Danni A. Gadd

Protein biomarkers have been identified across many age-related morbidities. However, characterising epigenetic influences could further inform disease predictions. Here, we leverage epigenome-wide data to study links between the DNA methylation (DNAm) signatures of the circulating proteome and incident diseases. Using data from four cohorts, we trained and tested epigenetic scores (EpiScores) for 953 plasma proteins, identifying 109 scores that explained between 1% and 58% of the variance in protein levels after adjusting for known protein quantitative trait loci (pQTL) genetic effects. By projecting these EpiScores into an independent sample (Generation Scotland; n = 9537) and relating them to incident morbidities over a follow-up of 14 years, we uncovered 137 EpiScore-disease associations. These associations were largely independent of immune cell proportions, common lifestyle and health factors, and biological aging. Notably, we found that our diabetes-associated EpiScores highlighted previous top biomarker associations from proteome-wide assessments of diabetes. These EpiScores for protein levels can therefore be a valuable resource for disease prediction and risk stratification.

The circulating proteome offers insights into the biological pathways that underlie disease. Here, we test relationships between 1,468 Olink protein levels and the incidence of 23 age-related diseases and mortality in the UK Biobank (n = 47,600). We report 3,209 associations between 963 protein levels and 21 incident outcomes. Next, protein-based scores (ProteinScores) are developed using penalized Cox regression. When applied to test sets, six ProteinScores improve the area under the curve estimates for the 10-year onset of incident outcomes beyond age, sex and a comprehensive set of 24 lifestyle factors, clinically relevant biomarkers and physical measures. Furthermore, the ProteinScore for type 2 diabetes outperforms a polygenic risk score and HbA1c-a clinical marker used to monitor and diagnose type 2 diabetes. The performance of scores using metabolomic and proteomic features is also compared. These data characterize early proteomic contributions to major age-related diseases, demonstrating the value of the plasma proteome for risk stratification.

Abstract Background Epigenetic clocks can track both chronological age (cAge) and biological age (bAge). The latter is typically defined by physiological biomarkers and risk of adverse health outcomes, including all-cause mortality. As cohort sample sizes increase, estimates of cAge and bAge become more precise. Here, we aim to develop accurate epigenetic predictors of cAge and bAge, whilst improving our understanding of their epigenomic architecture. Methods First, we perform large-scale ( N = 18,413) epigenome-wide association studies (EWAS) of chronological age and all-cause mortality. Next, to create a cAge predictor, we use methylation data from 24,674 participants from the Generation Scotland study, the Lothian Birth Cohorts (LBC) of 1921 and 1936, and 8 other cohorts with publicly available data. In addition, we train a predictor of time to all-cause mortality as a proxy for bAge using the Generation Scotland cohort (1214 observed deaths). For this purpose, we use epigenetic surrogates (EpiScores) for 109 plasma proteins and the 8 component parts of GrimAge, one of the current best epigenetic predictors of survival. We test this bAge predictor in four external cohorts (LBC1921, LBC1936, the Framingham Heart Study and the Women’s Health Initiative study). Results Through the inclusion of linear and non-linear age-CpG associations from the EWAS, feature pre-selection in advance of elastic net regression, and a leave-one-cohort-out (LOCO) cross-validation framework, we obtain cAge prediction with a median absolute error equal to 2.3 years. Our bAge predictor was found to slightly outperform GrimAge in terms of the strength of its association to survival (HR GrimAge = 1.47 [1.40, 1.54] with p = 1.08 × 10 −52 , and HR bAge = 1.52 [1.44, 1.59] with p = 2.20 × 10 −60 ). Finally, we introduce MethylBrowsR, an online tool to visualise epigenome-wide CpG-age associations. Conclusions The integration of multiple large datasets, EpiScores, non-linear DNAm effects, and new approaches to feature selection has facilitated improvements to the blood-based epigenetic prediction of biological and chronological age.

BACKGROUND: Metabolic differences have been reported between individuals with and without major depressive disorder (MDD), but their consistency and causal relevance have been unclear. METHODS: We conducted a metabolome-wide association study of MDD with 249 metabolomic measures available in the UK Biobank (n = 29,757). We then applied two-sample bidirectional Mendelian randomization and colocalization analysis to identify potentially causal relationships between each metabolite and MDD. RESULTS: A total of 191 metabolites tested were significantly associated with MDD (false discovery rate-corrected p < .05), which decreased to 129 after adjustment for likely confounders. Lower abundance of omega-3 fatty acid measures and a higher omega-6 to omega-3 ratio showed potentially causal effects on liability to MDD. There was no evidence of a causal effect of MDD on metabolite levels. Furthermore, genetic signals associated with docosahexaenoic acid colocalized with loci associated with MDD within the fatty acid desaturase gene cluster. Post hoc Mendelian randomization of gene-transcript abundance within the fatty acid desaturase cluster demonstrated a potentially causal association with MDD. In contrast, colocalization analysis did not suggest a single causal variant for both transcript abundance and MDD liability, but rather the likely existence of two variants in linkage disequilibrium with one another. CONCLUSIONS: Our findings suggest that decreased docosahexaenoic acid and increased omega-6 to omega-3 fatty acids ratio may be causally related to MDD. These findings provide further support for the causal involvement of fatty acids in MDD.

Microarray technology has been used to measure genome-wide DNA methylation in thousands of individuals. These studies typically test the associations between individual DNA methylation sites ("probes") and complex traits or diseases. The results can be used to generate methylation profile scores (MPS) to predict outcomes in independent data sets. Although there are many parallels between MPS and polygenic (risk) scores (PGS), there are key differences. Here, we review motivations, methods, and applications of DNA methylation-based trait prediction, with a focus on common diseases. We contrast MPS with PGS, highlighting where assumptions made in genetic modeling may not hold in epigenetic data.