Epigenetic-based age prediction in blood samples: Model development

Area of Science:

• Epigenetics
• Genomics
• Biogerontology

Background:

• Aging is a multifaceted process involving genetic, epigenetic, and environmental influences, leading to physiological decline.
• Epigenetic modifications, particularly DNA methylation, are crucial regulators of gene expression and are implicated in the aging process.
• Accurate age estimation is vital for various applications, including forensic science and clinical diagnostics.

Purpose of the Study:

• To develop and validate a novel epigenetic age estimation model for the Turkish population.
• To identify specific CpG sites in DNA methylation patterns correlated with chronological age.
• To assess the model's accuracy, reliability, and potential applications using blood samples.

Main Methods:

• Selection of eight CpG sites (TOM1L1, ELOVL2, ASPA, FHL2, C1orf132, CCDC102B, cg07082267, RASSF5) based on age correlation.
• Analysis of DNA methylation patterns in blood samples from 100 Turkish volunteers across different age groups.
• Development of a predictive model using Multiple Linear Regression, focusing on a 6-CpG subset for optimal performance.
• Validation through sensitivity analysis, training-test split, and fivefold cross-validation.

Main Results:

• A 6-CpG model (excluding cg07082267 and TOM1L1) demonstrated high accuracy in age prediction (r=0.941).
• The model explained 88% of age variance with low mean absolute error (MAE=4.07) and root mean square error (RMSE=5.73 years).
• Reliable performance was observed with DNA inputs as low as 1 ng, with consistent accuracy across age cohorts and genders.

Conclusions:

• The developed DNA methylation-based age estimation model is accurate and reliable for the Turkish population.
• The model shows potential for practical applications in forensic investigations and clinical assessments.
• Epigenetic markers provide a robust tool for estimating biological age and understanding aging processes.