STANFORD, CALIFORNIA
Somewhere inside your own body, hiding among 20,000 genes' worth of proteins, there may be a molecule that already knows how to do what Ozempic does — quiet the hunger, shrink the fat, spare the muscle — without making you sick to your stomach in the process. Stanford Medicine researchers say they've found it. They're calling it BRP, and they built an artificial intelligence system just to track it down.
The discovery, first published in the journal Nature and freshly recirculating through the medical press this week as scientists detail exactly how the AI-powered hunt worked, has reignited a question millions of Americans on GLP-1 drugs have been asking for years: is there a way to get Ozempic's results without Ozempic's baggage? For a growing body of obesity researchers, the answer just got a lot more interesting.
"The lack of effective drugs to treat obesity in humans has been a problem for decades," said Katrin Svensson, PhD, assistant professor of pathology at Stanford Medicine and senior author of the study. "Nothing we've tested before has compared to semaglutide's ability to decrease appetite and body weight. We are very eager to learn if it is safe and effective in humans."
The Problem With the Drugs Everyone's Already Taking
Semaglutide — sold as Ozempic and Wegovy — along with its cousins tirzepatide and the rest of the GLP-1 drug family, has arguably reshaped American culture more than any pharmaceutical in a generation. Millions of people have used it to lose significant weight, control blood sugar and, for some, fundamentally change their relationship with food. But anyone who has taken one, or knows someone who has, is familiar with the tradeoffs: nausea that can be severe enough to send people to the ER, chronic constipation, and a well-documented tendency for these drugs to burn through lean muscle mass along with fat — a side effect doctors increasingly warn could leave long-term users frailer, not just thinner.
The reason for all that collateral damage, according to Svensson, comes down to where these drugs act in the body. "The receptors targeted by semaglutide are found in the brain but also in the gut, pancreas and other tissues," Svensson explained. "That's why Ozempic has widespread effects including slowing the movement of food through the digestive tract and lowering blood sugar levels."
In other words, semaglutide doesn't just talk to the part of your brain that controls hunger — it talks to nearly everything, all at once, which is part of why it works so powerfully and also why it makes so many people miserable. Stanford's team went looking for something more surgical: a molecule that would hit the hunger switch and nothing else.
How an AI Found a Needle Made of 12 Amino Acids
The path to BRP started with an entirely different kind of hunt — not through a patient population, but through the human genome itself.
Researchers, led by senior research scientist Laetitia Coassolo, PhD, focused on a particular enzyme called prohormone convertase 1/3, which has previously been linked to human obesity. This enzyme's job is to snip larger, inactive proteins called prohormones into smaller, biologically active fragments called peptides — some of which act as hormones that regulate hunger, blood sugar and energy use. GLP-1, the hormone semaglutide is designed to mimic, is itself one of these fragments, produced when the same enzyme cuts a specific prohormone.
The logic was simple in theory and nightmarish in practice: if that one enzyme produces GLP-1, maybe it produces other useful appetite-regulating peptides too, hiding among the tens of thousands of possible fragments the human genome could theoretically generate. Finding them the old-fashioned way — extracting proteins from tissue and picking through the results with mass spectrometry — would mean sorting through hundreds of thousands of molecules to find the handful that actually matter.
So the team built a computer program instead. They called it Peptide Predictor, and its job was to scan all 20,000 human protein-coding genes for the exact sequences that prohormone convertase 1/3 likes to cut, then narrow the list down to proteins that get secreted outside the cell — a telltale sign of a hormone — with at least four possible cutting sites. That process alone narrowed a nearly impossible field down to 373 candidate prohormones.
From there, Peptide Predictor estimated that those 373 proteins could yield 2,683 distinct peptide fragments. Coassolo and Svensson zeroed in on the sequences most likely to influence the brain, selected 100 of them — including GLP-1 itself as a benchmark — and tested each one on lab-grown neuron-like cells to see which ones lit them up.
GLP-1 performed as expected, tripling the activity of the test cells. But one candidate blew that result away. Made from just 12 amino acids — a fraction of the size of most hormone peptides — it sent neuronal activity up tenfold compared to untreated cells. The researchers named it BRP, short for BRINP2-related peptide, after the larger prohormone it's sliced from.
"The algorithm was absolutely key to our findings," Svensson said.
Cutting Food Intake in Half — Without the Misery
A strong reaction in a petri dish is one thing. What happened next is why obesity researchers are paying attention.
The team tested BRP in lean mice and in minipigs, whose metabolism and eating behavior more closely mirror human biology than a standard lab mouse. A single injection given before feeding time cut food intake over the following hour by as much as 50% in both species — a strikingly fast, strikingly large effect for a molecule nobody had ever studied before.
