Machine Intelligence for Aging Biomarker and Clock Discovery

Biological aging is shaped by changes in genetic regulation, genome integrity and epigenetic control. These signals are complex, interdependent and difficult to interpret with conventional analytical approaches. Machine intelligence offers new ways to detect patterns across large-scale genomic and epigenomic datasets, identify candidate biomarkers, and model how molecular features relate to age-associated change.This Research Topic will bring together studies that develop or apply computational methods to discover and evaluate aging biomarkers and clocks, with a particular focus on genetic, genomic and epigenomic data. We welcome studies covering, but not limited to, the following areas:• Machine-learning and artificial-intelligence approaches for biomarker discovery, including feature selection, representation learning, deep learning and network-based methods.• Integration of complementary molecular data, including bulk, single-cell or spatial datasets.• Longitudinal or cross-sectional studies using human or model-organism data, where the work advances understanding of aging biology.• Methods that reveal which molecular features drive model predictions, distinguish age-associated signals from technical or cohort effects, and support reproducible analysis of high-dimensional data.• Approaches to identify informative genetic or epigenetic signatures from noisy, sparse or high-dimensional datasets.• Models that reveal interactions among genetic variation, gene regulation, DNA methylation, chromatin state and other molecular features.• Explainable or causal approaches that distinguish biomarkers that track age from features that may illuminate biological processes.• Methods for constructing genomic or epigenomic clocks while limiting overfitting and improving interpretability.• Comparisons of computational strategies and evaluations of their performance in independent datasets.This Topic foregrounds the computational discovery process and the biological interpretation of model-derived features. It does not seek broad reviews of aging clocks, general assessments of multi-omics clock robustness, or studies focused mainly on intervention response or clinical implementation. By connecting machine-learning methods with genetics, genomics and epigenomics, the Topic aims to advance transparent, biologically informative approaches for measuring and understanding aging. Topic editors: Micah Jasey Savin, Gerardo Pepe Source: https://www.frontiersin.org/research-topics/86682/machine-intelligence-for-aging-biomarker-and-clock-discovery