Using machine learning to map brain aging at the cellular level

Using machine learning to map brain aging at the cellular level

In a wide-ranging genomic press interview, Stanford University researcher Eric Sun shows how machine learning is revolutionizing our understanding of brain aging at unprecedented cellular resolution. Dr. Sun, who will establish his independent laboratory at MIT's Department of Biological Engineering and the Ragon Institute in 2026, represents a new generation of computer scientists who are transforming the aging of research through innovative machine learning.

Breakthrough discovery in cellular aging mechanisms

Dr. Sun's groundbreaking work focuses on developing "spatial aging clocks" - sophisticated machine learning models that can measure biological age at the individual cellular level. This represents a quantum leap from traditional aging research, which typically examines tissues or organs as entire units. His recent Nature Publication (2025) shows how these computational tools can identify specific cell types that dramatically influence the aging trajectory of their cellular neighbors, acting in either pro-aging or pro-rejuvenating directions.

“I have always been fascinated by the biology of aging,” explains Dr. Sun in an interview. "Why do we get wrinkles as we get older? Why does it become harder to learn and easier to forget? Why is it that some animals live significantly longer than others, but seemingly all animals experience aging?" These fundamental questions sparked his early interest in aging research, which crystallized after the discovery of Cynthia Kenyon's work on dramatic lifespan extension in C. elegans during his elementary school years.

Revolutionary computational framework for aging research

The Stanford researcher's approach represents a fundamental shift in the way scientists study aging. Traditional methods often provide broad snapshots of aging processes, but Dr. Sun's spatial aging clocks can pinpoint which cells age faster or slower in complex tissue environments. This detailed understanding opens up new opportunities for targeted interventions. Finally, could researchers identify and modify the specific cellular “bad actors” that accelerate aging in brain tissue? Could it be possible to enhance the activity of cells that promote youthful function in their neighbors?

The research methodology of Dr. Combining spatial transcriptomics with single-cell analysis, Sun produces detailed maps of how aging progresses through brain tissue. His machine learning models don't simply identify aged cells - they reveal the complex intercellular communication networks that determine whether neighboring cells age rapidly or maintain youthful characteristics.

From mathematical fundamentals to biological discovery

The path to this breakthrough is reflected by Dr. Sun's unique interdisciplinary background. Growing up in Pueblo, Colorado, he spent countless hours in the public library and became fascinated with dinosaurs and space exploration before focusing on mathematics. “Mathematics was my favorite subject through high school,” he notes, “and while it may not have directly sparked my passion for science, my early love of mathematics shaped the research areas and approaches I was drawn to.”

This mathematical foundation proved crucial when Dr. Sun began developing computational models during his undergraduate years at Harvard, where he applied chemistry, physics, and mathematics. His projects ranged from simulating chromosome development to building mathematical models of aging and using machine learning to predict age from multi-AMICS data. These experiences established the computational literacy that would later enable his revolutionary development of spatial aging.

Implications for dementia and neurodegeneration research

The practical applications of Dr. Sun's work goes far beyond basic research. Its computational framework could transform how researchers approach age-related diseases, particularly dementia and other neurodegenerative diseases. By identifying the specific cellular mechanisms that drive brain aging, scientists can develop more precise therapeutic targets. What if treatments could be designed so that the rejuvenating signals from beneficial cells simultaneously suppress the pro-aging influences of problematic cellular populations?

Dr. Sun's research also raises fascinating questions about the nature of aging itself. If individual cells can influence the aging trajectories of their neighbors, how might environmental factors or therapeutic interventions exploit these cellular communication networks? Could understanding these mechanisms lead to treatments that not only slow aging, but actually reverse it in specific brain regions?

Building the next generation of aging researchers

In addition to his research contributions, Dr. Sun the importance of mentoring future scientists. “Outside of my research, I am excited to establish my own laboratory and mentor students and postdocs,” he explains. “I want to support and cultivate the next generation of scientists, both in the area of ​​aging research and beyond.”

His commitment to scientific mentorship reflects broader concerns about supporting young researchers through the inevitable challenges of scientific discovery. Dr. Sun notes that the scientific community often emphasizes success over failure, even though failure is "extraordinarily more common than the former, and often a series of failures is the catalyst for eventual research creation or success."

Future Directions in Computer Aging

Looking to the future, Dr. His lab will focus on building large-scale AI models to predict the effects of biological perturbations at multiple scales, potentially enabling high-throughput computational screens for rejuvenation efforts.

The researcher's long-term vision includes translating computational discoveries into effective therapeutics. His work suggests a future in which aging research goes beyond describing what happens during aging to controlling exactly how it occurs. Could its spatial aging clocks eventually guide personalized anti-aging treatments tailored to an individual's specific cellular aging patterns?

The research of Dr. Sun also highlights the evolving relationship between artificial intelligence and biological discovery. His spatial aging clocks show how machine learning can not only analyze complex biological data, but also generate completely new insights into fundamental life processes. What other biological mysteries might yield similar AI-driven approaches as computing power continues to advance?

The genomic press interview by Dr. Eric Sun is part of a larger series called Innovators & Ideas, which highlights the people behind today's most influential scientific breakthroughs. Each interview in the series features a mix of cutting-edge research and personal reflections, giving readers a comprehensive look at the scientists shaping the future. By combining a focus on professional achievements with personal insights, this interview style invites a richer narrative that both engages and enlightens readers. This format provides an ideal starting point for profiles that explore the scientist's impact on the field while touching on broader human themes. For more information about the research leaders and rising stars in our Innovators and Ideas - Genomic Press Interview Series, visit our publications website:

Sources: