Deep learning uncovers gene targets and potential drugs to slow brain aging microbiologystudy

In a new study reported in Science Advances, scientists analyzed MRI data stored at the UK Biobank and identified seven genes responsible for fast biological brain aging and 13 existing drugs that can target those genes.

Slowing the aging process is a powerful strategy to prevent many diseases and enhance longevity. Previous research has suggested that the ability to delay aging by even 2% could result in $7.1 trillion in health care savings in less than half a century. Over the years, it has become evident that the brain aging pattern significantly impacts overall human aging, as it is responsible for the rise in the risk of neurodegeneration and decline in both physical and cognitive health.

A crucial parameter in brain health research is the brain age gap (BAG), which is the difference between a person’s estimated biological brain age and their chronological age. In other words, the BAG measures how old a person’s brain appears on MRI or other brain age measurement techniques compared to their actual age.

The brain age gap is also a reliable biomarker (or proxy) for studying brain health. A larger BAG is often seen in individuals with brain disorders like Alzheimer’s, demyelination, and schizophrenia and is also linked to lower cognitive test scores.

While the effects of the BAG are well explored, identifying the factors driving the brain’s aging process remains a challenge. Genes are known to play a crucial role in shaping how the brain ages.

In this study, researchers used deep learning models trained on MRI scans, lifestyle data, health records, and genetic information from nearly 39,000 UK Biobank participants—averaging 64 years old, with an equal gender distribution—to pinpoint specific genes that contribute to a widening BAG.

Their findings uncovered that seven genes (MAPT, TNFSF12, GZMB, SIRPB1, GNLY, NMB, and C1RL) were promising targets for brain aging.

The 3D-ViT model accurately predicted the biological brain age of the participants by analyzing key signatures in their MRI scans. Researchers used saliency map analysis, a technique that highlights the most influential areas in an image or dataset, to identify the brain regions critical for brain age estimation.

The findings pointed to the lentiform nucleus, a region of the brain responsible for cognition such as attention and working memory, and the posterior limb of the internal capsule, which connects many parts of the brain to the cerebral cortex—the brain’s outer layer that controls for thinking, memory, and learning.

By combining insights on specific gene targets, brain regions linked to aging, and existing clinical trial data, the researchers identified 13 drugs and supplements, including hydrocortisone, testosterone, diclofenac, and metformin, that can be repurposed to slow down brain aging.

The researchers noted that the genetic basis of aging uncovered in this study can facilitate the development of new drugs to slow brain aging and improve overall health. However, the results of this study were obtained from a population of a specific region. More research needs to be conducted across diverse populations to evaluate the true extent of these findings.

© 2025 Science X Network