Chronological age prediction based on DNA methylation: Massive parallel sequencing and random forest regression

Abstract

The use of DNA methylation (DNAm) to obtain additional information in forensic investigations showed to be a promising and increasing field of interest. Prediction of the chronological age based on age-dependent changes in the DNAm of specific CpG sites within the genome is one such potential application. Here we present an age-prediction tool for whole blood based on massive parallel sequencing (MPS) and a random forest machine learning algorithm. MPS allows accurate DNAm determination of pre-selected markers and neighboring CpG-sites to identify the best age-predictive markers for the age-prediction tool. 15 age-dependent markers of different loci were initially chosen based on publicly available 450K microarray data, and 13 finally selected for the age tool based on MPS (DDO, ELOVL2, F5, GRM2, HOXC4, KLF14, LDB2, MEIS1-AS3, NKIRAS2, RPA2, SAMD10, TRIM59, ZYG11A). Whole blood samples of 208 individuals were used for training of the algorithm and a further 104 individuals were used for model evaluation (age 18-69). In the case of KLF14, LDB2, SAMD10, and GRM2, neighboring CpG sites and not the initial 450K sites were chosen for the final model. Cross-validation of the training set leads to a mean absolute deviation (MAD) of 3.21 years and a root-mean square error (RMSE) of 3.97 years. Evaluation of model performance using the test set showed a comparable result (MAD 3.16 years, RMSE 3.93 years). A reduced model based on only the top 4 markers (ELOVL2, F5, KLF14, and TRIM59) resulted in a RMSE of 4.19 years and MAD of 3.24 years for the test set (cross validation training set: RMSE 4.63 years, MAD 3.64 years). The amplified region was additionally investigated for occurrence of SNPs in case of an aberrant DNAm result, which in some cases can be an indication for a deviation in DNAm. Our approach uncovered well-known DNAm age-dependent markers, as well as additional new age-dependent sites for improvement of the model, and allowed the creation of a reliable and accurate epigenetic tool for age-prediction without restriction to a linear change in DNAm with age.

Age Estimation Based on DNA Methylation Using Blood Samples From Deceased Individuals

“…In this study of Naue et al. , all the analyzed markers in blood ( DDO , ELOVL2 , F5 , GRM2 , HOXC4 , KLF14 , LDB2 , NKIRAS2 , RPA2 , SAMD10 , TRIM59 , MEIS1, and ZYG11A ) showed a comparable age dependency in comparison to a previous study using whole blood of living individuals . Interestingly, the ELOVL2 position (Chr6: 11044644; cg16867657) showed a Pearson correlation coefficient of 0.88, which is similar to the value ( r = 0.76) obtained in our data for the same position (CpG6).…”

“…The cross‐validation of the training set leads to a MAD of 7.22 years and evaluation of model performance using an independent test set of 19 individuals showed a MAD of 8.84 years. Previous forensic studies, most of them with blood samples of living individuals, using different loci and different number of markers, covering different age ranges and using different methodologies, gave values of MAD between 3.5 to 7.5 years . Thus, in our study the obtained MAD value, although less accurate, is in the range of other research approaches.…”

“… investigated 13 CpGs (located in genes DDO, ELOVL2 , F5, GRM2, HOXC4, KLF14, LDB2, MEIS1‐AS3, NKIRAS2, RPA2, SAMD10, TRIM59, and ZYG11A ) in blood samples from 29 deceased individuals by massively parallel sequencing. These markers were previously selected as strong age‐dependent loci on whole blood from living individuals . Hamano et al.…”

“…Several DNA methylation markers have been investigated in various tissues and body fluids using DNA-based methodologies such as bisulfite pyrosequencing (3)(4)(5)(6)(7)(8)(9), EpiTYPER technology (10), massively parallel sequencing (11)(12)(13), or SNaPshot assays (14)(15)(16).This allowed the identification of many CpG markers showing high correlations with chronological age, potentially useful as forensic age predictors. Thus, a number of high accurate age prediction models have been proposed for specific tissues, including blood (6,7,12), teeth (17), buccal swabs (8) or saliva (15), or as multi-tissue models (13,16,18,19).…”

Age Estimation with DNA: From Forensic DNA Fingerprinting to Forensic (Epi)Genomics: A Mini-Review

“…With the emergence of MPS in the forensic community, more recent research is going to focus on this new method [38]. Based on a selection of age-informative CpG sites using publicly available methylation data from 1,156 whole blood samples (2–90 years), Vidaki et al [31] applied stepwise regression for variable selection and identified 23 CpG sites that provided an age prediction accuracy of 0.92 (MAD ±4.6 years).…”