A new meeting report published in Aging (Volume 18) on May 14, 2026, introduces the emerging field of Gerophysics, which seeks to explain aging by combining biology with the quantitative principles of physics. Titled “Foundations of Gerophysics,” the report was led by first author Maximilian Unfried and corresponding authors Maximilian Unfried and Brian K. Kennedy from the National University of Singapore. While aging has traditionally been studied through biology, genetics, and medicine, many fundamental questions remain unresolved, including why organisms age at different rates, why resilience declines over time, and whether aging trajectories can be predicted before disease develops. Researchers argue that addressing these questions may require a new interdisciplinary framework grounded in physical laws.
The report summarises the inaugural Global Conference on Gerophysics, held in Singapore on March 5–6, 2025, which brought together 160 researchers from physics, biology, computational science, and medicine. Across 31 presentations, participants explored how concepts such as dynamical systems, thermodynamics, network theory, stochastic processes, and artificial intelligence could transform aging research into a predictive science. A central theme was the search for simple mathematical principles capable of explaining the complex biological processes underlying aging. For example, Uri Alon presented the “saturated removal” model, which explains hallmark aging patterns through the balance between damage production and repair. At the same time, Yifan Yang demonstrated how the same framework could distinguish interventions that extend lifespan from those that specifically reduce the years spent in poor health.
Several presentations examined whether aging behaves like a physical phase transition. Peter Fedichev and Jan Gruber proposed that aging emerges from growing instability within gene regulatory networks, linking resilience, entropy, and mortality patterns across species. Other researchers showed how network instability, loss of robustness, and critical transitions may help explain the progression from healthy aging to frailty and disease. Advances in measuring biological age were also highlighted. Andrew Teschendorff discussed improvements in epigenetic clocks, while Steffen Rulands presented evidence that age-related DNA methylation changes may reflect collective behaviours across the genome, suggesting that aging can be understood as a multi-scale physical process extending from molecular interactions to whole organisms.
Artificial intelligence also featured prominently throughout the meeting. Matt Kaeberlein described the “Million Molecule Challenge,” an ambitious initiative combining automated lifespan experiments with machine learning to screen more than one million compounds for longevity-promoting effects. Andrei Tarkhov presented AI-guided protein design strategies that substantially improve cellular reprogramming factors used in age-reversal research. Beyond AI, researchers emphasised systems-level approaches, including the use of network science to model cascading biological failures, entropy-based measures as potential biomarkers of aging, and computational analyses of large clinical and molecular datasets to identify promising geroprotective interventions. Additional presentations explored reproductive aging, skeletal muscle aging, comparative longevity across species, and tissue-specific metabolic changes linked to aging.
Research on metabolism and longevity further illustrated the breadth of the emerging field. Peter James Mullen presented multi-organ metabolomic analyses revealing tissue-specific metabolic signatures associated with aging. At the same time, Maximilian Unfried described comparative lipidomics studies showing that longer-lived species possess more resilient lipid interaction networks. Brian K. Kennedy discussed the challenges of translating aging biomarkers into reliable clinical tools capable of assessing biological age and evaluating interventions. Throughout the conference, participants repeatedly stressed that meaningful progress will require close collaboration between theoretical modelling and experimental research, with mathematical models generating testable predictions that are continually refined by new biological data.
The conference concluded with broad agreement on four priorities for advancing Gerophysics: developing shared multi-modal datasets, establishing physics-based definitions of aging and rejuvenation, building predictive models that forecast responses to interventions, and strengthening the translation of findings from animal models to humans. Participants also emphasised the need for interdisciplinary training to bridge the languages of biology, physics, and computational science. By integrating quantitative physical principles with modern biological research and artificial intelligence, Gerophysics aims to shift aging research from a largely descriptive discipline towards a predictive science capable of accelerating the development of interventions that promote healthier aging throughout life.
More information: Maximilian Unfried et al, Foundations of Gerophysics, Aging-US. DOI: 10.18632/aging.206378
Journal information: Aging-US Provided by Impact Journals LLC
