Researchers at Stanford Medicine have used a computational screening pipeline to identify a naturally occurring peptide that curbs appetite in animals without the digestive side effects that make semaglutide — the active ingredient in Ozempic and Wegovy — difficult for some patients to tolerate. The molecule, a 12-amino-acid peptide dubbed BRP, was described in a study published in Nature in 2025. The essential caveat: it has never been tested in people, and the distance between encouraging animal data and an approved therapy is measured in years and in failures.
- AI-guided discovery: A tool called Peptide Predictor screened roughly 20,000 human genes and narrowed the field to 100 candidates for laboratory testing.
- Animal-only results: BRP cut short-term food intake by up to 50% in mice and miniature pigs, and reduced body fat in obese mice over 14 days.
- A distinct mechanism: The peptide works through the hypothalamus independently of the GLP-1, leptin and melanocortin 4 receptors.
- No human timeline: The receptor it binds to is still unknown, and no clinical trial has been announced.
Screening 20,000 genes for overlooked signals
BRP takes its name from BRINP2, the prohormone it is cleaved from. Prohormones are proteins that enzymes slice into smaller fragments with biological roles of their own; GLP-1, the hormone mimicked by semaglutide and its rivals, is the best-known example. The enzyme responsible for many of those cuts is prohormone convertase 1/3, or PCSK1, and according to the study, a large share of the peptides it generates had never been characterized.
The team, led by researcher Katrin Svensson, built a computational tool called Peptide Predictor to close that gap. It scanned roughly 20,000 human protein-coding genes for the sequence patterns where PCSK1 tends to cut. The funnel narrowed fast: 373 candidate prohormones, 2,683 possible peptides, and finally 100 selected for testing in cultured neuronal cells. BRP boosted the activity of those cells roughly tenfold versus the control; GLP-1 managed about threefold.
“The algorithm was absolutely key to our findings,” Svensson said.
It is worth being precise about the kind of artificial intelligence involved. This is not a generative model inventing novel molecules; it is a prediction system that narrows a vast search space so that a laboratory tests 100 candidates instead of thousands.
What the animal studies found
In lean mice and miniature pigs, a single injection before a meal cut food intake over the following hour by as much as 50%. In obese mice given daily injections for 14 days, treated animals lost roughly 3 grams on average, mostly fat, while controls gained about the same amount. Glucose and insulin tolerance also improved.
Just as notable is what did not appear. Behavioral tests found no differences in movement, water intake, anxiety or stool output, according to Stanford, and the peptide produced none of the effects tied to semaglutide, such as nausea, constipation or significant muscle loss. For scale, in the STEP 1 trial of semaglutide in people, published in the New England Journal of Medicine in 2021, 44.2% of participants reported nausea versus 17.4% on placebo, while mean weight loss at 68 weeks reached 14.9% against 2.4% for placebo. That is the benchmark, and BRP is nowhere near it yet.
A separate pathway in the brain
According to the authors, BRP activates the FOS protein in the brain and acts primarily on the hypothalamus, the region governing appetite and metabolism. It does so independently of the GLP-1 receptor, leptin and the melanocortin 4 receptor. Semaglutide, by contrast, engages receptors in the brain, gut and pancreas, which helps explain its broader effects.
That independence is what makes BRP more than another Ozempic clone: if it works through a different route, it could serve patients who cannot tolerate current drugs, or be combined with them. These remain hypotheses, however — the receptor BRP binds to has not been identified.
No timeline, and plenty of unfinished work
There is no date for human testing. Stanford says Svensson co-founded a company, Merrifield Therapeutics, to move the peptide into clinical trials “in the near future,” and she and first author Laetitia Coassolo are listed as inventors on BRP patents.
“We’re very eager to learn whether it is safe and effective in humans,” Svensson said.
Coverage from 2026 reviewed so far records no trial underway. Before one can begin, the team must identify the receptor, map the signaling cascade and extend the peptide’s duration in the body to allow a more convenient dosing schedule.
The real story here may be the method rather than the molecule. A computational sweep of the human genome surfaced a bioactive peptide the body already makes and nobody had characterized, and the authors argue the same approach can uncover others. The “Ozempic rival” framing that accompanied the study is harder to defend: three grams in a mouse and one hour of reduced eating are not an obesity treatment, and rodent success guarantees nothing in humans. Weight biology is stubborn, and the first human trials will decide whether BRP becomes an alternative or simply a compelling story.