AI Study Identifies 64 Genetic Drivers and Potential Therapies for Brain Aging

Mapping the Genetic Landscape of Brain Aging

Researchers defined brain aging by the “Brain Age Gap” (BAG), which measures the discrepancy between an individual’s predicted brain age and their actual chronological age. By analyzing the UK Biobank dataset of 29,097 healthy participants, the team trained seven AI models to estimate brain age. The study determined that a three-dimensional visual Transformer (3D-ViT) model outperformed other algorithms in accuracy.

Following this, scientists performed Genome-Wide Association Studies (GWAS) on 31,520 healthy participants to correlate genetic variations with BAG. The analysis revealed 64 genes associated with brain aging, which function within biological pathways such as programmed cell death, vascular wall cell surface interactions, and extracellular matrix organization. Among these, seven genes—MAPT, TNFSF12, GZMB, SIRPB1, GNLY, NMB, and C1RL—were identified as strong causal candidates for brain aging.

Drug Repurposing and Future Therapeutic Potential

By cross-referencing these 64 genes with the Drug-Gene Interaction Database (DGIdb), the study identified 466 drugs with potential anti-aging properties. Further filtering narrowed this list to 29 candidates that may help delay brain aging. Of these, 20 had been cited in previous research, and 13—including dasatinib, diclofenac, and didanosine—are already being tested in clinical trials. Neuroscientist Agustín Ibáñez of Trinity College Dublin noted that the findings could pave the way for new brain anti-aging medications. However, researchers cautioned that because the study participants were primarily of European descent, the findings require further validation across more diverse populations.

Context: Structural Mechanisms of Neurodegeneration

While the AI study focuses on genetic drivers, other recent research highlights the physical mechanisms that may accelerate brain aging and neurodegenerative conditions.

According to toutiao.com, the MPS regulates the timing and quantity of nutrient and protein intake via endocytosis. When this structure weakens—a process observed in aging—neurons accelerate the uptake of amyloid precursor protein (APP). Once inside the cell, APP is cleaved into toxic Aβ42 fragments, which are closely linked to Alzheimer’s disease. The researchers suggest that stabilizing the MPS could provide a new therapeutic target to prevent the early, hidden cellular changes that precede disease symptoms.

Metabolic Signaling and Protein Interactions

Separate research has further clarified the link between brain function and systemic health in Alzheimer’s disease. Findings indicate that β-amyloid (Aβ) interferes with the PTPRD receptor, disrupting a metabolic signaling axis that connects cognitive health with physical functions like muscle mass and endurance.

As reported by familydoctor.cn, restoring levels of the protein asprosin can reconstruct PTPRD signaling. In late-stage Alzheimer’s models, this intervention was shown to improve memory and alleviate symptoms such as weight loss, muscle atrophy, and physical weakness.

Additionally, the role of mTOR—a protein kinase involved in metabolism and neural plasticity—remains a subject of investigation. While most research focuses on cytoplasmic mTOR, studies in the Journal of Neurochemistry (Ebiotrade) note that mTOR also functions within the nucleus. Although its specific role in neuronal pathology is still being defined, researchers are currently analyzing protein interaction data to determine how nuclear mTOR might contribute to neuronal dysfunction and neurodevelopmental processes.