Decoding disease-specific ageing mechanisms through pathway-level epigenetic clock: insights from multi-cohort...

Area of Science:

• Epigenetics and Aging Research
• Computational Biology and Bioinformatics
• Genomics and Disease Association Studies

Background:

• Aging is a complex process linked to chronic diseases.
• Existing epigenetic clocks often use isolated CpG sites, limiting biological insight.
• PathwayAge was developed to provide a biologically interpretable model of aging at the pathway level.

Purpose of the Study:

• To develop and validate PathwayAge, a novel epigenetic clock model.
• To assess the biological interpretability of pathway-level methylation changes in aging.
• To investigate the association between epigenetic age acceleration and various chronic diseases.

Main Methods:

• Utilized genome-wide DNA methylation and transcriptomic data from over 10,000 individuals across multiple cohorts.
• Employed a two-stage machine learning approach to aggregate CpG sites into Gene Ontology (GO) or KEGG pathways for age prediction.
• Calculated age acceleration residuals (AgeAcc) and tested associations with nine diseases.

Main Results:

• PathwayAge demonstrated high accuracy in age prediction across diverse cohorts (Rho up to 0.979).
• The model outperformed established epigenetic clocks in age estimation and disease association analyses.
• Significant age acceleration was linked to nine diseases, with specific pathways identified via permutation tests and cross-omics validation.

Conclusions:

• PathwayAge offers an interpretable, biologically grounded framework for epigenetic age estimation.
• The model reveals mechanistic links between aging pathways and disease development.
• PathwayAge holds promise for biomarker development and advancing precision aging medicine.