The team set out to solve this problem using artificial intelligence (AI) and routine lab tests like cholesterol, blood counts, and kidney function. Details of the findings were reported in the August 28 online issue of Science. Their new method combines machine learning with electronic health records to offer a more accurate, data-driven view of genetic risk.

Traditional genetic studies often rely on a simple yes/no diagnosis to classify patients. But many diseases, like high blood pressure, diabetes, or cancer, don’t fit neatly into binary categories. The Mount Sinai researchers trained AI models to quantify disease on a spectrum, offering more nuanced insight into how disease risk plays out in real life.

“We wanted to move beyond black-and-white answers that often leave patients and providers uncertain about what a genetic test result actually means,” says Ron Do, PhD, senior study author and the Charles Bronfman Professor in Personalized Medicine at the Icahn School of Medicine at Mount Sinai. “By using artificial intelligence and real-world lab data, such as cholesterol levels or blood counts that are already part of most medical records, we can now better estimate how likely disease will develop in an individual with a specific genetic variant. It’s a much more nuanced, scalable, and accessible way to support precision medicine, especially when dealing with rare or ambiguous findings.”

Using more than 1 million electronic health records, the researchers built AI models for 10 common diseases. They then applied these models to people known to have rare genetic variants, generating a score between 0 and 1 that reflects the likelihood of developing the disease.

A higher score, closer to 1, suggests a variant may be more likely to contribute to disease, while a lower score indicates minimal or no risk. The team calculated “ML penetrance” scores for more than 1,600 genetic variants.

Some of the results were surprising, say the investigators. Variants previously labeled as “uncertain” showed clear disease signals, while others thought to cause disease had little effect in real-world data.

“While our AI model is not meant to replace clinical judgment, it can potentially serve as an important guide, especially when test results are unclear. Doctors could in the future use the ML penetrance score to decide whether patients should receive earlier screenings or take preventive steps, or to avoid unnecessary worry or intervention if the variant is low-risk,” says lead study author Iain S. Forrest, MD, PhD, in the lab of Dr. Do at the Icahn School of Medicine at Mount Sinai. “If a patient has a rare variant associated with Lynch syndrome, for instance, and it scores high, that could trigger earlier cancer screening, but if the risk appears low, jumping to conclusions or overtreatment might be avoided.”

The team is now working to expand the model to include more diseases, a wider range of genetic changes, and more diverse populations. They also plan to track how well these predictions hold up over time, whether people with high-risk variants actually go on to develop disease, and whether early action can make a difference.

“Ultimately, our study points to a potential future where AI and routine clinical data work hand in hand to provide more personalized, actionable insights for patients and families navigating genetic test results,” says Dr. Do. “Our hope is that this becomes a scalable way to support better decisions, clearer communication, and more confidence in what genetic information really means.”

The paper is titled “Machine learning-based penetrance of genetic variants.”

The study’s authors, as listed in the journal, are Iain S. Forrest, Ha My T. Vy, Ghislain Rocheleau, Daniel M. Jordan, Ben O. Petrazzini, Girish N. Nadkarni, Judy H. Cho, Mythily Ganapathi, Kuan-Lin Huang, Wendy K. Chung, and Ron Do.

This work was supported in part by the following grants: National Institute of General Medical Sciences of the National Institutes of Health (NIH) (T32-GM007280); the National Institute of General Medical Sciences of the NIH (R35-GM124836); the National Institute of Diabetes and Digestive and Kidney Diseases (U24-DK062429); the National Human Genome Research Institute of the NIH (R01-HG010365); the National Institute of General Medical Sciences of the NIH (R35-GM138113); and the National Institute of Diabetes and Digestive and Kidney Diseases (U24-DK062429).

* Mount Sinai Health System member hospitals: The Mount Sinai Hospital; Mount Sinai Brooklyn; Mount Sinai Morningside; Mount Sinai Queens; Mount Sinai South Nassau; Mount Sinai West; and New York Eye and Ear Infirmary of Mount Sinai

Now, a new study in mice led by researchers at Washington University School of Medicine in St. Louis and the Baylor College of Medicine reveals a previously unknown cellular purging process that may help injured cells revert to a stem cell-like state more rapidly. The investigators dubbed this newly discovered response cathartocytosis, taking from Greek root words that mean cellular cleansing.

Published online in the journal Cell Reports, the study used a mouse model of stomach injury to provide new insights into how cells heal, or fail to heal, in response to damage, such as from an infection or inflammatory disease.

“After an injury, the cell’s job is to repair that injury. But the cell’s mature cellular machinery for doing its normal job gets in the way,” said first author Jeffrey W. Brown, MD, PhD, an assistant professor of medicine in the Division of Gastroenterology at WashU Medicine. “So, this cellular cleanse is a quick way of getting rid of that machinery so it can rapidly become a small, primitive cell capable of proliferating and repairing the injury. We identified this process in the GI tract, but we suspect it is relevant in other tissues as well.”

Brown likened the process to a “vomiting” or jettisoning of waste that essentially adds a shortcut, helping the cell declutter and focus on regrowing healthy tissues faster than it would be able to if it could only perform a gradual, controlled degradation of waste.

As with many shortcuts, this one has potential downsides: According to the investigators, cathartocytosis is fast but messy, which may help shed light on how injury responses can go wrong, especially in the setting of chronic injury. For example, ongoing cathartocytosis in response to an infection is a sign of chronic inflammation and recurring cell damage that is a breeding ground for cancer. In fact, the festering mess of ejected cellular waste that results from all that cathartocytosis may also be a way to identify or track cancer, according to the researchers.

