Integrated genetic and epigenetic prediction of coronary heart disease in the Framingham Heart Study

ZSCAN25 methylation predicts seizures and severe alcohol withdrawal syndrome

Currently, clinicians use their judgement and indices such as the Prediction of Alcohol Withdrawal Syndrome Scale (PAWSS) to determine whether patients are admitted to hospitals for consideration of withdrawal syndrome (AWS). However, only a fraction of those admitted will experience severe AWS. Previously, we and others have shown that epigenetic indices, such as the Alcohol T-Score (ATS), can quantify recent alcohol consumption. However, whether these or other alcohol biomarkers, such as carbohydrate deficient transferrin (CDT), could identify those at risk for severe AWS is unknown. To determine this, we first conducted genome-wide DNA methylation analyses of subjects entering and exiting alcohol treatment to identify loci whose methylation quickly reverted as a function of abstinence. We then tested whether methylation at a rapidly reverting locus, cg07375256, or other existing metrics including PAWSS scores, CDT levels, or ATS, could predict outcome in 125 subjects admitted for consideration of AWS. We found that PAWSS did not significantly predict severe AWS nor seizures. However, methylation at cg07375256 (ZSCAN25) and CDT strongly predicted severe AWS with ATS (p < 0.007) and cg07375256 (p < 6 × 10-5) methylation also predicting AWS associated seizures. We conclude that epigenetic methods can predict those likely to experience severe AWS and that the use of these or similar Precision Epigenetic approaches could better guide AWS management.

A review of disease risk prediction methods and applications in the omics era

Risk prediction and disease prevention are the innovative care challenges of the 21st century. Apart from freeing the individual from the pain of disease, it will lead to low medical costs for society. Until very recently, risk assessments have ushered in a new era with the emergence of omics technologies, including genomics, transcriptomics, epigenomics, proteomics, and so on, which potentially advance the ability of biomarkers to aid prediction models. While risk prediction has achieved great success, there are still some challenges and limitations. We reviewed the general process of omics-based disease risk model construction and the applications in four typical diseases. Meanwhile, we highlighted the problems in current studies and explored the potential opportunities and challenges for future clinical practice.

Applications of Machine Learning (ML) and Mathematical Modeling (MM) in Healthcare with Special Focus on Cancer Prognosis and Anticancer Therapy: Current Status and Challenges

The use of data-driven high-throughput analytical techniques, which has given rise to computational oncology, is undisputed. The widespread use of machine learning (ML) and mathematical modeling (MM)-based techniques is widely acknowledged. These two approaches have fueled the advancement in cancer research and eventually led to the uptake of telemedicine in cancer care. For diagnostic, prognostic, and treatment purposes concerning different types of cancer research, vast databases of varied information with manifold dimensions are required, and indeed, all this information can only be managed by an automated system developed utilizing ML and MM. In addition, MM is being used to probe the relationship between the pharmacokinetics and pharmacodynamics (PK/PD interactions) of anti-cancer substances to improve cancer treatment, and also to refine the quality of existing treatment models by being incorporated at all steps of research and development related to cancer and in routine patient care. This review will serve as a consolidation of the advancement and benefits of ML and MM techniques with a special focus on the area of cancer prognosis and anticancer therapy, leading to the identification of challenges (data quantity, ethical consideration, and data privacy) which are yet to be fully addressed in current studies.

Machine Learning in Cardiovascular Risk Prediction and Precision Preventive Approaches

PURPOSE OF REVIEW

In this review, we sought to provide an overview of ML and focus on the contemporary applications of ML in cardiovascular risk prediction and precision preventive approaches. We end the review by highlighting the limitations of ML while projecting on the potential of ML in assimilating these multifaceted aspects of CAD in order to improve patient-level outcomes and further population health.

RECENT FINDINGS

Coronary artery disease (CAD) is estimated to affect 20.5 million adults across the USA, while also impacting a significant burden at the socio-economic level. While the knowledge of the mechanistic pathways that govern the onset and progression of clinical CAD has improved over the past decade, contemporary patient-level risk models lag in accuracy and utility. Recently, there has been renewed interest in combining advanced analytic techniques that utilize artificial intelligence (AI) with a big data approach in order to improve risk prediction within the realm of CAD. By virtue of being able to combine diverse amounts of multidimensional horizontal data, machine learning has been employed to build models for improved risk prediction and personalized patient care approaches. The use of ML-based algorithms has been used to leverage individualized patient-specific data and the associated metabolic/genomic profile to improve CAD risk assessment. While the tool can be visualized to shift the paradigm toward a patient-specific care, it is crucial to acknowledge and address several challenges inherent to ML and its integration into healthcare before it can be significantly incorporated in the daily clinical practice.

Validation of an Integrated Genetic-Epigenetic Test for the Assessment of Coronary Heart Disease

BACKGROUND

Coronary heart disease (CHD) is the leading cause of death in the world. Unfortunately, many of the key diagnostic tools for CHD are insensitive, invasive, and costly; require significant specialized infrastructure investments; and do not provide information to guide postdiagnosis therapy. In prior work using data from the Framingham Heart Study, we provided in silico evidence that integrated genetic-epigenetic tools may provide a new avenue for assessing CHD.

METHODS AND RESULTS

In this communication, we use an improved machine learning approach and data from 2 additional cohorts, totaling 449 cases and 2067 controls, to develop a better model for ascertaining symptomatic CHD. Using the DNA from the 2 new cohorts, we translate and validate the in silico findings into an artificial intelligence-guided, clinically implementable method that uses input from 6 methylation-sensitive digital polymerase chain reaction and 10 genotyping assays. Using this method, the overall average area under the curve, sensitivity, and specificity in the 3 test cohorts is 82%, 79%, and 76%, respectively. Analysis of targeted cytosine-phospho-guanine loci shows that they map to key risk pathways involved in atherosclerosis that suggest specific therapeutic approaches.

CONCLUSIONS

We conclude that this scalable integrated genetic-epigenetic approach is useful for the diagnosis of symptomatic CHD, performs favorably as compared with many existing methods, and may provide personalized insight to CHD therapy. Furthermore, given the dynamic nature of DNA methylation and the ease of methylation-sensitive digital polymerase chain reaction methodologies, these findings may pave a pathway for precision epigenetic approaches for monitoring CHD treatment response.

Precision epigenetics provides a scalable pathway for improving coronary heart disease care globally

Coronary heart disease (CHD) is the world's leading cause of death. Up to 90% of all CHD deaths are preventable, but effective prevention of this mortality requires more scalable, precise methods for assessing CHD status and monitoring treatment response. Unfortunately, current diagnostic methods have barriers to implementation, particularly in rural areas and lower-income countries. This gap may be bridged by highly scalable advances in DNA methylation testing methods and artificial intelligence. Herein, we review prior studies of CHD related to methylation alone and in combination with other biovariables. We compare these new methods with established methods for diagnosing CHD. Finally, we outline pathways through which methylation-based testing methods may allow the democratization of improved methods for assessing CHD globally.