Then came the real test: could it produce sustained weight loss, the way semaglutide does over weeks and months? Researchers gave daily BRP injections to obese mice for 14 days straight. The treated animals lost an average of 3 grams — with almost all of that loss coming from fat, not muscle — while untreated control mice gained roughly 3 grams over the same period. The treated mice also showed measurably improved glucose and insulin tolerance, meaning their bodies were getting better, not just thinner, at managing blood sugar.
And then came the part that has generated the most excitement: what didn't happen. Researchers ran behavioral tests checking for the tell-tale fingerprints of GLP-1 drug side effects — reduced movement, changes in water consumption, anxiety-like behavior, and, critically, changes in fecal output, a proxy for the constipation that plagues many Ozempic and Wegovy users. Across all of it, the BRP-treated animals looked no different from untreated ones. No signs of the nausea-related behaviors researchers look for in animal models, either, and none of the substantial muscle loss that has become one of the most cited downsides of GLP-1 therapy in humans.
Follow-up measurements of brain and body activity confirmed that BRP was working through pathways distinct from those used by GLP-1 or semaglutide — evidence, the researchers say, that this isn't just a weaker copy of an existing drug, but a genuinely different mechanism for controlling appetite.
Why the Hypothalamus Matters
The key to BRP's apparently cleaner side-effect profile, according to Svensson, is location. The molecule appears to act specifically in the hypothalamus — a small, deep structure in the brain responsible for regulating hunger, body temperature, hormone release and energy expenditure — rather than fanning out across the gut, pancreas and other organs the way semaglutide's target receptors do.
That distinction matters enormously for anyone who has ever had to grit their teeth through a wave of Ozempic-induced nausea. If BRP truly stays contained to the brain's appetite control center, it could, in theory, deliver the hunger suppression without dragging the digestive system and skeletal muscle along for the ride.
Researchers caution there's still real science left to do before anyone can say that for certain. The Stanford team is now racing to identify the specific cell-surface receptor that BRP binds to — the molecular lock the peptide fits into — which will help explain exactly how it changes behavior in the brain and confirm whether the clean side-effect profile holds up under closer scrutiny. They're also working on a familiar problem for small peptides: BRP, like many molecules its size, likely gets broken down quickly in the body, and researchers need to engineer a longer-lasting version before it could be practical as a once-daily or once-weekly treatment the way current GLP-1 drugs are.
What This Means for You
If you're currently on Ozempic, Wegovy, Zepbound or any other GLP-1 medication, nothing changes today — BRP has not been tested in humans, and Svensson's team has been careful to frame these as animal results that still need to clear the enormous hurdle of human clinical trials before anyone can say whether it's safe or effective in people.
That said, the timeline may move faster than typical academic research. Svensson has co-founded a company, Merrifield Therapeutics, built around developing BRP-based peptides for metabolic disorders, and has said human clinical trials are expected to begin in the near future. Anyone hoping for a next-generation obesity treatment with a gentler side-effect profile should watch for updates on those trials over the next one to two years — realistically the earliest point at which meaningful human safety and efficacy data would emerge.
In the meantime, patients considering or currently using GLP-1 drugs should keep working with their prescribers to manage existing side effects rather than waiting on a still-experimental alternative. Diet counseling, dose titration schedules, and management strategies for nausea and muscle preservation — including resistance training and adequate protein intake — remain the tools available today.
The Bigger Picture
The BRP discovery lands at a moment when the obesity drug market, already worth tens of billions of dollars a year, is entering its next phase: not just proving that GLP-1 drugs work, but racing to build something that works just as well with fewer catches. Pharmaceutical giants and biotech startups alike are chasing "next-generation" obesity treatments — oral GLP-1 pills, muscle-sparing combination therapies, and now, AI-discovered natural peptides pulled straight from the human genome.
What makes the Stanford discovery notable isn't just the molecule itself, but the method used to find it. Peptide Predictor represents a new way of hunting for hormones — using computational prediction to search the entire genome for hidden biological signals, rather than relying on chance discovery or laborious lab-by-lab screening. Researchers involved in the project believe the same AI approach could be turned loose on other prohormones to uncover peptides relevant to entirely different diseases, from diabetes to inflammatory conditions.
For now, BRP remains a promising animal study, not a medicine. But in a field that has spent decades searching for anything that comes close to matching semaglutide's power, a naturally occurring molecule that appears to do the job with a fraction of the misery is exactly the kind of lead obesity researchers have been waiting for.
