Both products draw on the same landmark study. Here's how they use it differently -- and what that means for you.
If you've been following genetics and weight-loss drugs at all this year, you've probably seen the headlines: in April 2026, 23andMe's research arm published the largest genetic study ever conducted on GLP-1 receptor agonists -- the drug class behind Ozempic, Wegovy, Mounjaro, and Zepbound. The study, led by researchers including Q.J. Su and J.R. Ashenhurst and published in Nature, analyzed genetic data from nearly 28,000 people who had taken these medications. Shortly after, 23andMe rolled out a new report -- GLP-1 Medications: Weight Loss and Nausea -- to members of its Total Health service, built directly on those findings.
We read the same paper. We built a different report. Here's an honest look at both.
What the Nature Study Actually Found
Two genetic signals came out of the study as the clearest findings.
The first sits in GLP1R, the gene that encodes the receptor these drugs are designed to activate. A missense variant here -- one that changes a single amino acid in the receptor's structure -- was associated with a modest but statistically robust boost in weight-loss efficacy, on the order of a fraction of a kilogram per copy of the variant. The effect held up when researchers checked it against an independent cohort in the All of Us research program, which is what gives it real credibility rather than a one-off statistical fluke.
The second sits in GIPR, the receptor that tirzepatide (Mounjaro, Zepbound) activates in addition to the GLP-1 receptor -- the thing that actually makes tirzepatide a "dual agonist" rather than a GLP-1-only drug like semaglutide. A variant here was linked to a meaningfully higher risk of nausea and vomiting, but only in people taking tirzepatide specifically. People on semaglutide with the same variant showed no elevated risk, which makes biological sense: semaglutide doesn't touch the GIP receptor at all, so a GIPR variant shouldn't matter for it.
The most striking number in the whole paper is what happens when someone carries risk-associated versions of both genes: their odds of vomiting on tirzepatide climb roughly 15-fold compared to people without either variant. That's not a subtle effect -- it's the kind of finding that could genuinely change a prescribing conversation.
One thing the paper is careful to say clearly, and that's worth repeating here: genetics is a real piece of the puzzle, but not the dominant one. The study found no single genetic signal that explained overall BMI loss on its own, and the authors are explicit that non-genetic factors -- which drug you're on, your starting weight, your sex, whether you have type 2 diabetes, your dose -- do most of the explanatory work. Anyone telling you a genetic report can tell you exactly how much weight you'll lose is overselling what the science supports.
Where the Two Reports Actually Differ
Coverage. 23andMe's report, understandably, is built around the two headline findings from their own study: the GLP1R efficacy variant and the GIPR tirzepatide-nausea variant. AuraGen's GLP-1 & GIP Metabolic Response report covers those same two variants -- we cite the same paper -- but adds two gene families that don't appear in 23andMe's panel at all: LEP/LEPR, the leptin signaling pathway that governs appetite regulation and satiety, and CYP2D6, a drug-metabolism gene relevant to interaction risk with other medications people on GLP-1 therapy are often also taking.
Structure. Our report is organized around three practical questions rather than a flat list of variants: Will it work (efficacy-related genes), why you eat the way you do (appetite and satiety biology), and which drug, what risk (drug-specific tolerability signals, including the GLP1R × GIPR interaction). The idea is that a genetic report should map to decisions you and your prescriber can actually make, not just hand you a list of rsIDs.
Transparency about null findings. This is the one we're proudest of. Our panel includes CTRB1, a gene some earlier, smaller studies suggested might affect drug response. When we ran it against the current evidence, it didn't hold up as an actionable signal -- so we say that plainly in the report instead of quietly dropping it or dressing it up as a finding. We'd rather show you what we tested and ruled out than pad the panel with genes that sound impressive but don't tell you anything real.
Who's asking the question. This one is worth naming directly, without throwing shade: 23andMe conducted this study in-house, using their own proprietary dataset, and then built a consumer product from it. That's not a conflict of interest in any sinister sense -- it's just worth knowing that the company publishing the research and the company selling you the report based on it are the same company. AuraGen didn't run this study. We're a second, independent read of the same peer-reviewed literature, which is a different kind of check on the findings, not a better one, just different.
GLP-1 & GIP Metabolic Response Report
9 genes, 20+ variants across three clinical layers -- efficacy, appetite biology, and drug-specific risk. Includes the GLP1R and GIPR findings from Su et al. Nature 2026, plus LEP/LEPR and CYP2D6 coverage not found in any competitor panel. Available as an instant unlock from your existing sequencing data.
View Report Details →Who Each Is Actually Better For
If you're already a 23andMe customer with existing genetic data and you want one more report layered onto a service you're already paying for, their GLP-1 report is a reasonable, low-friction add -- you already have the data, the report shows up in an app you're already using.
If you're starting from scratch, or if you want the fuller metabolic picture -- appetite biology alongside drug response, not just drug response -- AuraGen's report covers more ground from a single sequencing run, and folds into the rest of our panel (skin, wellness, pharmacogenomics) rather than living as a single standalone add-on.
Neither report is a substitute for talking to the physician who's actually prescribing your medication. Both are genuinely useful context to bring into that conversation.
Research Use Only. AuraGen reports are not FDA-cleared and do not constitute medical advice. Always discuss genetic test results with a licensed healthcare provider before making any changes to your treatment.
The Bottom Line
The science here is real and it's new -- this is one of the first studies to put solid genetic evidence behind why the same GLP-1 drug can feel completely different for two people. But the effect sizes are modest, the biggest number in the paper (the 15-fold interaction effect) applies to a specific two-variant combination that most people won't carry, and no genetic test -- ours or anyone else's -- can tell you with certainty how much weight you'll lose.
What it can do is give you and your doctor a more informed starting point, especially around the tirzepatide tolerability question, where the evidence is genuinely strong.
Reference: Su, Q.J., Ashenhurst, J.R., Xu, W. et al. Genetic predictors of GLP1 receptor agonist weight loss and side effects. Nature (2026).