Epigenetics of Early Cardiometabolic Disease: Mechanisms and Precision Medicine

Epigenetics has transformed our understanding of the molecular basis of complex diseases, including cardiovascular and metabolic disorders. This review offers a comprehensive overview of the current state of knowledge on epigenetic processes implicated in cardiovascular and metabolic diseases, highlighting the potential of DNA methylation as a precision medicine biomarker and examining the impact of social determinants of health, gut bacterial epigenomics, noncoding RNA, and epitranscriptomics on disease development and progression. We discuss challenges and barriers to advancing cardiometabolic epigenetics research, along with the opportunities for novel preventive strategies, targeted therapies, and personalized medicine approaches that may arise from a better understanding of epigenetic processes. Emerging technologies, such as single-cell sequencing and epigenetic editing, hold the potential to further enhance our ability to dissect the complex interplay between genetic, environmental, and lifestyle factors. To translate research findings into clinical practice, interdisciplinary collaborations, technical and ethical considerations, and accessibility of resources and knowledge are crucial. Ultimately, the field of epigenetics has the potential to revolutionize the way we approach cardiovascular and metabolic diseases, paving the way for precision medicine and personalized health care, and improving the lives of millions of individuals worldwide affected by these conditions.

The Reversion of the Epigenetic Signature of Coronary Heart Disease in Response to Smoking Cessation

Coronary heart disease (CHD) is the leading cause of death worldwide. However, current diagnostic tools for CHD, such as coronary computed tomography angiography (CCTA), are poorly suited for monitoring treatment response. Recently, we have introduced an artificial-intelligence-guided integrated genetic-epigenetic test for CHD whose core consists of six assays that determine methylation in pathways known to moderate the pathogenesis of CHD. However, whether methylation at these six loci is sufficiently dynamic to guide CHD treatment response is unknown. To test that hypothesis, we examined the relationship of changes in these six loci to changes in cg05575921, a generally accepted marker of smoking intensity, using DNA from a cohort of 39 subjects undergoing a 90-day smoking cessation intervention and methylation-sensitive digital PCR (MSdPCR). We found that changes in epigenetic smoking intensity were significantly associated with reversion of the CHD-associated methylation signature at five of the six MSdPCR predictor sites: cg03725309, cg12586707, cg04988978, cg17901584, and cg21161138. We conclude that methylation-based approaches could be a scalable method for assessing the clinical effectiveness of CHD interventions, and that further studies to understand the responsiveness of these epigenetic measures to other forms of CHD treatment are in order.

Artificial intelligence in cardiovascular prevention: new ways will open new doors

Prevention and effective treatment of cardiovascular disease are progressive issues that grow in tandem with the average age of the world population. Over recent decades, the potential role of artificial intelligence in cardiovascular medicine has been increasingly recognized because of the incredible amount of real-world data (RWD) regarding patient health status and healthcare delivery that can be collated from a variety of sources wherein patient information is routinely collected, including patient registries, clinical case reports, reimbursement claims and billing reports, medical devices, and electronic health records. Like any other (health) data, RWD can be analysed in accordance with high-quality research methods, and its analysis can deliver valuable patient-centric insights complementing the information obtained from conventional clinical trials. Artificial intelligence application on RWD has the potential to detect a patient's health trajectory leading to personalized medicine and tailored treatment. This article reviews the benefits of artificial intelligence in cardiovascular prevention and management, focusing on diagnostic and therapeutic improvements without neglecting the limitations of this new scientific approach.

A comprehensive investigation of statistical and machine learning approaches for predicting complex human diseases on genomic variants

Quantifying an individual's risk for common diseases is an important goal of precision health. The polygenic risk score (PRS), which aggregates multiple risk alleles of candidate diseases, has emerged as a standard approach for identifying high-risk individuals. Although several studies have been performed to benchmark the PRS calculation tools and assess their potential to guide future clinical applications, some issues remain to be further investigated, such as lacking (i) various simulated data with different genetic effects; (ii) evaluation of machine learning models and (iii) evaluation on multiple ancestries studies. In this study, we systematically validated and compared 13 statistical methods, 5 machine learning models and 2 ensemble models using simulated data with additive and genetic interaction models, 22 common diseases with internal training sets, 4 common diseases with external summary statistics and 3 common diseases for trans-ancestry studies in UK Biobank. The statistical methods were better in simulated data from additive models and machine learning models have edges for data that include genetic interactions. Ensemble models are generally the best choice by integrating various statistical methods. LDpred2 outperformed the other standalone tools, whereas PRS-CS, lassosum and DBSLMM showed comparable performance. We also identified that disease heritability strongly affected the predictive performance of all methods. Both the number and effect sizes of risk SNPs are important; and sample size strongly influences the performance of all methods. For the trans-ancestry studies, we found that the performance of most methods became worse when training and testing sets were from different populations.

Towards AI-driven longevity research: An overview

While in the past technology has mostly been utilized to store information about the structural configuration of proteins and molecules for research and medical purposes, Artificial Intelligence is nowadays able to learn from the existing data how to predict and model properties and interactions, revealing important knowledge about complex biological processes, such as aging. Modern technologies, moreover, can rely on a broader set of information, including those derived from the next-generation sequencing (e.g., proteomics, lipidomics, and other omics), to understand the interactions between human body and the external environment. This is especially relevant as external factors have been shown to have a key role in aging. As the field of computational systems biology keeps improving and new biomarkers of aging are being developed, artificial intelligence promises to become a major ally of aging research.

The Interplay of Epigenetic, Genetic, and Traditional Risk Factors on Blood Pressure: Findings from the Health and Retirement Study

The epigenome likely interacts with traditional and genetic risk factors to influence blood pressure. We evaluated whether 13 previously reported DNA methylation sites (CpGs) are associated with systolic (SBP) or diastolic (DBP) blood pressure, both individually and aggregated into methylation risk scores (MRS), in 3070 participants (including 437 African ancestry (AA) and 2021 European ancestry (EA), mean age = 70.5 years) from the Health and Retirement Study. Nine CpGs were at least nominally associated with SBP and/or DBP after adjusting for traditional hypertension risk factors (p < 0.05). MRS_SBP was positively associated with SBP in the full sample (β = 1.7 mmHg per 1 standard deviation in MRS_SBP; p = 2.7 × 10^-5) and in EA (β = 1.6; p = 0.001), and MRS_DBP with DBP in the full sample (β = 1.1; p = 1.8 × 10^-6), EA (β = 1.1; p = 7.2 × 10^-5), and AA (β = 1.4; p = 0.03). The MRS and BP-genetic risk scores were independently associated with blood pressure in EA. The effects of both MRSs were weaker with increased age (p_interaction < 0.01), and the effect of MRS_DBP was higher among individuals with at least some college education (p_interaction = 0.02). In AA, increasing MRS_SBP was associated with higher SBP in females only (p_interaction = 0.01). Our work shows that MRS is a potential biomarker of blood pressure that may be modified by traditional hypertension risk factors.

