Stanford Medicine researchers have used artificial intelligence to identify a naturally occurring peptide that reduced appetite and promoted fat loss in animals without producing several effects commonly associated with GLP-1 medications. The discovery has quickly attracted attention as a possible “natural Ozempic,” although that description requires an important qualification: the molecule has not yet been proven safe or effective in humans.
The experimental molecule is known as BRINP2-related peptide, or BRP. It contains only 12 amino acids and is derived from a protein naturally encoded by the human body. In laboratory studies, BRP appeared to influence appetite through a brain pathway distinct from the one targeted by semaglutide, the active ingredient used in Ozempic and Wegovy. The findings were published in the peer-reviewed journal Nature in March 2025.
The results are encouraging because BRP reduced food intake in both mice and minipigs, while longer treatment helped obese mice lose primarily fat rather than significant muscle. However, BRP remains an early-stage research candidate rather than a treatment that patients can currently obtain.
How AI Found a Previously Unknown Peptide
The discovery began with a difficult biological search problem. The human body produces large precursor proteins called prohormones. Enzymes can cut these proteins into smaller peptides, some of which function as hormones that control appetite, metabolism and other processes.
Traditional laboratory methods make it difficult to distinguish meaningful peptide hormones from the enormous number of fragments created during ordinary protein processing. Stanford researchers therefore developed a computational system called Peptide Predictor to search for likely cleavage sites across approximately 20,000 human protein-coding genes.
According to Stanford Medicine’s explanation of the discovery, the algorithm narrowed the search to 373 potential prohormones and predicted that they could generate 2,683 unique peptide sequences. Researchers then selected 100 promising candidates and tested whether they activated cultured neuronal cells.
GLP-1 increased neuronal activity approximately threefold compared with control cells. BRP produced a much stronger response in this initial screening, increasing activity tenfold. That result did not automatically prove that BRP would cause weight loss, but it gave researchers a strong reason to test the molecule in living animals.
This is where artificial intelligence provided its greatest value. It did not independently invent or approve a weight-loss drug. Instead, it helped scientists examine a biological search space that would have been extremely slow and expensive to investigate manually.
What Happened When BRP Was Tested in Animals
Researchers administered BRP before feeding lean mice and minipigs. During the following hour, food consumption fell by as much as 50% in both animal models.
Minipigs were included because their eating behaviour and metabolic characteristics can provide information that is sometimes more relevant to humans than results obtained from rodents alone. Even so, successful testing in pigs does not guarantee that the same dose, safety profile or appetite response will occur in people.
The team also gave daily BRP injections to obese mice for 14 days. Treated animals lost an average of about three grams, while mice in the control group gained approximately three grams over the same period. The reported weight reduction came almost entirely from fat, and the treated mice also showed improved glucose and insulin tolerance.
These results made BRP particularly interesting because some weight lost during aggressive appetite suppression can come from lean tissue. Preserving muscle matters for strength, mobility, metabolic health and the long-term maintenance of weight loss.
Why BRP May Avoid Certain GLP-1 Side Effects
Semaglutide works by activating GLP-1 receptors found in several parts of the body, including the brain, pancreas and digestive system. This broad activity helps regulate appetite and blood glucose, but it also slows gastric emptying.
That delayed movement of food through the stomach contributes to fullness, while also helping explain why patients may experience nausea, vomiting, constipation, diarrhoea or acid reflux. The Stanford Medicine guide to GLP-1 medications explains that these gastrointestinal effects are closely connected to how the drugs operate.
BRP appears to follow a more targeted route. The Stanford team found that it activated different neuronal and metabolic pathways and seemed to act strongly in the hypothalamus, the brain region involved in appetite and energy regulation.
During the animal experiments, researchers did not find meaningful differences in movement, water consumption, anxiety-like behaviour or faecal output between treated and control groups. The animals also did not display the behavioural signs researchers commonly use to identify nausea or food aversion.
Those findings support the possibility of a treatment that reduces hunger without extensively affecting the digestive system. They do not establish that BRP has no side effects. Human biology may respond differently, and uncommon or long-term problems cannot be detected through a relatively short animal study.
Why Calling It “Natural Ozempic” Can Be Misleading
BRP is natural in the sense that its underlying sequence comes from a protein encoded by the human body. That does not mean people can obtain an effective dose by eating a particular food, taking an herbal supplement or following a special diet.
Any future BRP medicine would probably need to be manufactured, purified, formulated and administered at a controlled dose. The animal studies used injections rather than an ordinary food-based source.
BRP is also not simply a natural version of semaglutide. Semaglutide mimics GLP-1 and has effects across multiple organs. The Nature research paper reported that BRP acted independently of the GLP-1 receptor, leptin and the melanocortin-4 receptor. It should therefore be viewed as a possible new class of appetite-regulating medicine rather than a direct copy of Ozempic.
The comparison is useful for communicating its potential, but it can create unrealistic expectations. Ozempic and related medicines have been evaluated in large human trials, prescribed extensively and studied for recognised medical uses. BRP has so far been tested mainly in cells, mice and pigs.
Important Questions Still Need Answers
One of the largest scientific unknowns is the identity of the receptor that BRP binds to. Researchers observed the pathways activated after treatment, but they had not fully established the precise cell-surface target responsible for the peptide’s effects.
Identifying that receptor will help determine whether BRP could influence other brain functions beyond appetite. It may also allow scientists to design improved versions with greater potency, stability or selectivity.
Duration is another challenge. Naturally occurring peptides can break down rapidly inside the body. The research team is investigating ways to extend BRP’s activity so that a future medicine would not require inconveniently frequent dosing.
Clinical development would also need to examine cardiovascular effects, blood glucose changes, neurological safety, gastrointestinal symptoms, drug interactions and the possibility that appetite returns after treatment ends. Long-term trials would be required to determine whether any weight reduction can be maintained.
Human Trials Will Decide Whether the Excitement Is Justified
Senior researcher Katrin Svensson co-founded Merrifield Therapeutics to support the molecule’s development. Stanford also disclosed that Svensson and lead researcher Laetitia Coassolo are inventors on patents involving BRP peptides for metabolic conditions. Those commercial interests do not invalidate the research, but they are relevant when evaluating optimistic claims about its future.
No publicly confirmed human results were identified at the time of writing. Until properly controlled clinical trials are completed, nobody knows whether BRP will reproduce its animal benefits in people or whether new safety concerns will emerge.
Drug development is filled with treatments that performed impressively in animal models but failed during human testing. Differences in metabolism, dosage, immune response and disease complexity can change both effectiveness and safety.
AI Could Transform More Than Weight-Loss Research
The wider significance of the study extends beyond BRP. Peptide Predictor mapped thousands of previously uncharacterised fragments that may contain other biologically active molecules.
Some could eventually influence inflammation, pain, cardiovascular function, behaviour or hormone regulation. The research therefore demonstrates how computational tools can uncover potential medicines already hidden within human proteins.
AI did not replace experimental science in this case. It reduced the number of candidates that scientists needed to test, after which laboratory and animal experiments established which molecule deserved further attention.
BRP may eventually become a more targeted obesity treatment, or later research may reveal limitations that prevent clinical use. For now, it represents a promising starting point rather than an available alternative to semaglutide.
The discovery’s strongest message is not that a side-effect-free Ozempic has arrived. It is that scientists may have uncovered a completely different appetite-control pathway and AI helped them find it far sooner than traditional methods could have done.
