A research team has developed a novel biological aging clock named gtAge by integrating the immunoglobulin G (IgG) N-glycome and blood transcriptome, using a deep reinforcement learning-driven multi-omics integration method called AlphaSnake, according to a study published in Engineering. The work provides a new approach to estimate biological age by combining complementary molecular signatures that change with aging.

The study notes that both the IgG N-glycome and transcriptome serve as potential biochemical markers for chronological and biological ages, yet few attempts have been made to build an aging clock using integrated multi-omics data. To address this, researchers established gtAge based on data from 302 individuals in the Busselton Healthy Ageing Study cohort, with a mean age of (56.98 ± 5.23) years. They adopted least angle regression within a bootstrap-based feature selection framework to identify predictive features from each omics layer and then designed AlphaSnake, a deep Q network-based agent, to dynamically select the most informative features across the two omics sets during forward feature selection.

Performance evaluation using ten-fold cross-validation showed that the AlphaSnake-derived gtAge achieved a coefficient of determination R² of 0.853, outperforming the traditional concatenation-based integration method which yielded an R² of 0.820. The gtAge model explained up to 85.3% of the variance in chronological age, exceeding the performance of age predictions from the IgG N-glycome alone, referred to as gAge with an R² of 0.290, and from the transcriptome alone, referred to as tAge with an R² of 0.812. After feature refinement, the final model included 144 features comprising 137 genes and 7 glycan traits, and removing glycan features led to a notable drop in predictive performance, confirming the contribution of IgG glycosylation signatures.

Further analysis examined delta age, the difference between predicted age and chronological age, and its links to age-related phenotypes. Delta gtAge and delta tAge were both negatively associated with high-density lipoprotein, with p values of 0.02 and 0.022, respectively, while delta gAge showed positive correlations with total cholesterol, triglyceride, fasting plasma glucose, low-density lipoprotein, and glycated hemoglobin, with respective p values of 0.006, 0.002, 0.014, 0.006, and 0.039, respectively. These associations suggest that gtAge, tAge, and gAge carry complementary information about biological aging beyond chronological age.

The researchers also conducted feature importance analysis using SHAP values and pathway enrichment analysis. Genes in the tAge model were enriched in immune and inflammatory pathways such as chemokine activity and natural killer cell-mediated immunity, while the integrated gtAge model included key immune regulatory genes that support the role of immunosenescence and chronic inflammation in aging. The findings indicate that multi‑omics integration captures a more comprehensive molecular landscape of aging, and AlphaSnake offers a flexible framework for feature selection in high-dimensional heterogeneous omics data beyond aging research. The study was based on a middle-aged cohort, and future validation in larger and more diverse populations is needed to improve generalizability.

The paper “Deep Reinforcement Learning-Driven Multi-Omics Integration for Constructing gtAge: A Novel Aging Clock from the IgG N-Glycome and Blood Transcriptome,” is authored by Yao Xia, Syed Mohammed Shamsul Islam, Xingang Li, Abdul Baten, Xuerui Tan, Wei Wang. Full text of the open access paper: https://doi.org/10.1016/j.eng.2025.08.016. For more information about Engineering, visit the website at https://www.sciencedirect.com/journal/engineering.