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.

Do Loneliness and Per Capita Income Combine to Increase the Pace of Biological Aging for Black Adults across Late Middle Age?

In a sample of 685 late middle-aged Black adults (M age at 2019 = 57.17 years), we examined the effects of loneliness and per capita income on accelerated aging using a newly developed DNA-methylation based index: the DunedinPACE. First, using linear, mixed effects regression in a growth curve framework, we found that change in DunedinPACE was dependent on age, with a linear model best fitting the data (b = 0.004, p < 0.001), indicating that average pace of change increased among older participants. A quadratic effect was also tested, but was non-significant. Beyond the effect of age, both change in loneliness (b = 0.009, p < 0.05) and change in per capita income (b = -0.016, p < 0.001) were significantly associated with change in DunedinPACE across an 11-year period, accounting for significant between person variability observed in the unconditional model. Including non-self-report indices of smoking and alcohol use did not reduce the association of loneliness or per capita income with DunedinPACE. However, change in smoking was strongly associated with change in DunedinPACE such that those reducing their smoking aged less rapidly than those continuing to smoke. In addition, both loneliness and per capita income were associated with DunedinPACE after controlling for variation in cell-types.

Identification of DNA methylation-regulated genes as potential biomarkers for coronary heart disease via machine learning in the Framingham Heart Study

Background

DNA methylation-regulated genes have been demonstrated as the crucial participants in the occurrence of coronary heart disease (CHD). The machine learning based on DNA methylation-regulated genes has tremendous potential for mining non-invasive predictive biomarkers and exploring underlying new mechanisms of CHD.

Results

First, the 2085 age-gender-matched individuals in Framingham Heart Study (FHS) were randomly divided into training set and validation set. We then integrated methylome and transcriptome data of peripheral blood leukocytes (PBLs) from the training set to probe into the methylation and expression patterns of CHD-related genes. A total of five hub DNA methylation-regulated genes were identified in CHD through dimensionality reduction, including ATG7, BACH2, CDKN1B, DHCR24 and MPO. Subsequently, methylation and expression features of the hub DNA methylation-regulated genes were used to construct machine learning models for CHD prediction by LightGBM, XGBoost and Random Forest. The optimal model established by LightGBM exhibited favorable predictive capacity, whose AUC, sensitivity, and specificity were 0.834, 0.672, 0.864 in the validation set, respectively. Furthermore, the methylation and expression statuses of the hub genes were verified in monocytes using methylation microarray and transcriptome sequencing. The methylation statuses of ATG7, DHCR24 and MPO and the expression statuses of ATG7, BACH2 and DHCR24 in monocytes of our study population were consistent with those in PBLs from FHS.

Conclusions

We identified five DNA methylation-regulated genes based on a predictive model for CHD using machine learning, which may clue the new epigenetic mechanism for CHD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13148-022-01343-2.

DNA methylation-based predictors of health: applications and statistical considerations

DNA methylation data have become a valuable source of information for biomarker development, because, unlike static genetic risk estimates, DNA methylation varies dynamically in relation to diverse exogenous and endogenous factors, including environmental risk factors and complex disease pathology. Reliable methods for genome-wide measurement at scale have led to the proliferation of epigenome-wide association studies and subsequently to the development of DNA methylation-based predictors across a wide range of health-related applications, from the identification of risk factors or exposures, such as age and smoking, to early detection of disease or progression in cancer, cardiovascular and neurological disease. This Review evaluates the progress of existing DNA methylation-based predictors, including the contribution of machine learning techniques, and assesses the uptake of key statistical best practices needed to ensure their reliable performance, such as data-driven feature selection, elimination of data leakage in performance estimates and use of generalizable, adequately powered training samples.

Tissue-Specific Methylation Biosignatures for Monitoring Diseases: An In Silico Approach

Tissue-specific gene methylation events are key to the pathogenesis of several diseases and can be utilized for diagnosis and monitoring. Here, we established an in silico pipeline to analyze high-throughput methylome datasets to identify specific methylation fingerprints in three pathological entities of major burden, i.e., breast cancer (BrCa), osteoarthritis (OA) and diabetes mellitus (DM). Differential methylation analysis was conducted to compare tissues/cells related to the pathology and different types of healthy tissues, revealing Differentially Methylated Genes (DMGs). Highly performing and low feature number biosignatures were built with automated machine learning, including: (1) a five-gene biosignature discriminating BrCa tissue from healthy tissues (AUC 0.987 and precision 0.987), (2) three equivalent OA cartilage-specific biosignatures containing four genes each (AUC 0.978 and precision 0.986) and (3) a four-gene pancreatic β-cell-specific biosignature (AUC 0.984 and precision 0.995). Next, the BrCa biosignature was validated using an independent ccfDNA dataset showing an AUC and precision of 1.000, verifying the biosignature’s applicability in liquid biopsy. Functional and protein interaction prediction analysis revealed that most DMGs identified are involved in pathways known to be related to the studied diseases or pointed to new ones. Overall, our data-driven approach contributes to the maximum exploitation of high-throughput methylome readings, helping to establish specific disease profiles to be applied in clinical practice and to understand human pathology.

Artificial Intelligence and Cardiovascular Genetics

Polygenic diseases, which are genetic disorders caused by the combined action of multiple genes, pose unique and significant challenges for the diagnosis and management of affected patients. A major goal of cardiovascular medicine has been to understand how genetic variation leads to the clinical heterogeneity seen in polygenic cardiovascular diseases (CVDs). Recent advances and emerging technologies in artificial intelligence (AI), coupled with the ever-increasing availability of next generation sequencing (NGS) technologies, now provide researchers with unprecedented possibilities for dynamic and complex biological genomic analyses. Combining these technologies may lead to a deeper understanding of heterogeneous polygenic CVDs, better prognostic guidance, and, ultimately, greater personalized medicine. Advances will likely be achieved through increasingly frequent and robust genomic characterization of patients, as well the integration of genomic data with other clinical data, such as cardiac imaging, coronary angiography, and clinical biomarkers. This review discusses the current opportunities and limitations of genomics; provides a brief overview of AI; and identifies the current applications, limitations, and future directions of AI in genomics.

Current and Future Applications of Artificial Intelligence in Coronary Artery Disease

Cardiovascular diseases (CVDs) carry significant morbidity and mortality and are associated with substantial economic burden on healthcare systems around the world. Coronary artery disease, as one disease entity under the CVDs umbrella, had a prevalence of 7.2% among adults in the United States and incurred a financial burden of 360 billion US dollars in the years 2016–2017. The introduction of artificial intelligence (AI) and machine learning over the last two decades has unlocked new dimensions in the field of cardiovascular medicine. From automatic interpretations of heart rhythm disorders via smartwatches, to assisting in complex decision-making, AI has quickly expanded its realms in medicine and has demonstrated itself as a promising tool in helping clinicians guide treatment decisions. Understanding complex genetic interactions and developing clinical risk prediction models, advanced cardiac imaging, and improving mortality outcomes are just a few areas where AI has been applied in the domain of coronary artery disease. Through this review, we sought to summarize the advances in AI relating to coronary artery disease, current limitations, and future perspectives.

