Chronological age estimation using CpG methylation signatures in Pakistani population

Area of Science:

• Forensic Science
• Epigenetics
• Genomics

Background:

• DNA methylation (DNAm) patterns offer a promising avenue for forensic age estimation.
• Epigenetic aging studies have historically underrepresented the Pakistani population.
• Accurate age assessment is crucial in forensic investigations.

Purpose of the Study:

• To develop and validate an epigenetic age-prediction model using DNA methylation markers.
• To assess the reliability of age estimation specifically within the Pakistani population.
• To identify robust age-associated CpG sites for forensic applications.

Main Methods:

• Analysis of seven age-associated CpG sites (KLF14, CCDC102B, TRIM59, ASPA, C1ORF132, FHL2, ELOVL2) in 181 Pakistani individuals (1-76 years).
• Utilized a methylation SNaPshot™ multiplex assay.
• Developed and compared three age prediction models: stepwise regression, multivariate linear regression (MVLR), and support vector machine (SVM).

Main Results:

• The SVM model demonstrated the highest prediction accuracy with a Mean Absolute Deviation (MAD) of 3.40 in the training set and 3.44 in the validation set.
• Correct age predictions within ±4 years were 77.35% for SVM, 71.69% for MVLR, and 69.81% for stepwise regression.
• ELOVL2 and FHL2 were identified as robust and reliable CpG sites for age estimation in the Pakistani population, while CCDC102B showed the lowest association.

Conclusions:

• The developed epigenetic age-prediction models, particularly the SVM model, show promise for forensic age estimation in the Pakistani population.
• Age prediction accuracy decreases with increasing age, suggesting the influence of environmental and lifestyle factors (e.g., air pollution) in Pakistan.
• Further optimization using diverse population samples and forensic materials is necessary to enhance the model's real-world applicability.