Development and validation of DNA Methylation scores in two European cohorts augment 10-year risk prediction of type 2 diabetes

Abstract

Type 2 diabetes mellitus (T2D) is one of the most prevalent diseases in the world and presents a major health and economic burden, a notable proportion of which could be alleviated with improved early prediction and intervention. While standard risk factors including age, obesity, and hypertension have shown good predictive performance, we show that the use of CpG DNA methylation information leads to a significant improvement in the prediction of 10-year T2D incidence risk.Whilst previous studies have been largely constrained by linear assumptions and the use of CpGs one-at-the-time, we have adopted a more flexible approach based on a range of linear and tree-ensemble models for classification and time-to-event prediction. Using the Generation Scotland cohort (n=9,537) our best performing model (Area Under the Curve (AUC)=0.880, Precision Recall AUC (PRAUC)=0.539, McFadden’s R2=0.316) used a LASSO Cox proportional-hazards predictor and showed notable improvement in onset prediction, above and beyond standard risk factors (AUC=0.860, PRAUC=0.444 R2=0.261). Replication of the main finding was observed in an external test dataset (the German-based KORA study, p=3.7×10−4). Tree-ensemble methods provided comparable performance and future improvements to these models are discussed.Finally, we introduce MethylPipeR, an R package with accompanying user interface, for systematic and reproducible development of complex trait and incident disease predictors. While MethylPipeR was applied to incident T2D prediction with DNA methylation in our experiments, the package is designed for generalised development of predictive models and is applicable to a wide range of omics data and target traits.

An overview of DNA methylation-derived trait score methods and applications

“…Although the EPIC array includes 850K probes, 120K are reported as non-variable between individuals in the blood [40], and further QC steps that retain only variable sites useful for use in MPS have reduced the number of probes used in practice to ~ 370K [45]. Currently, many studies use a combination of Illumina EPIC and 450K arrays and so use the intersection of probes (~ 450K) and of these ~ 200K probes are retained as being sufficiently variable for use in blood-based MWAS and MPS [45,46].…”

“…Major depressive disorder [64] Smoking [65] Alzheimer's disease a [66] Incident diabetes [46] Alcohol consumption, body fat percent, body mass index, lipoprotein cholesterol, waist-to-hip ratio [15,67] 109 proteins [20] Electronic health records [68] Linear mixed model BLUP b : OSCA [52] All probes have a predicted effect size with effect sizes assumed to be drawn from a normal distribution with the total variance attributed to DNAm estimated from a restricted maximum likelihood (REML) analysis of the data.…”

“…MethylDetectR provides scripts and/or an online tool for the calculation of MPS for many traits, housing weights generated by several MPS studies [67]. MethylPipeR is a tool that facilitates the automated application of penalized regression models for MPS generation, in addition to tree-based statistical learning methods [46]. The primary issue facing those using these tree-based neural net methods is that the number of features is too large to build networks and current research is investigating if the list of probes considered can be reduced.…”

“…1). DNAm differences associated with incident diseases such as type 2 diabetes are detectable many years prior to formal diagnoses [46]. These signatures may represent the consequences of risk exposures; DNAm at smoking-associated genes have been linked to lung cancer development [95].…”

Epigenetic contributions to clinical risk prediction of cardiovascular disease

“…DNAm was profiled in blood samples using the Illumina EPIC array. Quality control details have been described previously 21 . Filtering for poorly detected probes and samples, non-blood (e.g., saliva) samples and outliers was performed.…”