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

Obesity is associated with many chronic diseases that impair healthy aging and is governed by genetic, epigenetic, and environmental factors and their complex interactions. This study aimed to develop a model that predicts an individual's risk of obesity by better characterizing these complex relations and interactions focusing on dietary factors. For this purpose, we conducted a combined genome-wide and epigenome-wide scan for body mass index (BMI) and up to three-way interactions among 402,793 single nucleotide polymorphisms (SNPs), 415,202 DNA methylation sites (DMSs), and 397 dietary and lifestyle factors using the generalized multifactor dimensionality reduction (GMDR) method. The training set consisted of 1,573 participants in exam 8 of the Framingham Offspring Study (FOS) cohort. After identifying genetic, epigenetic, and dietary factors that passed statistical significance, we applied machine learning (ML) algorithms to predict participants' obesity status in the test set, taken as a subset of independent samples (n = 394) from the same cohort. The quality and accuracy of prediction models were evaluated using the area under the receiver operating characteristic curve (ROC-AUC). GMDR identified 213 SNPs, 530 DMSs, and 49 dietary and lifestyle factors as significant predictors of obesity. Comparing several ML algorithms, we found that the stochastic gradient boosting model provided the best prediction accuracy for obesity with an overall accuracy of 70%, with ROC-AUC of 0.72 in test set samples. Top predictors of the best-fit model were 21 SNPs, 230 DMSs in genes such as CPT1A, ABCG1, SLC7A11, RNF145, and SREBF1, and 26 dietary factors, including processed meat, diet soda, French fries, high-fat dairy, artificial sweeteners, alcohol intake, and specific nutrients and food components, such as calcium and flavonols. In conclusion, we developed an integrated approach with ML to predict obesity using omics and dietary data. This extends our knowledge of the drivers of obesity, which can inform precision nutrition strategies for the prevention and treatment of obesity. Clinical Trial Registration: [www.ClinicalTrials.gov], the Framingham Heart Study (FHS), [NCT00005121].

Machine Learning in Epigenomics: Insights into Cancer Biology and Medicine

The recent deluge of genome-wide technologies for the mapping of the epigenome and resulting data in cancer samples has provided the opportunity for gaining insights into and understanding the roles of epigenetic processes in cancer. However, the complexity, high-dimensionality, sparsity, and noise associated with these data pose challenges for extensive integrative analyses. Machine Learning (ML) algorithms are particularly suited for epigenomic data analyses due to their flexibility and ability to learn underlying hidden structures. We will discuss four overlapping but distinct major categories under ML: dimensionality reduction, unsupervised methods, supervised methods, and deep learning (DL). We review the preferred use cases of these algorithms in analyses of cancer epigenomics data with the hope to provide an overview of how ML approaches can be used to explore fundamental questions on the roles of epigenome in cancer biology and medicine.

DNA methylation signatures of incident coronary heart disease: findings from epigenome-wide association studies

Coronary heart disease (CHD) is a type of cardiovascular disease (CVD) that affects the coronary arteries, which provide oxygenated blood to the heart. It is a major cause of mortality worldwide. Various prediction methods have been developed to assess the likelihood of developing CHD, including those based on clinical features and genetic variation. Recent epigenome-wide studies have identified DNA methylation signatures associated with the development of CHD, indicating that DNA methylation may play a role in predicting future CHD. This narrative review summarises recent findings from DNA methylation studies of incident CHD (iCHD) events from epigenome-wide association studies (EWASs). The results suggest that DNA methylation signatures may identify new mechanisms involved in CHD progression and could prove a useful adjunct for the prediction of future CHD.

Epigenetic Age Acceleration Reflects Long-Term Cardiovascular Health

[Figure: see text].

WITHDRAWN: Health risk appraisal of rural population in poverty

Ahead of Print article withdrawn by publisher.

External validation of integrated genetic-epigenetic biomarkers for predicting incident coronary heart disease

Aim: The Framingham Risk Score (FRS) and atherosclerotic cardiovascular disease (ASCVD) Pooled Cohort Equation (PCE) for predicting risk for incident coronary heart disease (CHD) work poorly. To improve risk stratification for CHD, we developed a novel integrated genetic-epigenetic tool. Materials & methods: Using machine learning techniques and datasets from the Framingham Heart Study (FHS) and Intermountain Healthcare (IM), we developed and validated an integrated genetic-epigenetic model for predicting 3-year incident CHD. Results: Our approach was more sensitive than FRS and PCE and had high generalizability across cohorts. It performed with sensitivity/specificity of 79/75% in the FHS test set and 75/72% in the IM set. The sensitivity/specificity was 15/93% in FHS and 31/89% in IM for FRS, and sensitivity/specificity was 41/74% in FHS and 69/55% in IM for PCE. Conclusion: The use of our tool in a clinical setting could better identify patients at high risk for a heart attack.

Genetic information improves the prediction of major adverse cardiovascular events in the GENEMACOR population

The inclusion of a genetic risk score (GRS) can modify the risk prediction of coronary artery disease (CAD), providing an advantage over the use of traditional models. The predictive value of the genetic information on the recurrence of major adverse cardiovascular events (MACE) remains controversial. A total of 33 genetic variants previously associated with CAD were genotyped in 1587 CAD patients from the GENEMACOR study. Of these, 18 variants presented an hazard ratio >1, so they were selected to construct a weighted GRS (wGRS). MACE discrimination and reclassification were evaluated by C-Statistic, Net Reclassification Index and Integrated Discrimination Improvement methodologies. After the addition of wGRS to traditional predictors, the C-index increased from 0.566 to 0.572 (p=0.0003). Subsequently, adding wGRS to traditional plus clinical risk factors, this model slightly improved from 0.620 to 0.622 but with statistical significance (p=0.004). NRI showed that 17.9% of the cohort was better reclassified when the primary model was associated with wGRS. The Kaplan-Meier estimator showed that, at 15-year follow-up, the group with a higher number of risk alleles had a significantly higher MACE occurrence (p=0.011). In CAD patients, wGRS improved MACE risk prediction, discrimination and reclassification over the conventional factors, providing better cost-effective therapeutic strategies.

