A new computational tool developed by scientists at Sanford Burnham Prebys Medical Discovery Institute can infer changes at the ends of chromosomes using images from routine medical biopsies, according to findings published on March 16 in Cell Reports Methods.
The tool, called TLPath, is designed to predict the length of telomeres—repeating sections of DNA that cap chromosome ends—by detecting structural alterations in cells and tissues. Telomere length has been linked to aging and risk for chronic diseases, making this research significant for understanding human health over time.
Sanju Sinha, PhD, assistant professor in the Cancer Metabolism and Microenvironment Program at Sanford Burnham Prebys, said, "Whenever DNA gets replicated as our cells grow and divide, the part at the end of the DNA cannot be replicated." Sinha explained that instead of losing essential genetic information with each cell division, cells use telomeres as protective caps. "This would be a problem if our DNA was degraded bit by bit from birth, but instead our cells evolved a unique solution of capping the ends of DNA with repeating regions called telomeres that can be whittled down instead of more essential genetic information."
Researchers trained TLPath using data from the Genotype-Tissue Expression Project—a National Institutes of Health initiative launched in 2010—which provided high-resolution scans paired with laboratory tests for telomere length. The dataset included 5,263 histopathology slides from 18 tissue types donated by 919 individuals. "The dataset pairs these high-resolution images with laboratory tests of telomere length, enabling us to train TLPath to find predictive features in the cells and tissue," said Sinha.
The model segments each slide into an average of 1,387 fragments or patches and analyzes up to 1,024 structural features per patch. By assigning statistical weights to these features and comparing them with measured telomere lengths, TLPath learns how to make predictions based on image data alone. After training on specific tissue types, it successfully predicted telomere lengths on samples not included in its training set.
Sinha said recent advances in computer vision made this work possible: "The key to our work was building on recent developments in computer vision for histopathology slides, which is the creation of foundation models." He added that these models identify higher order features beyond what humans can interpret but are validated for their predictive power.
Testing showed that TLPath predicted telomere length more accurately than simply using patient age. It could also distinguish differences between individuals who were exactly the same chronological age. "This opens up new opportunities based on the conceptual advancement that measurable structural changes in cells can predict the length of telomeres," said Sinha. He noted that direct measurement requires complex and costly tests: "The only limit to using an approach such as TLPath is the availability of scanned histopathology slides."
While such slides are routinely created during clinical care, they are rarely digitized or shared widely among researchers outside projects like those funded by NIH. Sinha concluded: "Whether it is new slides being developed today or those preserved in biobanks, all we need is for them to be properly scanned, stored and shared in order to enable large-scale studies." He added: "This has the potential to transform our ability to study telomere biology, learn more about human aging and ultimately help people preserve better health as they age."
