decrease the number of features, without detriment to the classification accuracy. The best imputation methods achieve almost the same classification accuracy for data with missing values as for the original data. XAI approaches have allowed us to explain model predictions from both populational and individual perspectives.

CONCLUSIONS

We propose a methodologically valid and comprehensive approach to the classification of healthy individuals and patients with various diseases based on whole-blood DNA methylation data using Parkinson's disease and schizophrenia as examples. The proposed algorithm works better for the former pathology, characterized by a complex set of symptoms. It allows to solve data harmonization problems for meta-analysis of many different datasets, impute missing values, and build classification models of small dimensionality.

Holistic Integration of Omics Tools for Precision Nutrition in Health and Disease

The combination of multiple omics approaches has emerged as an innovative holistic scope to provide a more comprehensive view of the molecular and physiological events underlying human diseases (including obesity, dyslipidemias, fatty liver, insulin resistance, and inflammation), as well as for elucidating unique and specific metabolic phenotypes. These omics technologies include genomics (polymorphisms and other structural genetic variants), epigenomics (DNA methylation, histone modifications, long non-coding RNA, telomere length), metagenomics (gut microbiota composition, enterotypes), transcriptomics (RNA expression patterns), proteomics (protein quantities), and metabolomics (metabolite profiles), as well as interactions with dietary/nutritional factors. Although more evidence is still necessary, it is expected that the incorporation of integrative omics could be useful not only for risk prediction and early diagnosis but also for guiding tailored dietary treatments and prognosis schemes. Some challenges include ethical and regulatory issues, the lack of robust and reproducible results due to methodological aspects, the high cost of omics methodologies, and high-dimensional data analyses and interpretation. In this review, we provide examples of system biology studies using multi-omics methodologies to unravel novel insights into the mechanisms and pathways connecting the genotype to clinically relevant traits and therapy outcomes for precision nutrition applications in health and disease.

The potential of DNA methylation as a biomarker for obesity and smoking

DNA methylation is an epigenetic modification that has consistently been shown to be linked with a variety of human traits and diseases. Because DNA methylation is dynamic and potentially reversible in nature and can reflect environmental exposures and predict the onset of diseases, it has piqued interest as a potential disease biomarker. DNA methylation patterns are more stable than transcriptomic or proteomic patterns, and they are relatively easy to measure to track exposure to different environments and risk factors. Importantly, technologies for DNA methylation quantification have become increasingly cost effective-accelerating new research in the field-and have enabled the development of novel DNA methylation biomarkers. Quite a few DNA methylation-based predictors for a number of traits and diseases already exist. Such predictors show potential for being more accurate than self-reported or measured phenotypes (such as smoking behavior and body mass index) and may even hold potential for applications in clinics. In this review, we will first discuss the advantages and challenges of DNA methylation biomarkers in general. We will then review the current state and future potential of DNA methylation biomarkers in two human traits that show rather consistent alterations in methylome-obesity and smoking. Lastly, we will briefly speculate about the future prospects of DNA methylation biomarkers, and possible ways to achieve them.

Genetic and epigenetic nutritional interactions influencing obesity risk and adiposity outcomes

PURPOSE OF REVIEW

This article aims to critically overview the current interplay of genetic/epigenetic factors and several nutritional aspects influencing obesity susceptibility and adiposity outcomes for obesity management and weight status monitoring.

RECENT FINDINGS

Single nucleotide polymorphisms located in or near genes participating in energy homeostasis, fatty acid metabolism, appetite control, brain regulation, and thermogenesis have been associated with body composition measures (body weight, body mass index, waist circumference, body fat percentage, and visceral adipose tissue) depending on nutrient intakes, dietary patterns, and eating behaviors. Moreover, studies analyzing interactions between the epigenome and dietary intakes in relation to adiposity outcomes are reported. The main epigenetic mechanisms include methylation levels of promoter sequences, telomere length, and micro-ribonucleic acid expression profiles, whereas covalent histone modifications remain less studied.

SUMMARY

Exploring potential interactions between the genetic/epigenetic background and nutritional features is improving the current understanding of the obesity physiopathogenesis and the usefulness of translating this precision information in the clinical setting for weight gain prediction, the design of personalized nutrition therapies as well as individual responsiveness estimation to dietary advice. The analysis of further relationships between the genotype, the epigenotype and other precision markers including the gut microbiota and the metabolome is warranted.

Association between alcohol consumption and DNA methylation in blood: a systematic review of observational studies

Aim: We systematically reviewed and evaluated current literature on alcohol consumption and DNA methylation (DNAm) at the genome-wide and probe-wise level in blood of adults. Materials & methods: Five databases (PubMed, Embase, Web of Science, CINAHL and PsycInfo) were searched until 20 December 2020. Studies assessing the effect of alcohol dependence on DNAm were not eligible. Results: 11 cross-sectional studies were included with 88 to 9643 participants. Overall, all studies had a risk of bias criteria unclear or unmet. Epigenome-wide association studies identified between 0 and 5458 differentially methylated positions, and 15 were observed in at least four studies. Conclusion: Potential methylation markers for alcohol consumption have been identified, but further validation in large cohorts is needed.