Age Prediction based on Blood DNA Methylation

Abstract

ObjectiveIn the judicial field, traditional DNA methylation age prediction models have low accuracy and poor stability. Additionally, the use of linear regression models for detection is inefficient and costly. This study aims to utilize the prediction principles of the Support Vector Regression (SVR) model, based on preliminary laboratory data from blood DNA methylation detection using the Illumina 850K chip. By selecting low-dimensional and highly linear loci, we aim to establish a highly stable and accurate blood DNA methylation age prediction model. MethodsThis research is based on Illumina 850K chip technology. We conducted a literature review to select CpG sites and related primers, then employed SVR for model construction and age prediction. The model was built on the Matlab2022a platform. Standard parameters were selected, and optimal values for C and g were determined using grid search and cross-validation methods. During data processing, numerical values were normalized before calculation and de-normalized to obtain the predicted values. ResultsThe constructed model achieved an R2 of 0.91563 and a Mean Absolute Error (MAE) of 2.77 years. This indicates that the prediction accuracy for blood samples reached 91.56%, with an error of 2.77 years. Moreover, the accuracy of the models predictions decreases with increasing age.