Comparison of genetic risk prediction models to improve prediction of coronary heart disease in two large cohorts of the MONICA/KORA study

It is still unclear how genetic information, provided as single-nucleotide polymorphisms (SNPs), can be most effectively integrated into risk prediction models for coronary heart disease (CHD) to add significant predictive value beyond clinical risk models. For the present study, a population-based case-cohort was used as a trainingset (451 incident cases, 1488 noncases) and an independent cohort as testset (160 incident cases, 2749 noncases). The following strategies to quantify genetic information were compared: A weighted genetic risk score including Metabochip SNPs associated with CHD in the literature (GRS_Metabo ); selection of the most predictive SNPs among these literature-confirmed variants using priority-Lasso (PL_Metabo ); validation of two comprehensive polygenic risk scores: GRS_Gola based on Metabochip data, and GRS_Khera (available in the testset only) based on cross-validated genome-wide genotyping data. We used Cox regression to assess associations with incident CHD. C-index, category-free net reclassification index (cfNRI) and relative integrated discrimination improvement (IDI_rel ) were used to quantify the predictive performance of genetic information beyond Framingham risk score variables. In contrast to GRS_Metabo and PL_Metabo , GRS_Gola significantly improved the prediction (delta C-index [95% confidence interval]: 0.0087 [0.0044, 0.0130]; IDI_rel : 0.0509 [0.0131, 0.0894]; cfNRI improved only in cases: 0.1761 [0.0253, 0.3219]). GRS_Khera yielded slightly worse prediction results than GRS_Gola .

Panomics: New Databases for Advancing Cardiology

The multifactorial nature of cardiology makes it challenging to separate noisy signals from confounders and real markers or drivers of disease. Panomics, the combination of various omic methods, provides the deepest insights into the underlying biological mechanisms to develop tools for personalized medicine under a systems biology approach. Questions remain about current findings and anticipated developments of omics. Here, we search for omic databases, investigate the types of data they provide, and give some examples of panomic applications in health care. We identified 104 omic databases, of which 72 met the inclusion criteria: genomic and clinical measurements on a subset of the database population plus one or more omic datasets. Of those, 65 were methylomic, 59 transcriptomic, 41 proteomic, 42 metabolomic, and 22 microbiomic databases. Larger database sample sizes and longer follow-up are often better suited for panomic analyses due to statistical power calculations. They are often more complete, which is important when dealing with large biological variability. Thus, the UK BioBank rises as the most comprehensive panomic resource, at present, but certain study designs may benefit from other databases.

DNA methylation and gene expression integration in cardiovascular disease

BACKGROUND

The integration of different layers of omics information is an opportunity to tackle the complexity of cardiovascular diseases (CVD) and to identify new predictive biomarkers and potential therapeutic targets. Our aim was to integrate DNA methylation and gene expression data in an effort to identify biomarkers related to cardiovascular disease risk in a community-based population. We accessed data from the Framingham Offspring Study, a cohort study with data on DNA methylation (Infinium HumanMethylation450 BeadChip; Illumina) and gene expression (Human Exon 1.0 ST Array; Affymetrix). Using the MOFA2 R package, we integrated these data to identify biomarkers related to the risk of presenting a cardiovascular event.

RESULTS

Four independent latent factors (9, 19, 21-only in women-and 27), driven by DNA methylation, were associated with cardiovascular disease independently of classical risk factors and cell-type counts. In a sensitivity analysis, we also identified factor 21 as associated with CVD in women. Factors 9, 21 and 27 were also associated with coronary heart disease risk. Moreover, in a replication effort in an independent study three of the genes included in factor 27 were also present in a factor identified to be associated with myocardial infarction (CDC42BPB, MAN2A2 and RPTOR). Factor 9 was related to age and cell-type proportions; factor 19 was related to age and B cells count; factor 21 pointed to human immunodeficiency virus infection-related pathways and inflammation; and factor 27 was related to lifestyle factors such as alcohol consumption, smoking and body mass index. Inclusion of factor 21 (only in women) improved the discriminative and reclassification capacity of the Framingham classical risk function and factor 27 improved its discrimination.

CONCLUSIONS

Unsupervised multi-omics data integration methods have the potential to provide insights into the pathogenesis of cardiovascular diseases. We identified four independent factors (one only in women) pointing to inflammation, endothelium homeostasis, visceral fat, cardiac remodeling and lifestyles as key players in the determination of cardiovascular risk. Moreover, two of these factors improved the predictive capacity of a classical risk function.

A hybrid cost-sensitive ensemble for heart disease prediction

BACKGROUND

Heart disease is the primary cause of morbidity and mortality in the world. It includes numerous problems and symptoms. The diagnosis of heart disease is difficult because there are too many factors to analyze. What's more, the misclassification cost could be very high.

METHODS

A cost-sensitive ensemble method was proposed to improve the efficiency of diagnosis and reduce the misclassification cost. The proposed method contains five heterogeneous classifiers: random forest, logistic regression, support vector machine, extreme learning machine and k-nearest neighbor. T-test was used to investigate if the performance of the ensemble was better than individual classifiers and the contribution of Relief algorithm.

RESULTS

The best performance was achieved by the proposed method according to ten-fold cross validation. The statistical tests demonstrated that the performance of the proposed ensemble was significantly superior to individual classifiers, and the efficiency of classification was distinctively improved by Relief algorithm.

CONCLUSIONS

The proposed ensemble gained significantly better results compared with individual classifiers and previous studies, which implies that it can be used as a promising alternative tool in medical decision making for heart disease diagnosis.

Associations of combined genetic and epigenetic scores with muscle size and muscle strength: a pilot study in older women

BACKGROUND

Inter-individual variance in skeletal muscle is closely related to genetic architecture and epigenetic regulation. Studies have examined genetic and epigenetic relationships with characteristics of ageing muscle separately, while no study has combined both genetic and epigenetic profiles in ageing muscle research. The aim of this study was to evaluate the association between combined genetic and methylation scores and skeletal muscle phenotypes in older women.

METHODS

Forty-eight older Caucasian women (aged 65-79 years) were included in this study. Biceps brachii thickness and vastus lateralis anatomical cross-sectional area (ACSA_VL ) were measured by ultrasonography. Maximum isometric elbow flexion (MVC_EF ) and knee extension (MVC_KE ) torques were measured by a customized dynamometer. The muscle-driven genetic predisposition score (GPS_SNP ) was calculated based on seven muscle-related single nucleotide polymorphisms (SNPs). DNA methylation levels of whole blood samples were analysed using Infinium MethylationEPIC BeadChip arrays. The DNA methylation score was calculated as a weighted sum of methylation levels of sarcopenia-driven CpG sites (MS_SAR ) or an overall gene-wise methylation score (MS_SNP , the mean methylation level of CpG sites located in muscle-related genes). Linear regression models were built to study genetic and epigenetic associations with muscle size and strength. Three models were built with both genetic and methylation scores: (1) MS_SAR  + GPS_SNP , (2) MS_SNP  + GPS_SNP , and (3) gene-wise combined scores which were calculated as the ratio of the SNP score to the mean methylation level of promoters in the corresponding gene. Additional models with only a genetic or methylation score were also built. All models were adjusted for age and BMI.

RESULTS

MS_SAR was negatively associated with ACSA_VL , MVC_EF , and MVC_KE and explained 10.1%, 35.5%, and 40.1% of the variance, respectively. MS_SAR explained more variance in these muscular phenotypes than GPS_SNP , MS_SNP , and models including both genetic and methylation scores. MS_SNP and GPS_SNP accounted for less than 8% and 5% of the variance in all muscular phenotypes, respectively. The genotype and methylation level of CNTF was positively related to MVC_KE (P = 0.03) and explained 12.2% of the variance. The adjusted R² and Akaike information criterion showed that models with only a MS_SAR performed the best in explaining inter-individual variance in muscular phenotypes.

