An Aging Clock Integrating Epigenetic and Inflammatory Measures
Researchers here present an interesting approach to epigenetic clock development. Based on a large set of training data, the researchers used epigenetic data to predict clinical biomarkers, in this case circulating proteins measured in a blood sample that are relevant to the chronic inflammation of aging, an assessment of the inflammatory state of the immune system. Then the researchers used the predicted biomarkers of inflammation as a basis for predicting age. This approach to clock development has the advantage of producing results that are more explicable than a direct prediction of age from epigenetic data, as one can theorize more readily about the role of specific inflammatory markers than is the case for specific epigenetic changes. We will likely see more of these two-stage clocks developed in the future.
We introduce EpInflammAge, a novel deep learning framework that bridges the epigenetic and inflammatory aspects of aging. Our results demonstrate three key advances: (1) successful prediction of inflammatory markers from DNA methylation data, (2) accurate age estimation using synthetic inflammatory profiles, and (3) robust disease sensitivity across multiple pathological conditions
One of the primary objectives of this research was to integrate the two hallmarks of aging - namely, epigenetic modifications and immunosenescence. To this end, we conducted a simultaneous examination of DNA methylation data and levels of cytokines and chemokines. We developed models for estimating inflammatory marker levels from epigenetic profiles and subsequently evaluated their performance on a large cohort of healthy and diseased samples. As measuring inflammation is clinically significant, the developed model enables the acquisition of epigenetic data and the prediction of inflammatory biomarkers based on methylation. This development presents an opportunity to progress in the direction of evaluating inflammaging, which is characterized by low-grade inflammation associated with age and age-related diseases.
EpInflammAge achieves competitive performance metrics against 34 epigenetic clock models, including an overall mean absolute error of 7 years and a Pearson correlation coefficient of 0.85 in healthy controls, while demonstrating robust sensitivity across multiple disease categories. Explainable AI revealed the contribution of each feature to the age prediction. The sensitivity to multiple diseases due to combining inflammatory and epigenetic profiles is promising for both research and clinical applications. EpInflammAge is released as an easy-to-use web tool that generates the age estimates and levels of inflammatory parameters for methylation data, with the detailed report on the contribution of input variables to the model output for each sample.