Male-specific age estimation based on Y-chromosomal DNA methylation

Abstract

Although DNA methylation variation of autosomal CpGs provides robust age predictive biomarkers, no male-specific age predictor exists based on Y-CpGs yet. Since sex chromosomes play an important role in aging, a Y-chromosome-based age predictor would allow studying male-specific aging effects and would also be useful in forensics. Here, we used blood-based DNA methylation microarray data of 1,057 males from six cohorts aged 15-87 and identified 75 Y-CpGs with an interquartile range of ≥0.1. Of these, 22 and six were significantly hyper- and hypomethylated with age (p(cor)<0.05, Bonferroni), respectively. Amongst several machine learning algorithms, a model based on support vector machines with radial kernel performed best in male-specific age prediction. We achieved a mean absolute deviation (MAD) between true and predicted age of 7.54 years (cor=0.81, validation) when using all 75 Y-CpGs, and a MAD of 8.46 years (cor=0.73, validation) based on the most predictive 19 Y-CpGs. The accuracies of both age predictors did not worsen with increased age, in contrast to autosomal CpG-based age predictors that are known to predict age with reduced accuracy in the elderly. Overall, we introduce the first-of-its-kind male-specific epigenetic age predictor for future applications in aging research and forensics.

Keywords: DNA methylation; Y-chromosome; epigenetic age prediction; epigenetics; machine learning.

Figure 1

Examples of age-associated Y-CpG methylation in blood. (A) Histogram showing the age distribution in all samples colour-coded per training (n = 758), internal validation (n = 172) and external testing (n = 127) datasets, (B) DNA methylation levels of cg13308744 showing the strongest negative correlation with age (ρ = -0.3197, p-value = 1.545E-26), (C) DNA methylation levels of cg04691144 showing the strongest positive correlation with age (ρ = 0.3192, p- = 1.820E-26). ρ: Spearman correlation coefficient, Bonferroni threshold: α/n= 0.05/75 = 6.667E-4, Loess: locally estimating scatterplot smoothing curve.

Figure 2

IGV screenshot on the Y Chromosome including the location of all reference Y-genes, the 416 Y-CpGs included in the Illumina® Human Methylation450 BeadChip array and the 75 age-predictive Y-CpGs used in this study (with highlighted the 19 Y-CpGs that were further selected) as well as their Spearman correlation coefficients.

Figure 3

Male-specific epigenetic age prediction in blood based on 75 Y-CpGs using support vector machine (radial kernel). Validation dataset (n = 172): (A) Predicted vs. true age and (B) age prediction errors per age category; Testing dataset (n = 127): (C) Predicted vs. true age and (D) age prediction errors per age category. ρ: Spearman correlation coefficient, RMSE: root mean square error, MAD: mean absolute deviation.

Figure 4

Age prediction of male samples included in the testing set of this study (n = 127) using the publically available Horvath age predictor based on 353 autosomal CpGs [1]. (A) Predicted vs. true age and (B) age prediction errors per age category. ρ: Spearman correlation coefficient, RMSE: root mean square error, MAD: mean absolute deviation.