CONCLUSIONS

Our results improve the understanding of inter-individual variance in muscular characteristics of older women and suggest a possible application of a sarcopenia-driven methylation score to muscle strength estimation in older women while the combination with a genetic score still needs to be further studied.

A simple, rapid, interpretable, actionable and implementable digital PCR based mortality index

Mortality assessments are conducted for both civil and commercial purposes. Recent advances in epigenetics have resulted in DNA methylation tools to assess risk and aid in this task. However, widely available array-based algorithms are not readily translatable into clinical tools and do not provide a good foundation for clinical recommendations. Further, recent work shows evidence of heritability and possible racial bias in these indices. Using a publicly available array data set, the Framingham Heart Study (FHS), we develop and test a five-locus mortality-risk algorithm using only previously validated methylation biomarkers that have been shown to be free of racial bias, and that provide specific assessments of smoking, alcohol consumption, diabetes and heart disease. We show that a model using age, sex and methylation measurements at these five loci outperforms the 513 probe Levine index and approximates the predictive power of the 1030 probe GrimAge index. We then show each of the five loci in our algorithm can be assessed using a more powerful, reference-free digital PCR approach, further demonstrating that it is readily clinically translatable. Finally, we show the loci do not reflect ethnically specific variation. We conclude that this algorithm is a simple, yet powerful tool for assessing mortality risk. We further suggest that the output from this or similarly derived algorithms using either array or digital PCR can be used to provide powerful feedback to patients, guide recommendations for additional medical assessments, and help monitor the effect of public health prevention interventions.

A Comparison of the Predictive Power of DNA Methylation with Carbohydrate Deficient Transferrin for Heavy Alcohol Consumption

Currently, the most commonly used biomarker of alcohol consumption patterns is carbohydrate-deficient transferrin (CDT). However, the CDT has limited sensitivity and requires the use of blood. Recently, we have shown that digital DNA methylation techniques can both sensitively and specifically detect heavy alcohol consumption (HAC) using DNA from blood or saliva. In order to better understand the relative performance characteristics of these two tests, we compared an Alcohol T-Score (ATS) derived from our prior study and serum CDT levels in 313 (182 controls and 131 HAC cases) subjects discordant for HAC. Overall, the Receiver Operating Characteristic (ROC) area under the curve (AUC) analyses showed that DNA methylation predicted HAC status better than CDT with AUCs of 0.96 and 0.87, respectively (p < 0.0001). The performance of the CDT was affected by gender while the ATS was not, while both were affected by age. We conclude that DNA methylation is a promising method for quantifying HAC and that further studies to better refine its strengths and limitations are in order.

Detection and Severity Assessment of Peripheral Occlusive Artery Disease via Deep Learning Analysis of Arterial Pulse Waveforms: Proof-of-Concept and Potential Challenges

Toward the ultimate goal of affordable and non-invasive screening of peripheral occlusive artery disease (PAD), the objective of this work is to investigate the potential of deep learning-based arterial pulse waveform analysis in detecting and assessing the severity of PAD. Using an established transmission line model of arterial hemodynamics, a large number of virtual patients associated with PAD of a wide range of severity and the corresponding arterial pulse waveform data were created. A deep convolutional neural network capable of detecting and assessing the severity of PAD based on the analysis of brachial and ankle arterial pulse waveforms was constructed, evaluated for efficacy, and compared with the state-of-the-art ankle-brachial index (ABI) using the virtual patients. The results suggested that deep learning may diagnose PAD more accurately and robustly than ABI. In sum, this work demonstrates the initial proof-of-concept of deep learning-based arterial pulse waveform analysis for affordable and convenient PAD screening as well as presents challenges that must be addressed for real-world clinical applications.

Nutrigenetics-personalized nutrition in obesity and cardiovascular diseases

Epidemiological data support the view that both obesity and cardiovascular diseases (CVD) account for a high proportion of total morbidity and mortality in adults throughout the world. Obesity and CVD have complex interplay mechanisms of genetic and environmental factors, including diet. Nutrition is an environmental factor and it has a predominant and recognizable role in health management and in the prevention of obesity and obesity-related diseases, including CVD. However, there is a marked variation in CVD in patients with obesity and the same dietary pattern. The different genetic polymorphisms could explain this variation, which leads to the emergence of the concept of nutrigenetics. Nutritional genomics or nutrigenetics is the science that studies and characterizes gene variants associated with differential response to specific nutrients and relating this variation to various diseases, such as CVD related to obesity. Thus, the personalized nutrition recommendations, based on the knowledge of an individual's genetic background, might improve the outcomes of a specific dietary intervention and represent a new dietary approach to improve health, reducing obesity and CVD. Given these premises, it is intuitive to suppose that the elucidation of diet and gene interactions could support more specific and effective dietary interventions in both obesity and CVD prevention through personalized nutrition based on nutrigenetics. This review aims to briefly summarize the role of the most important genes associated with obesity and CVD and to clarify the knowledge about the relation between nutrition and gene expression and the role of the main nutrition-related genes in obesity and CVD.

Machine learning and clinical epigenetics: a review of challenges for diagnosis and classification

BACKGROUND

Machine learning is a sub-field of artificial intelligence, which utilises large data sets to make predictions for future events. Although most algorithms used in machine learning were developed as far back as the 1950s, the advent of big data in combination with dramatically increased computing power has spurred renewed interest in this technology over the last two decades.

MAIN BODY

Within the medical field, machine learning is promising in the development of assistive clinical tools for detection of e.g. cancers and prediction of disease. Recent advances in deep learning technologies, a sub-discipline of machine learning that requires less user input but more data and processing power, has provided even greater promise in assisting physicians to achieve accurate diagnoses. Within the fields of genetics and its sub-field epigenetics, both prime examples of complex data, machine learning methods are on the rise, as the field of personalised medicine is aiming for treatment of the individual based on their genetic and epigenetic profiles.

CONCLUSION

We now have an ever-growing number of reported epigenetic alterations in disease, and this offers a chance to increase sensitivity and specificity of future diagnostics and therapies. Currently, there are limited studies using machine learning applied to epigenetics. They pertain to a wide variety of disease states and have used mostly supervised machine learning methods.

Epigenetic Link Between Statin Therapy and Type 2 Diabetes

OBJECTIVE

To investigate the role of epigenetics in statins' diabetogenic effect comparing DNA methylation (DNAm) between statin users and nonusers in an epigenome-wide association study in blood.

RESEARCH DESIGN AND METHODS

Five cohort studies' participants (n = 8,270) were classified as statin users when they were on statin therapy at the time of DNAm assessment with Illumina 450K or EPIC array or noncurrent users otherwise. Associations of DNAm with various outcomes like incident type 2 diabetes, plasma glucose, insulin, and insulin resistance (HOMA of insulin resistance [HOMA-IR]) as well as with gene expression were investigated.