A novel cellular process

The researchers identified cathartocytosis within an important regenerative injury response called paligenosis, which was first described in 2018 by the current study’s senior author, Jason C. Mills, MD, PhD. Now at the Baylor College of Medicine, Mills began this work while he was a faculty member in the Division of Gastroenterology at WashU Medicine and Brown was a postdoctoral researcher in his lab.

In paligenosis, injured cells shift away from their normal roles and undergo a reprogramming process to an immature state, behaving like rapidly dividing stem cells, as happens during development. Originally, the researchers assumed the decluttering of cellular machinery in preparation for this reprogramming happens entirely inside cellular compartments called lysosomes, where waste is digested in a slow and contained process.

From the start, though, the researchers noticed debris outside the cells. They initially dismissed this as unimportant, but the more external waste they saw in their early studies, the more Brown began to suspect that something deliberate was going on. He utilized a model of mouse stomach injury that triggered the reprogramming of mature cells to a stem cell state all at once, making it obvious that the “vomiting” response — now happening in all the stomach cells simultaneously — was a feature of paligenosis, not a bug. In other words, the vomiting process was not just an accidental spill here and there but a newly identified, standard way cells behaved in response to injury.

Although they discovered cathartocytosis happening during paligenosis, the researchers said cells could potentially use cathartocytosis to jettison waste in other, more worrisome situations, like giving mature cells that ability to start to act like cancer cells.

The downside to downsizing

While the newly discovered cathartocytosis process may help injured cells proceed through paligenosis and regenerate healthy tissue more rapidly, the tradeoff comes in the form of additional waste products that could fuel inflammatory states, making chronic injuries harder to resolve and correlating with increased risk of cancer development.

“In these gastric cells, paligenosis — reversion to a stem cell state for healing — is a risky process, especially now that we’ve identified the potentially inflammatory downsizing of cathartocytosis within it,” Mills said. “These cells in the stomach are long-lived, and aging cells acquire mutations. If many older mutated cells revert to stem cell states in an effort to repair an injury — and injuries also often fuel inflammation, such as during an infection — there’s an increased risk of acquiring, perpetuating and expanding harmful mutations that lead to cancer as those stem cells multiply.”

More research is needed, but the authors suspect that cathartocytosis could play a role in perpetuating injury and inflammation in Helicobacter pylori infections in the gut. H. pylori is a type of bacteria known to infect and damage the stomach, causing ulcers and increasing the risk of stomach cancer.

The findings also could point to new treatment strategies for stomach cancer and perhaps other GI cancers. Brown and WashU Medicine collaborator Koushik K. Das, MD, an associate professor of medicine, have developed an antibody that binds to parts of the cellular waste ejected during cathartocytosis, providing a way to detect when this process may be happening, especially in large quantities. In this way, cathartocytosis might be used as a marker of precancerous states that could allow for early detection and treatment.

“If we have a better understanding of this process, we could develop ways to help encourage the healing response and perhaps, in the context of chronic injury, block the damaged cells undergoing chronic cathartocytosis from contributing to cancer formation,” Brown said.

This work was supported by the National Institutes of Health (NIH), grant numbers K08DK132496, R21AI156236, P30DK052574, P30DK056338, R01DK105129, R01CA239645, F31DK136205, K99GM159354 and F31CA236506; the Department of Defense, grant number W81XWH-20-1-0630; the American Gastroenterological Association, grant numbers AGA2021-5101 and AGA2024-13-01; and a Philip and Sima Needleman Student Fellowship in Regenerative Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

An international team of scientists has now discovered that people gain more weight on an ultra-processed diet compared to a minimally processed diet, even when they eat the same number of calories. The study in humans also revealed a diet high in ultra-processed foods introduces higher levels of pollutants that are known to affect sperm quality. The findings were published in the journal Cell Metabolism.

“Our results prove that ultra-processed foods harm our reproductive and metabolic health, even if they’re not eaten in excess. This indicates that it is the processed nature of these foods that makes them harmful,” says Jessica Preston, lead author of the study, who carried out the research during her PhD at the University of Copenhagen’s NNF Center for Basic Metabolic Research (CBMR).

Same calories, different outcomes

To get the best possible data, the scientists compared the health impact of unprocessed and ultra-processed diets on the same person. They recruited 43 men aged 20 to 35, who spent three weeks on each of the two diets, with three months ‘washout’ in between. Half started on the ultra-processed and half started on the unprocessed diet. Half of the men also received a high-calorie diet with an extra 500 daily calories, while half received the normal amount of calories for their size, age and physical activity levels. They were not told which diet they were on. Both the unprocessed and ultra-processed diets had the same amount of calories, protein, carbs and fats.

Men gained around 1 kg more of fat mass while on the ultra-processed diet compared to the unprocessed diet, regardless of whether they were on the normal or excess calorie diet. Several other markers of cardiovascular health were also affected.

Ultra-processed foods polluted with endocrine disruptors

The scientists also discovered a worrying increase in the level of the hormone-disrupting phthalate cxMINP, a substance used in plastics, in men on the ultra-processed diet. Men on this diet also saw decreases in their levels of testosterone and follicle-stimulating hormone, which are crucial for sperm production.

“We were shocked by how many body functions were disrupted by ultra-processed foods, even in healthy young men. The long-term implications are alarming and highlight the need to revise nutritional guidelines to better protect against chronic disease.” says the study’s senior author Professor Romain Barrès from the University of Copenhagen’s NNF Center for Basic Metabolic Research, and the Université Côte d’Azur.