The emerging field of epigenetic clocks offers a revolutionary way to enhance personalized training and strategic sports management. The core question is: Could these biological age predictors be utilized to serve as biomarkers for enhancing personalized training and sports management? To tackle this, a team at VUGENE, in collaboration with the Epigenetic Clock Development Foundation and other partners, initiated a focused study on professional soccer players.
This research specifically tracks the impact of intensive physical activity on DNA methylation age across pre- and post-match intervals throughout the season. The connection between physical activity and epigenetic aging was measured by dynamic changes in DNAmGrimAge2 (a future mortality predictor) and DNAmFitAge (a physical fitness age predictor). The researchers aimed to determine if the epigenetic clocks can serve as reliable biomarkers for optimizing individual training loads, forecasting personalized recovery windows, and refining long-term athletic development strategies.
Intense physical activity causes rapid changes in biological age predictors. (A–C) Epigenetic profiles (DNAm) of saliva samples collected from athletes (n ≤ 19) during medium to low physical stress states (yellow: Before game, blue: Rested) or immediately after intensive physical activity (red: after game) were used to estimate biological/chronological age in years; The reference time point for all comparisons was After Game; (A) GrimAge2 cal. (before vs. after game β = −7.07, 95% CI: [−10.32, −3.71], p = 6.13e-05, f2 = 0.76; after game vs. rested β = 4.58, 95% CI: [1.62, 7.47], p = 0.00281, f2 = 1.50); (B) FitAge cal. (before vs. after game β = −4.76, 95% CI: [−7.10, −2.36], p = 0.00016, f2 = 1.50; after game vs. rested β = 3.88, 95% CI: [1.76, 5.96], p = 0.000476, f2 = 1.90); (C) Chronological age predictor Skin & Blood Clock (before vs. after game β = −2.99, 95% CI: [−4.53, −1.41], p = 0.000307, f2 = 1.90; after game vs. rested β = 3.88, 95% CI: [2.48, 5.25], p = 3.19e-07, f2 = 2.30); (A–C) Each dot represents predicted age in years for a specific time point and player; Significance levels are indicated by ** (p ≤ 0.01); Statistical significance in predicted age differences was evaluated using a linear mixed-effects model with chronological age and timepoints (before, after game, or rested) as fixed effects and player ID as well as batch number as a random effects; plots show median (bold line) with interquartile range (box) and 1.5-fold interquartile range (whiskers); Cal.: GrimAge2 and FitAge predictions were calibrated to the actual age range of players.
Results revealed that intensive match play immediately and significantly altered the epigenetic markers of biological age in professional soccer players. Post-game samples (n ≤ 19) showed a remarkable reduction in age predictors: DNAmGrimAge2 decreased by 32% and DNAmFitAge dropped by 18%, a rejuvenation effect that was most pronounced among midfielders, while support staff (n = 5) experienced no such changes. This temporary epigenetic shift is associated with proteins governing inflammation – specifically a decrease in CRP and an increase in IL-6 – suggesting these changes are a direct reflection of acute physiological stress induced by exercise. Crucially, the effects were short-term, as both biological age predictors observed returned to their baseline levels within 24 hours after the game.
The effects of player injuries were analyzed to assess whether epigenetic clocks could predict injury risk. Injured players showed opposite DNA methylation and CRP response patterns compared to uninjured peers: after the game, the injury group exhibited rising CRP and declining IL-6 levels based on methylation estimates. These patterns suggest a reduced fitness and slower recovery in injured athletes, whereas the decrease in DNAmGrimAge2 and DNAmFitAge in the non-injury group indicates better post-game recovery and overall health.
Bioinformatics beyond Analysis
For this analysis, we carried out the epigenetic analysis of over 850 000 CpG loci on Illumina HumanMethylationEPIC v1.0 Bead Chip data. The core task involved using the raw methylation patterns as inputs for epigenetic clock models (DNAmGrimAge2 (Lu et al., 2019), DNAmFitAge (McGreevy et al., 2023), and the Skin & Blood clock (Horvarth et al., 2018)) to estimate methylation-based biological age. We also analyzed DNA methylation-based immune cell subset predictors. All collected data were then analyzed using a Linear Mixed-Effects Model (Kuznetsova et al., 2017) to account for the repeated longitudinal measurements from the athletes. Age, sample collection group (rested, pre-, and post-match) and plate numbers served as batch effect estimators. As DNAmFitAge and DNAmGrimAge2 were trained on blood samples, to use them on saliva samples, the clocks were calibrated using a Clock Foundation trained model.
Why Does This Matter?
In athletic and physically active populations, the epigenetic clocks provide valuable insight into the physiological impact of training on aging trajectories. Combining cutting-edge molecular – epigenetic and inflammatory biomarkers into monitoring systems could enable the early identification of athletes at risk of overtraining, support personalized training and recovery plans, and optimize injury prevention strategies.
Robert T. Brooke, Thomas Kocher, Roland Zauner, Juozas Gordevičius, Milda Milčiūtė, Marc Nowakowski, Christian Haser, Thomas Blobel, Johanna Sieland, Daniel Langhoff, Winfried Banzer, Steve Horvath, Florian Pfab. Epigenetic Age Monitoring in Professional Soccer Players for Tracking Recovery and the Effects of Strenuous Exercise. Aging Cell, 2025 Oct, 24(10).
Read full article: Link
Written by: Austėja Jankevičiūtė
Cover image credits: tuiphotoengineer / Adobe Stock