RESULTS

Discovery (n = 6,820) and replication (n = 1,450) phases associated five DNAm sites with statin use: cg17901584 (1.12 × 10-25 [DHCR24]), cg10177197 (3.94 × 10-08 [DHCR24]), cg06500161 (2.67 × 10-23 [ABCG1]), cg27243685 (6.01 × 10-09 [ABCG1]), and cg05119988 (7.26 × 10-12 [SC4MOL]). Two sites were associated with at least one glycemic trait or type 2 diabetes. Higher cg06500161 methylation was associated with higher fasting glucose, insulin, HOMA-IR, and type 2 diabetes (odds ratio 1.34 [95% CI 1.22, 1.47]). Mediation analyses suggested that ABCG1 methylation partially mediates the effect of statins on high insulin and HOMA-IR. Gene expression analyses showed that statin exposure and ABCG1 methylation were associated with ABCG1 downregulation, suggesting epigenetic regulation of ABCG1 expression. Further, outcomes insulin and HOMA-IR were significantly associated with ABCG1 expression.

CONCLUSIONS

This study sheds light on potential mechanisms linking statins with type 2 diabetes risk, providing evidence on DNAm partially mediating statins' effects on insulin traits. Further efforts shall disentangle the molecular mechanisms through which statins may induce DNAm changes, potentially leading to ABCG1 epigenetic regulation.

The association of DNA methylation with body mass index: distinguishing between predictors and biomarkers

BACKGROUND

DNA methylation is associated with body mass index (BMI), but it is not clear if methylation scores are biomarkers for extant BMI or predictive of future BMI. Here, we explore the causal nature and predictive utility of DNA methylation measured in peripheral blood with BMI and cardiometabolic traits.

METHODS

Analyses were conducted across the life course using the ARIES cohort of mothers (n = 792) and children (n = 906), for whom DNA methylation and genetic profiles and BMI at multiple time points (3 in children at birth, in childhood and in adolescence; 2 in mothers during pregnancy and in middle age) were available. Genetic and DNA methylation scores for BMI were derived using published associations between BMI and DNA methylation and genotype. Causal relationships between methylation and BMI were assessed using Mendelian randomisation and cross-lagged models.

RESULTS

The DNA methylation scores in adult women explained 10% of extant BMI variance. However, less extant variance was explained by scores generated in the same women during pregnancy (2% BMI variance) and in older children (15-17 years; 3% BMI variance). Similarly, little extant variance was explained in younger children (at birth and at 7 years; 1% and 2%, respectively). These associations remained following adjustment for smoking exposure and education levels. The DNA methylation score was found to be a poor predictor of future BMI using linear and cross-lagged models, suggesting that DNA methylation variation does not cause later variation in BMI. However, there was some evidence to suggest that BMI is predictive of later DNA methylation. Mendelian randomisation analyses also support this direction of effect, although evidence is weak. Finally, we find that DNA methylation scores for BMI are associated with extant cardiometabolic traits independently of BMI and genetic score.

CONCLUSION

The age-specific nature of DNA methylation associations with BMI, lack of causal relationship and limited predictive ability of future BMI indicate that DNA methylation is likely influenced by BMI and might more accurately be considered a biomarker of BMI and related outcomes rather than a predictor. Future epigenome-wide association studies may benefit from further examining associations between early DNA methylation and later health outcomes.

DNA Methylation Biomarkers in Aging and Age-Related Diseases

Recent research efforts provided compelling evidence of genome-wide DNA methylation alterations in aging and age-related disease. It is currently well established that DNA methylation biomarkers can determine biological age of any tissue across the entire human lifespan, even during development. There is growing evidence suggesting epigenetic age acceleration to be strongly linked to common diseases or occurring in response to various environmental factors. DNA methylation based clocks are proposed as biomarkers of early disease risk as well as predictors of life expectancy and mortality. In this review, we will summarize key advances in epigenetic clocks and their potential application in precision health. We will also provide an overview of progresses in epigenetic biomarker discovery in Alzheimer's, type 2 diabetes, and cardiovascular disease. Furthermore, we will highlight the importance of prospective study designs to identify and confirm epigenetic biomarkers of disease.

Psychotropic drug-induced genetic-epigenetic modulation of CRTC1 gene is associated with early weight gain in a prospective study of psychiatric patients

BACKGROUND

Metabolic side effects induced by psychotropic drugs represent a major health issue in psychiatry. CREB-regulated transcription coactivator 1 (CRTC1) gene plays a major role in the regulation of energy homeostasis and epigenetic mechanisms may explain its association with obesity features previously described in psychiatric patients. This prospective study included 78 patients receiving psychotropic drugs that induce metabolic disturbances, with weight and other metabolic parameters monitored regularly. Methylation levels in 76 CRTC1 probes were assessed before and after 1 month of psychotropic treatment in blood samples.

RESULTS

Significant methylation changes were observed in three CRTC1 CpG sites (i.e., cg07015183, cg12034943, and cg 17006757) in patients with early and important weight gain (i.e., equal or higher than 5% after 1 month; FDR p value = 0.02). Multivariable models showed that methylation decrease in cg12034943 was more important in patients with early weight gain (≥ 5%) than in those who did not gain weight (p = 0.01). Further analyses combining genetic and methylation data showed that cg12034943 was significantly associated with early weight gain in patients carrying the G allele of rs4808844A>G (p = 0.03), a SNP associated with this methylation site (p = 0.03).

CONCLUSIONS

These findings give new insights on psychotropic-induced weight gain and underline the need of future larger prospective epigenetic studies to better understand the complex pathways involved in psychotropic-induced metabolic side effects.

AHRR Methylation is a Significant Predictor of Mortality Risk in Framingham Heart Study

Background.-The ability to predict mortality is useful to clinicians, policy makers and insurers. At the current time, prediction of future mortality is still an inexact process with some proposing that epigenetic assessments could play a role in improving prognostics. In past work, we and others have shown that DNA methylation status at cg05575921, a well-studied measure of smoking intensity, is also a predictor of mortality. However, the exact extent of that predictive capacity and its independence of other commonly measured mortality risk factors are unknown. Objective.-To determine the capacity of methylation to predict mortality. Method.-We analyzed the relationship of methylation at cg05575921 and cg04987734, a recently described quantitative marker of heavy alcohol consumption, to mortality in the Offspring Cohort of the Framingham Heart Study using proportional hazards survival analysis. Results.-In this group of participants (n = 2278) whose average age was 66 ± 9 years, we found that the inclusion of both cg05575921 and cg04987734 methylation to a base model consisting of age and sex only, or to a model containing 11 commonly used mortality risk factors, improved risk prediction. What is more, prediction accuracy for the base model plus methylation data was increased compared to the base model plus known predictors of mortality (CHD, COPD, or stroke). Conclusion.-Cg05575921, and to a smaller extent cg04987734, are strong predictors of mortality risk in older Americans and that incorporation of DNA methylation assessments to these and other loci may be useful to population scientists, actuaries and policymakers to better understand the relationship of environmental risk factors, such as smoking and drinking, to mortality.

