Early evaluation of five age-correlated DNA methylation markers in an Italian population sample

Abstract

Age prediction helps investigation to narrowing down the number of potential suspects and give more reliable predicted appearance.However, since it is not known if some of the widely used methylation markers (ELOVL2, C1orf132, TRIM59, KLF14 and FHL2) are applicable as age predictors in the Italian population, in this study we collected and analyzed a set of saliva samples from 96 Italian individuals aged 19–67 years.All DNA samples were quantified, bisulfite-converted, amplified with monoplex and analyzed using PyroMark Q48 Autoprep (Qiagen).The age prediction was performed according to reference (Zbieć-Piekarska et al. [1]).Preliminary results shown a linear correlation between the methylation markers studied and age in the Italian samples, with a Pearson correlation coefficient (PCC) of 0.75 and the value of median absolute deviation (MAD) of 5.4.The MAD value observed with group 1 (age 20 s or less) was 2.8, with group 2 (30 s) was 4.5, with group 3 (40 s) was 4.7 and with group 4 (50 s or more) was 5.4.If we take the assumption that age prediction is correct when predicted age is in the range of max ±5 years of the actual age, the number of correct prediction decrease with increasing of the age group. Moreover, in contrast of what expected, our data set has shown a bias in the underestimating of predicted age at the increasing of chronological age (20% group 1, 60% group 2, 79% group 3 and 88% group 4). This experiment suggest to increase the number of different population sampled since DNA methylation could be affected by genetic variation, environmental and lifestyle variation.