Media release
From:
How old are you really? Researchers find a new way to measure biological age
Researchers from Edith Cowan University (ECU) have developed an innovative new way to measure biological age, which could make it easier to detect and track age-related conditions.
A team from ECU, together with researchers from Royal Prince Alfred Hospital in Sydney and Shantou University Medical College in China, has studied elements in the blood that change with age, specifically the IgG N-glycome, which refers to sugar structure attached to antibodies, as well as a snapshot of gene activity within blood cells, called transcriptome.
By combining these two sets of data using an artificial intelligence (AI) technique called Deep Reinforcement Learning, the researchers created a new ageing clock called gtAge.
The gtAge method predicted a person’s age with 85 per cent accuracy - more precise than using just the IgG N-glycome or just the transcriptome alone.
They also found the difference between predicted age and actual age - called delta age – was linked to health markers related to ageing, such as cholesterol and blood sugar levels.
Age – is it just a number?
Co-author and Postdoctoral Research Fellow in ECU’s School of Medical and Health Sciences, Dr Xingang Li, explained although chronological age - the time elapsed since birth - is the most direct and commonly used metric, it does not entirely capture individual variability in the aging process.
“In reality, some individuals remain healthy until into their 80s and 90s, whereas others may experience age-related decline much earlier,” Dr Li said.
“This discrepancy can be attributed to differences in biological age, which integrates genetic, lifestyle, nutritional, disease-related, and general health factors to accurately reflect the true biological aging process.”
Dr Li noted gtAge explains 85.3 per cent of the variation in chronological age.
“By merging IgG N-glycome data and transcriptome data, we have elevated the accuracy of biological ageing estimation,” he said. “It links to real health risks and could help spot people at risk of age-related diseases earlier.”
Crunching the data
In an important example of cross-disciplinary work, co-author Dr Syed Islam, ECU Senior Lecturer of Computer Science, led the AI side of the study.
“To improve age prediction using integrated multiomics data, we developed a custom AI tool named AlphaSnake, powered by Deep Reinforcement Learning,” Dr Islam explained.
“This algorithm works by picking the most useful data points from the two different biological sources, avoiding the pitfalls of just blindly blending data.”
Where to from here?
The study involved testing gtAge on 302 middle-aged adults from the Busselton Healthy Ageing Study in Western Australia.
With Australia’s ageing population, the research team believes gtAge could serve as a valuable medical tool.
“By measuring biological age and not just looking at someone’s birthdate, it could be very useful to better understand their health,” Dr Islam said.
“If we know in advance, then we can change our lifestyle to better act on preserving our health and help prevent some of the damages our body may have experienced.”
Dr Yao Xia, Dr Syed Islam, Dr Xingang Li, Dr Abdul Baten, Dr Xuerui Tan and Professor Wei Wang were co-authors of the study, Deep Reinforcement Learning–Driven Multi-Omics Integration for Constructing gtAge: A Novel Aging Clock from IgG N-glycome and Blood Transcriptome, published in Engineering.