A “Natural Ozempic” Discovered by AI?

Researchers at Stanford Medicine used artificial intelligence to identify BRP (BRINP2-related peptide), a naturally occurring molecule that reduced food intake and body weight in mice and pigs without causing nausea or muscle loss, potentially offering a new class of obesity treatments. The AI tool Peptide Predictor analyzed 20,000 human genes to find the 12-amino-acid peptide, which acts through brain pathways different from GLP-1 drugs like Ozempic, though BRP has not yet been tested in humans.

Appetite /us/basics/appetite A “Natural Ozempic” Discovered by AI? Newly identified peptide may have fewer side effects than current GLP-1 drugs. Posted July 24, 2026 Reviewed by Monica Vilhauer Ph.D. /us/docs/editorial-process Key points - AI helped identify BRP, a naturally occurring peptide that may reduce appetite. - BRP reduced food intake and body fat in animals without clear nausea or muscle loss. - BRP appears to act through brain pathways different from those targeted by GLP-1 drugs. - Despite promising animal results, BRP has not yet been tested for safety or efficacy in humans. The success of drugs such as Ozempic and Wegovy has transformed how scientists think about obesity. These medications demonstrate that appetite https://www.psychologytoday.com/us/basics/appetite and body weight are not simply products of willpower https://www.psychologytoday.com/us/basics/self-control . They are regulated by biological signals connecting the brain, digestive system, pancreas, and other organs. But the same broad biological reach that makes these drugs effective can also produce unwanted effects, including nausea, constipation, slowed digestion, and loss of lean body mass. Now, researchers at Stanford Medicine have identified a naturally occurring molecule that may suppress appetite through a different, and potentially more targeted, pathway. The molecule, called BRP short for BRINP2-related peptide , reduced food intake and body weight in mice and pigs without producing several of the adverse effects associated with existing weight-loss medications. Headlines have understandably referred to BRP as a “natural Ozempic.” That comparison captures the excitement surrounding the discovery, but it also risks getting ahead of the science. BRP has not yet been shown to be safe or effective in humans. What may be most important about the research is not that scientists have found an immediate replacement for semaglutide. It is that artificial intelligence https://www.psychologytoday.com/us/basics/artificial-intelligence AI helped uncover a previously hidden biological signal that may lead to an entirely new class of obesity treatments. Searching the Body’s Hidden Pharmacy Our bodies manufacture thousands of proteins, many of which begin as larger, inactive precursors called prohormones. Enzymes cut these prohormones into smaller fragments known as peptides. Some peptides become hormones https://www.psychologytoday.com/us/basics/hormones that regulate appetite, metabolism, stress https://www.psychologytoday.com/us/basics/stress , growth, and other essential functions. The problem is that biologically meaningful peptides are difficult to find. They can be buried among thousands of fragments produced during ordinary protein processing. Instead of attempting to isolate every possible peptide in the laboratory, the Stanford team developed a computational tool called Peptide Predictor. The algorithm examined approximately 20,000 human protein-coding genes for locations where enzymes might cut prohormones into biologically active peptides. The researchers focused on cleavage patterns associated with prohormone convertase 1/3, an enzyme involved in producing several hormones, including glucagon-like peptide-1, or GLP-1. Semaglutide works by mimicking GLP-1’s effects. The algorithm narrowed the search to 373 possible prohormones and predicted that they could generate 2,683 peptides. Researchers selected 100 candidates that appeared likely to act in the brain and tested them in neuron-like cells. One peptide stood out: BRP. The molecule was only 12 amino acids long, yet it increased neuronal activity approximately tenfold, compared with untreated cells. The discovery illustrates one of AI’s most promising roles in medicine. AI did not independently invent a weight-loss drug or prove that the molecule works. It helped researchers navigate a vast biological search space and identify candidates that might otherwise have remained overlooked. What BRP Did in Animals The researchers next administered BRP to mice and minipigs. Pigs were included because aspects of their metabolism and eating behavior resemble those of humans more closely than mouse models alone. A BRP injection given before feeding reduced food consumption during the following hour by as much as 50 percent in both species. Obese mice receiving daily injections for 14 days lost an average of about three grams, almost entirely from body fat, while untreated mice gained approximately three grams. The treated mice also showed improved glucose and insulin tolerance. Equally intriguing was what the researchers did not observe. The animals did not display clear signs of nausea, food aversion, constipation, reduced water consumption, abnormal movement, or anxiety https://www.psychologytoday.com/us/basics/anxiety -like behavior. The weight reduction also did not appear to involve substantial muscle loss. These findings raise the possibility that BRP could influence appetite more selectively than current GLP-1 medications. GLP-1 receptors are distributed throughout the brain, digestive tract, pancreas, and other tissues. That widespread distribution helps explain semaglutide’s effects on appetite, blood glucose, and digestion. BRP appears to activate different neuronal and metabolic pathways and may act more specifically within the hypothalamus, a brain region central to hunger and energy regulation. A treatment that reduces appetite without broadly affecting the digestive system would represent an important advance. But that possibility remains hypothetical. Why “Natural” Does Not Mean Proven—or Safer Describing BRP as natural may make it sound inherently gentle or safe. But natural molecules can have powerful biological effects, and a substance naturally produced by the body can behave differently when administered at pharmacological doses. Researchers have not yet identified the receptor through which BRP acts. They also do not know whether humans will respond as mice and pigs did, whether repeated treatment could produce unforeseen effects, or whether appetite suppression would be sustained over time. Another challenge is that small peptides are often broken down rapidly in the body. The researchers are exploring ways to extend BRP’s activity, but modifying the molecule to last longer could also change its effectiveness or safety profile. A New Direction for Obesity Treatment Obesity is a complex, chronic condition shaped by genetics https://www.psychologytoday.com/us/basics/genetics , metabolism, medications, sleep, stress, environment, mental health, food access, and social conditions. No single molecule will address every cause or work for every person. Still, BRP represents an encouraging shift. Rather than merely modifying existing GLP-1 drugs, scientists may be able to identify previously unknown hormones and brain pathways that regulate appetite with greater precision. The discovery also challenges the persistent idea that hunger should be controllable through discipline alone. Appetite emerges from complex neural https://www.psychologytoday.com/us/basics/neuroscience and hormonal systems, many of which operate outside conscious awareness. BRP is not yet a natural alternative to Ozempic. It is an early-stage experimental molecule supported by promising animal data. Human trials will determine whether that promise survives the difficult transition from laboratory discovery to medical treatment. For now, the real breakthrough may be the method: using AI to reveal biological messages the human body has been sending all along. References Coassolo, L., B. Danneskiold-Samsøe, N., Nguyen, Q. et al. Prohormone cleavage prediction uncovers a non-incretin anti-obesity peptide. Nature 641, 192–201 2025 . https://doi.org/10.1038/s41586-025-08683-y https://doi.org/10.1038/s41586-025-08683-y Stanford Medicine. "AI helps Stanford scientists discover “natural Ozempic” without the usual side effects." ScienceDaily , 24 July 2026. < www.sciencedaily.com/releases/2026/07/260719230608.htm http://www.sciencedaily.com/releases/2026/07/260719230608.htm ;.