Proteomic Clocks and Biological Age
Key Takeaways
- Proteomic clocks estimate age-related patterns from circulating proteins, usually measured in plasma or serum.
- Different proteomic clocks are trained for different targets, including chronological age, mortality risk, inflammatory ageing, or organ-specific ageing.
- They can predict important outcomes in cohorts, but a proteomic age score is still a model output rather than a direct readout of a person's whole-body biological age.
- Interpretation depends heavily on the assay platform, training population, and outcome used to build the clock.
Who This Is Useful For
This page is useful for readers trying to understand what proteomic clocks are actually measuring before interpreting biological age claims, blood-based ageing panels, or headlines linking protein signatures to mortality and disease risk.
What Proteomic Clocks Measure
Proteomic clocks use patterns in circulating proteins to estimate age-related biological change. Instead of relying on one molecule, they combine many proteins into a statistical model that predicts a target such as chronological age, mortality risk, or another ageing-related outcome. Because proteins sit close to physiology and disease processes, proteomic models can capture signals related to inflammation, extracellular matrix remodeling, tissue injury, endocrine signaling, and immune state. [1] [2] [3] [4]
Why Researchers Are Interested
Proteins are attractive ageing biomarkers because they are downstream of gene regulation and often connect more directly to functional state and pathology than upstream molecular layers do. Reviews of omics ageing clocks and human proteomics studies describe recurring age-related signals in inflammatory pathways, extracellular matrix biology, growth-factor signaling, and other processes already implicated in ageing and age-related disease. [1] [4]
Plasma proteome studies also suggest that protein ageing is not purely linear across adulthood. Large-cohort work has found wave-like shifts across the lifespan, which helps explain why proteomic clocks may reflect different biological programs depending on age range, cohort composition, and model design. [2]
Proteomic Clock Types at a Glance
| Clock Type | Main Target | What It Can Be Useful For | Main Limitation |
|---|---|---|---|
| Chronological-age proteomic clocks | Calendar age prediction | Summarizing how strongly a plasma protein profile resembles typical age-related patterns | High age-prediction accuracy does not automatically mean strongest prediction of health outcomes |
| Outcome-trained proteomic clocks | Mortality, disease, or multimorbidity risk | Estimating ageing-related risk in cohorts more directly than age-only models | Outputs depend on the outcome used for training and are not interchangeable with age-trained clocks |
| Domain-specific clocks | Inflammatory or pathway-focused ageing features | Tracking one aspect of age-related biology such as inflammaging | May capture only one domain rather than whole-organism ageing |
| Organ-specific proteomic clocks | Relative ageing of particular organs | Estimating heterogeneity across brain, heart, kidney, artery, and other tissues | Interpretation depends on assumptions linking blood proteins to tissue-specific ageing |
What Studies Have Found
Early large plasma-proteome studies showed that age can be predicted from circulating proteins with high accuracy and that the resulting age gap relates to functional traits. In one widely cited study, age-related protein changes were non-linear across the lifespan, and the proteomic age gap was linked with measures such as cognition and physical function. [2]
Other work showed that proteomic signatures can predict broader ageing outcomes. A plasma proteomic biomarker signature developed in the Baltimore Longitudinal Study of Aging and InCHIANTI predicted age, multimorbidity, and mortality, supporting the view that proteomic age reflects more than simple calendar time. [3]
More recent models have been built explicitly around risk prediction. In the UK Biobank and external populations, a proteomic ageing clock trained on thousands of plasma proteins was associated with multimorbidity, all-cause mortality, frailty, reaction time, telomere length, and risk across a wide range of common age-related diseases. [7] A separate mortality-trained proteomic aging clock also predicted all-cause mortality and multiple incident diseases in middle-aged and older adults, illustrating that outcome-focused designs can outperform simple age prediction for some uses. [8]
Proteomic clocks have also become more specialized. The inflammatory clock iAge focuses on systemic inflammatory ageing and has been linked with multimorbidity, frailty, immunosenescence, and cardiovascular ageing. [5] Organ-specific proteomic ageing models similarly show that different organs can appear biologically older or younger within the same person, and these organ-level age gaps are associated with organ-relevant disease risk and mortality. [6]
Why Proteomic Age Is Not the Same as a True Biological Age
The phrase "biological age" is often used loosely, but proteomic clocks do not directly measure one hidden, universal age inside the body. They estimate age-related patterns from a specific data type using a specific model. That means a proteomic age score is best understood as proteomic age, or as one indicator of biological ageing, rather than a complete measurement of whole-body ageing. [1] [9]
