Aging, a complex and multifaceted biological process, represents one of the most profound challenges in modern biomedical research. Far from being a simple accumulation of years, aging manifests as a progressive functional decline characterized by a constellation of molecular, cellular, and physiological changes that ultimately increase vulnerability to disease and death. This universal phenomenon affects all living organisms, yet exhibits remarkable variability in its progression among individuals of the same chronological age, suggesting that biological aging can be modulated by both genetic and environmental factors.
The geroscience hypothesis, a concept in aging research, posits that targeting the fundamental processes of aging could simultaneously prevent or delay the onset of multiple chronic diseases. This approach challenges traditional medical models that typically address diseases individually, instead recognizing aging itself as the primary risk factor for most chronic conditions affecting older adults. By intervening in the aging process, geroscience aims to extend not merely lifespan, but more importantly, healthspan—the period of life spent in good health, free from chronic diseases and disabilities.
Among the various molecular mechanisms implicated in aging, epigenetic modifications have emerged as particularly promising targets for intervention. Epigenetics, the study of heritable changes in gene expression that do not involve alterations to the underlying DNA sequence, provides a crucial link between our genetic blueprint and environmental influences. DNA methylation, a key epigenetic mechanism involving the addition of methyl groups to DNA molecules, has been extensively studied as a biomarker of aging and a potential driver of age-related functional decline.
The development of epigenetic clocks—mathematical algorithms that predict biological age based on DNA methylation patterns—has revolutionized our ability to quantify aging at the molecular level. These clocks have demonstrated remarkable accuracy in predicting chronological age across diverse tissues and cell types, and more importantly, their deviation from chronological age (referred to as age acceleration) has been associated with various age-related conditions and mortality risk. Recent advances in machine learning approaches have further enhanced the precision and interpretability of these epigenetic clocks, enabling more sophisticated analyses of aging trajectories.
However, despite these significant advances, current epigenetic clocks face important limitations. They often lack generalizability when applied to populations or samples that differ systematically from the training data, such as super-centenarians—individuals who have reached the extreme upper limits of human lifespan. Additionally, many existing clocks have not been fully validated for use in rare samples, interventional studies, or in-vitro experimental systems, limiting their utility in translational research.
This dissertation aims to address these critical gaps in the current understanding of aging by focusing on DNA methylation and imaging-based biomarkers. The specific aims are three-fold:
1. To solve the transportability problem of epigenetic clocks, particularly when test data are systematically different from the training data. This will involve applications to super-centenarians, whose exceptional longevity represents a unique opportunity to understand the molecular signatures of successful aging.
2. To train epigenetic clocks that exhibit expected behavior with respect to literature-validated pro- and anti-aging interventions. This aim recognizes that a truly useful aging biomarker should not only correlate with chronological age but also respond appropriately to interventions known to modulate the aging process.
3. To develop biomarkers for in vitro studies based on in-vitro optic cell imaging data and DNA methylation data. This innovative approach combines multiple data modalities to create more robust and informative aging biomarkers for cellular models.
