DNA methylation-based health predictors: advances, applications, and perspectives.

Area of Science:

• Epigenetics
• Biomarkers
• Personalized Medicine

Background:

• DNA methylation (DNAm) is a dynamic epigenetic mark influenced by various factors.
• It serves as a powerful biomarker for health outcomes, biological aging, and disease risk.
• DNAm's dynamic nature makes it suitable for personalized medicine applications.

Purpose of the Study:

• To provide a comprehensive synthesis of recent advances in DNA methylation-based predictors.
• To describe the methodological frameworks and performance of these predictors.
• To highlight clinical applications, ethical considerations, and future directions.

Main Methods:

• Review of recent literature on DNAm-based predictors.
• Synthesis of diverse methodological frameworks including penalized regression, surrogate modeling, and deep learning.
• Discussion of performance across preprocessing strategies and study populations.

Main Results:

• DNAm predictors include epigenetic clocks, exposure biomarkers, disease risk models, and trait-specific estimators.
• Performance varies with preprocessing and study populations.
• Key applications and challenges such as reproducibility and interpretability are identified.

Conclusions:

• DNAm-based predictors offer significant potential for personalized medicine and health research.
• Addressing challenges in reproducibility, tissue specificity, generalizability, and interpretability is crucial.
• Future directions involve AI, multiomics integration, and longitudinal modeling for advancing DNAm predictor development and implementation.