Deciphering host immunity to malaria using systems immunology

A century of conceptual and technological advances in infectious disease research has changed the face of medicine. However, there remains a lack of effective interventions and a poor understanding of host immunity to the most significant and complex pathogens, including malaria. The development of successful interventions against such intractable diseases requires a comprehensive understanding of host-pathogen immune responses. A major advance of the past decade has been a paradigm switch in thinking from the contemporary reductionist (gene-by-gene or protein-by-protein) view to a more holistic (whole organism) view. Also, a recognition that host-pathogen immunity is composed of complex, dynamic interactions of cellular and molecular components and networks that cannot be represented by any individual component in isolation. Systems immunology integrates the field of immunology with omics technologies and computational sciences to comprehensively interrogate the immune response at a systems level. Herein, we describe the system immunology toolkit and report recent studies deploying systems-level approaches in the context of natural exposure to malaria or controlled human malaria infection. We contribute our perspective on the potential of systems immunity for the rational design and development of effective interventions to improve global public health.

A primer in artificial intelligence in cardiovascular medicine

Driven by recent developments in computational power, algorithms and web-based storage resources, machine learning (ML)-based artificial intelligence (AI) has quickly gained ground as the solution for many technological and societal challenges. AI education has become very popular and is oversubscribed at Dutch universities. Major investments were made in 2018 to develop and build the first AI-driven hospitals to improve patient care and reduce healthcare costs. AI has the potential to greatly enhance traditional statistical analyses in many domains and has been demonstrated to allow the discovery of 'hidden' information in highly complex datasets. As such, AI can also be of significant value in the diagnosis and treatment of cardiovascular disease, and the first applications of AI in the cardiovascular field are promising. However, many professionals in the cardiovascular field involved in patient care, education or science are unaware of the basics behind AI and the existing and expected applications in their field. In this review, we aim to introduce the broad cardiovascular community to the basics of modern ML-based AI and explain several of the commonly used algorithms. We also summarise their initial and future applications relevant to the cardiovascular field.

Data science, artificial intelligence, and machine learning: Opportunities for laboratory medicine and the value of positive regulation

Artificial intelligence (AI) and data science are rapidly developing in healthcare, as is their translation into laboratory medicine. Our review article presents an overview of the data science domain while discussing the reasons for its emergence. We also present several perspectives of its applications in clinical laboratories, along with potential ethical challenges related to AI and data science.

Preeclampsia-Associated Alteration of DNA Methylation in Fetal Endothelial Progenitor Cells

Objective

The pregnancy complication preeclampsia represents an independent risk factor for cardiovascular disease. Our previous research shows a diminished function of fetal endothelial colony-forming cells (ECFC), a proliferative subgroup of endothelial progenitor cells (EPC) in preeclampsia. The aim of this study was to further investigate whether DNA methylation of fetal EPC is affected in preeclampsia.

Methods

The genomic methylation pattern of fetal ECFC from uncomplicated and preeclamptic pregnancies was compared for 865918 CpG sites, and genes were classified into gene networks. Low and advanced cell culture passages were compared to explore whether expansion of fetal ECFC in cell culture leads to changes in global methylation status and if methylation characteristics in preeclampsia are maintained with increasing passage.

Results

A differential methylation pattern of fetal ECFC from preeclampsia compared to uncomplicated pregnancy was detected for a total of 1266 CpG sites in passage 3, and for 2362 sites in passage 5. Key features of primary networks implicated by methylation differences included cell metabolism, cell cycle and transcription and, more specifically, genes involved in cell-cell interaction and Wnt signaling. We identified an overlap between differentially regulated pathways in preeclampsia and cardiovascular system development and function. Cell culture passages 3 and 5 showed similar gene network profiles, and 1260 out of 1266 preeclampsia-associated methylation changes detected in passage 3 were confirmed in passage 5.

Conclusion

Methylation modification caused by preeclampsia is stable and detectable even in higher cell culture passages. An epigenetically modified endothelial precursor may influence both normal morphogenesis and postnatal vascular repair capacity. Further studies on epigenetic modifications in complicated pregnancies are needed to facilitate development of EPC based therapies for cardiovascular alterations.

A Direct Comparison of the Relationship of Epigenetic Aging and Epigenetic Substance Consumption Markers to Mortality in the Framingham Heart Study

A number of studies have examined the relationship of indices of epigenetic aging (EA) to key health outcomes. Unfortunately, our understanding of the relationship of EA to mortality and substance use-related health variables is unclear. In order to clarify these interpretations, we analyzed the relationship of the Levine EA index (LEA), as well as established epigenetic indices of cigarette (cg05575921) and alcohol consumption (cg04987734), to all-cause mortality in the Framingham Heart Study Offspring Cohort (n = 2256) Cox proportional hazards regression. We found that cg05575921 and cg04987734 had an independent effect relative to LEA and vice versa, with the model including all the predictors having better performance than models with either LEA or cg05575921 and cg04987734 alone. After correction for multiple comparisons, 195 and 327, respectively, of the 513 markers in the LEA index, as well as the overall index itself, were significantly associated with cg05575921 and cg04987734 methylation status. We conclude that the epigenetic indices of substance use have an independent effect over and above LEA, and are slightly stronger predictors of mortality in head-to-head comparisons. We also conclude that the majority of the strength of association conveyed by the LEA is secondary to smoking and drinking behaviors, and that efforts to promote healthy aging should continue to focus on addressing substance use.

Blood-Based Biomarkers for Predicting the Risk for Five-Year Incident Coronary Heart Disease in the Framingham Heart Study via Machine Learning

An improved approach for predicting the risk for incident coronary heart disease (CHD) could lead to substantial improvements in cardiovascular health. Previously, we have shown that genetic and epigenetic loci could predict CHD status more sensitively than conventional risk factors. Herein, we examine whether similar machine learning approaches could be used to develop a similar panel for predicting incident CHD. Training and test sets consisted of 1180 and 524 individuals, respectively. Data mining techniques were employed to mine for predictive biosignatures in the training set. An ensemble of Random Forest models consisting of four genetic and four epigenetic loci was trained on the training set and subsequently evaluated on the test set. The test sensitivity and specificity were 0.70 and 0.74, respectively. In contrast, the Framingham risk score and atherosclerotic cardiovascular disease (ASCVD) risk estimator performed with test sensitivities of 0.20 and 0.38, respectively. Notably, the integrated genetic-epigenetic model predicted risk better for both genders and very well in the three-year risk prediction window. We describe a novel DNA-based precision medicine tool capable of capturing the complex genetic and environmental relationships that contribute to the risk of CHD, and being mapped to actionable risk factors that may be leveraged to guide risk modification efforts.