This matters because different proteomic clocks can disagree for valid reasons. One model may be trained to predict chronological age, another mortality, another inflammatory burden, and another organ ageing. If they differ, that may reflect different targets rather than model failure. [1] [5] [6] [7] [8] [9]
Why Interpretation Is Hard
Proteomic clocks are highly dependent on how proteins are measured. Different studies use different assay platforms, protein panels, preprocessing pipelines, and training populations, which means models do not transfer perfectly across datasets. This is one reason external validation matters so much in proteomic biomarker research. [1] [7] [8]
Blood proteins are also influenced by current physiology, disease burden, medication use, and tissue injury. That makes proteomic clocks potentially informative, but it also means they are not pure readouts of ageing alone. Their outputs usually combine ageing biology with contemporaneous health state. [1] [3] [5] [7]
Evidence Quality and Interpretation
Confidence is strong that circulating protein profiles change systematically with age and can be used to build highly predictive age-related models in large cohorts. This is now supported by multiple independent proteomic clock studies and broader reviews of omics ageing clocks. [1] [2] [3] [4]
Confidence is also strong that some proteomic age-gap measures predict mortality, multimorbidity, and major age-related disease risk at the population level. Large 2024 cohort analyses strengthened this evidence substantially. [7] [8]
Confidence is moderate that specialized proteomic clocks, such as inflammatory and organ-specific models, capture biologically meaningful subdomains of ageing rather than only statistical variation. The signal is compelling, but these models are more specific in scope and not equivalent to a whole-body measure. [5] [6] [9]
Confidence is weaker for treating any single proteomic clock as a definitive individual clinical measure of biological age. Platform differences, model dependence, and the conceptual ambiguity of "biological age" still limit that interpretation. [1] [8] [9]
What This Does Not Mean
- It does not mean one proteomic clock can fully summarize whole-body ageing. [1] [9]
- It does not mean all proteomic age scores are measuring the same construct. [5] [6] [8]
- It does not mean strong cohort-level risk prediction automatically makes a test a clinical diagnosis. [7] [8]
- It does not mean proteins included in a clock are necessarily direct causal drivers of ageing. [1] [4]
Practical Interpretation Examples
- If a proteomic clock estimates someone as older than their chronological age: that means their protein profile resembles patterns associated with older or higher-risk cohorts in that model, not that a hidden true age has been directly measured. [7] [9]
- If two proteomic clocks disagree: they may have been trained on different targets such as chronological age, mortality, inflammation, or organ-specific ageing. [5] [6] [8]
- If a clock predicts mortality or disease well in cohorts: that supports prognostic value, but not automatic clinical readiness for individual decision-making. [7] [8]
Related Reading
Summary
Proteomic clocks use blood-protein patterns to estimate age-related biological states and, in some designs, age-related risk. They are among the more promising blood-based ageing biomarkers because they connect to physiology, disease processes, and tissue-level heterogeneity. Their main limitation is not lack of signal, but interpretation: a proteomic clock is a model built on a chosen protein panel and a chosen target, so its output should be read as one lens on biological ageing rather than a complete and final measurement of it. [1] [7] [9]
References
- Rutledge, J., Oh, H. S., & Wyss-Coray, T. (2022). Measuring biological age using omics data. Nature Reviews Genetics. https://www.nature.com/articles/s41576-022-00511-7
- Lehallier, B., et al. (2019). Undulating changes in human plasma proteome profiles across the lifespan. Nature Medicine. https://www.nature.com/articles/s41591-019-0673-2
- Tanaka, T., et al. (2020). Plasma proteomic biomarker signature of age predicts health and life span. eLife. https://pmc.ncbi.nlm.nih.gov/articles/PMC7723412/
- Johnson, A. A., Shokhirev, M. N., Wyss-Coray, T., & Lehallier, B. (2020). Systematic review and analysis of human proteomics aging studies unveils a novel proteomic aging clock and identifies key processes that change with age. Ageing Research Reviews. https://pubmed.ncbi.nlm.nih.gov/32311500/
- Sayed, N., et al. (2021). An inflammatory aging clock (iAge) based on deep learning tracks multimorbidity, immunosenescence, frailty and cardiovascular aging. Nature Aging. https://www.nature.com/articles/s43587-021-00082-y
- Oh, H. S. H., et al. (2023). Organ aging signatures in the plasma proteome track health and disease. Nature. https://www.nature.com/articles/s41586-023-06802-1
- Argentieri, M. A., et al. (2024). Proteomic aging clock predicts mortality and risk of common age-related diseases in diverse populations. Nature Medicine. https://www.nature.com/articles/s41591-024-03164-7
- Kuo, C.-L., et al. (2024). Proteomic aging clock (PAC) predicts age-related outcomes in middle-aged and older adults. Aging Cell. https://pmc.ncbi.nlm.nih.gov/articles/PMC11320350/
- Johnson, A. A., & Shokhirev, M. N. (2024). Contextualizing aging clocks and properly describing biological age. Aging Cell. https://pmc.ncbi.nlm.nih.gov/articles/PMC11634725/
This content is provided for educational purposes only and does not constitute medical advice.