Metabolomics data and AI are revolutionizing how we measure aging and predict healthy lifespan.
Study: Metabolomic Age (MileAge) Predicts Health and Lifespan: A Comparison of Multiple Machine Learning Algorithms. Image credit: Sergey Tarasov / Shutterstock
In a recent study published in the journal Science Advancesresearchers from King’s College London examined metabolomic aging clocks using machine learning models trained on plasma metabolite data from the UK Biobank. The objective of this research was to evaluate the potential of metabolomic aging clocks to predict health outcomes and lifespan, by measuring their accuracy, robustness and relevance to biological indicators of aging beyond chronological age.
Context
Biological aging, as distinct from chronological age, reflects molecular and cellular damage influencing health and susceptibility to disease. Chronological age alone cannot capture the variability in aging-related physiological states between individuals. However, recent advances in omics technologies, particularly metabolomics, have provided insights into biological aging through molecular profiling.
Metabolites, small molecules originating from metabolic pathways, help assess physiological health and are linked to aging-related outcomes, such as chronic disease and mortality. Previous studies have established correlations between metabolomics data and aging, but have been limited by small sample sizes and few markers.
Recent efforts to derive “aging clocks” from omics data using machine learning have demonstrated significant predictive power for health outcomes. However, challenges remain in optimizing these models for accuracy and interpretability, particularly with regard to metabolomics.
The current study
The study in question used nuclear magnetic resonance (NMR) spectroscopy to analyze plasma metabolite data from the UK Biobank, including 225,212 participants aged 37 to 73 years. Exclusion criteria included pregnancy, data inconsistencies, and extreme metabolite values. The dataset encompassed 168 metabolites representing lipid profiles, amino acids and glycolysis products.
The researchers applied 17 machine learning algorithms, including linear regression, tree-based models, and ensemble techniques, to develop metabolomic aging clocks. They also used a rigorous nested cross-validation method to ensure robust model evaluation.
Key preprocessing steps included handling metabolite outliers and correcting age prediction biases inherent in the models. The predictive models aimed to estimate chronological age using metabolite profiles, and the differences between predicted and actual ages were defined as the “MileAge delta”. Statistical corrections were applied extensively to remove systematic biases and improve forecast accuracy, particularly for younger and older age categories.
Models were evaluated for their predictive accuracy using criteria such as mean absolute error (MAE), root mean square error (RMSE), and correlation coefficients. For example, the Cubist regression model achieved an MAE of 5.31 years, outperforming other models like multivariate adaptive regression splines (MAE = 6.36 years). Additional analysis adjusted the predictions to remove systematic biases and better align them with chronological age.
Results
The results showed that metabolomic aging clocks developed from plasma metabolite profiles could effectively differentiate biological from chronological aging. Among the different models tested in the study, the Cubist rule-based regression model provided the strongest predictive associations with health and mortality markers, outperforming other algorithms in terms of accuracy and robustness.
Additionally, positive MileAge delta values, indicating accelerated aging, were linked to frailty, shorter telomeres, increased morbidity, and increased mortality risk. Specifically, a 1-year increase in MileAge delta corresponded to a 4% increase in all-cause mortality risk, with hazard ratios (HRs) exceeding 1.5 in extreme cases.
Additionally, the study showed that individuals with accelerated aging were more likely to report poorer self-rated health and suffer from chronic illnesses. Associations with telomere fragility and attrition were particularly pronounced, with some differences equivalent to an 18-year gap in fragility index scores. Interestingly, women had slightly higher MileAge deltas than men in most models.
The study also confirmed the non-linear nature of the relationships between metabolites and age, while highlighting the usefulness of statistical corrections to improve prediction accuracy. Furthermore, comparison with existing aging markers showed that metabolomic aging clocks captured health-relevant signals and often outperformed simpler predictors. However, the results highlighted that slowed aging (negative MileAge deltas) did not consistently translate into better health outcomes, highlighting the complexity of biological aging metrics.
Conclusions
Overall, the study demonstrated the utility of metabolomic aging clocks in predicting biological aging and associated health outcomes. By benchmarking multiple machine learning algorithms, the results also showed the superior performance of the Cubist rule-based model in linking metabolite-derived ages to health and mortality markers.
Results suggest that metabolomic aging clocks have the potential to provide support for proactive health management and risk stratification, while highlighting the need for further validations across diverse populations and longitudinal data for application. broader clinic. This study sets a new benchmark for algorithm development, illustrating how metabolic profiles can provide actionable insights into aging and health.
Takeaways
- Metabolomic aging clocks can predict biological aging with increased accuracy.
- The findings highlight associations between accelerated aging and increased health risks.
- The regression model based on the Cubist rule showed the best performance among those tested.
- Future research should expand the validation of metabolomic aging clocks in diverse clinical settings.
This study opens new perspectives on understanding the aging process. While metabolomic clocks show promise for their ability to predict health and longevity, it also raises questions about how these findings might influence clinical practices and how we approach aging in our societies.
