Machine learning techniques can predict biological age of the body
Muscle atrophy is one of the processes associated with aging and is associated with the deterioration of skeletal muscle and its function. American scientists have recently developed a new machine learning technique that can predict the biological age of muscles and help prevent muscle atrophy.
Deep learning-based models can be used to assess the importance of genetic and epigenetic factors influencing this process across multiple age groups.
“We believe that the most effective anti-aging therapies should be tissue-specific, so we focused on developing tissue-specific aging biomarkers. This study is an example of a marker for skeletal muscle tissue,” said lead researcher Polina Mamoshina of the US-based next-generation artificial intelligence company Insilico Medicine.
 |
In the study, published in Frontiers in Genetics, the team analyzed gene expression in young and old tissues from healthy donors.
Differential gene expression analysis was performed to compare the expression of both young and old muscle tissues and to preprocess the resulting data for a set of machine learning algorithms.
After using several machine learning methods, they predicted the age of the samples based on the transcriptome signature.
Finally, the trained age predictors were used to identify tissue-specific aging measures.
The combined data-driven approach demonstrated that the age prediction model could be a useful tool in identifying potential targets for longevity agents.
Age-related muscle loss remains a significant clinical challenge affecting hundreds of millions of older adults. It is associated with adverse health outcomes such as falls, impaired balance, physical disability, and death. Understanding muscle loss from aging research suggests that understanding the molecular mechanisms of muscle aging may point to potential rejuvenation approaches.
