Image: Mostphotos, Dmitry Berkut, Dmitry Berkut / www.dimaberkut.com
Image: Mostphotos, Dmitry Berkut, Dmitry Berkut / www.dimaberkut.com
Research project funded by the Swedish Research Council.
Unlike chronological age, the pace of our biological age is constantly changing. The most promising biomarker of biological age is the epigenetic clock, which measures changes in DNA methylation levels. However, because we lack a fundamental understanding of the methylation mechanisms, the clock is currently of limited use. We want to develop a mathematical model to increase this understanding and make the clock a forward-looking age indicator.
There is a paradigm shift in longevity research. As cells age, they undergo changes that prevent them from functioning normally. For years, scientists have been trying to find the causes of these changes and agreed on nine independent ageing features such as a disrupted cell cycle, an increased rate of DNA mutations and short telomeres. However, a recent study shed new light on these signs of ageing. A group of researchers restored vision in old mice using epigenetic programming. This breakthrough points to epigenetic changes as the underlying drivers of ageing.
The aim of this project is to develop methods to increase the understanding of the mechanisms behind epigenetic clocks and enable predictions of biological aging. We will apply our knowledge in computational and mathematical modeling that we have developed for other epigenetic systems in the fruit fly. To make our models realistic and calibrate the model parameters, we will work with experimental data provided by Associate Professor Sofie Degerman at Umeå University.
The data describes the methylation level of more than 800 million positions on DNA in several different cell types and individuals. In addition to methylation levels from normally developed individuals, through Sofie Degerman we have access to unique data from adolescents and adults with leukemia or lymphoma as well as samples from a 25-year longitudinal project on aging and dementia. Together, we have an infrastructure where we can compare our models with aging data, make predictions, and, if necessary, perform new experiments.
The project is structured around three objectives. In sub-goal 1, we will develop a theoretical model for DNA methylation. In sub-goal 2, we will fit the model to experimental data using Bayesian inference methods. In sub-objective 3, we will use simulations to analyze data from patients showing accelerated ageing symptoms due to diseases.
The innovation of the application is to apply simulation methods and mathematical modeling developed in theoretical physics to find important causal links for epigenetic clocks and biological aging. Calibrated on unique data, our models will help determine important parameters that are difficult to measure with current experimental methods (e.g. protein activity). Computer simulations will also allow us to study how the clock reacts if key mechanisms are out of balance and predict which DNA regions change the most. Of particular interest are regions with genes linked to cancer, dementia, or autoimmune diseases. We see these goals as realistic during the grant period. But looking ahead, we see that causal models will play an important role in developing epigenetic therapies to alleviate diseases and to slow, stop or even reverse accelerated aging.
