PHILADELPHIA, June 17, 2026 – Imagine if a doctor could look at your DNA and predict not just your disease risk, but also which foods you're drawn to – and whether those cravings are quietly protecting you. A new study from an international team of researchers found that a specific genetic variant linked to a love of onions is associated with lower blood pressure and a reduced risk of type 2 diabetes.
The research, led by scientists at the Monell Chemical Senses Center in Philadelphia, the National Institutes of Health, the University of Queensland, the University of Bristol, and the QIMR Berghofer Medical Research Institute, is published today in BMC Medicine.
The study advances understanding of how our senses of taste and smell may shape long-term health and introduces a sophisticated new method to determine why certain foods may be genuinely beneficial to health.
The research team reported that common genetic variants in taste and smell receptor genes affect one's liking and intake of a wide variety of foods. Using statistical analysis, they showed that these small DNA variants have direct health consequences for outcomes such as blood pressure.
The field of nutrition science has long wrestled with a question: just because people who eat a lot of vegetables happen to be healthier than those who don't, does that mean the vegetables are the cause? Healthy eaters also tend to exercise more, smoke less, and have higher incomes. Randomized controlled trials, the gold standard for establishing causation, are rarely feasible in dietary studies due to long follow-up periods, high costs, and logistical constraints.
"Long-term, randomized controlled trials are simply not feasible in nutrition, and the findings of observational studies – such as a recent study that found vitamin E supplements lower heart disease risk – are often not replicable through large-scale clinical trials," said Monell Chief Science Officer and co-author Danielle Reed, PhD. "We used Mendelian randomization – genetic analysis – to address this challenge."
Mendelian randomization harnesses the natural lottery of genetics: the version of a gene you inherit at birth is determined by chance, not by lifestyle or health. By using genetic variants as proxies, or "instruments", for dietary habits, the researchers tested causal questions about food and health. They focused specifically on genes encoding taste and smell receptors in the mouth and nose.
"We wanted instruments that were biologically meaningful, rooted in the fundamental biology of how we experience food," Reed said. "Taste and smell genes gave us exactly that: a way into the causal question of diet and health that doesn't depend on people accurately remembering what they ate, or on the assumption that their diet has or hasn't been changed by disease."
The team screened over 1,200 genetic variants across 325 taste and smell receptor genes in data from the UK Biobank, a landmark study of nearly 500,000 British adults. They then tested which variants were linked to food preferences.
They found that 268 genetic variants across 117 taste and smell receptor genes were associated with preferences for 96 different foods. Genetic variants shaped people's liking for everything from garlic and grapefruit to horseradish, broad beans, and aniseed.
The researchers confirmed the results in a completely separate, younger group of people (the children of the 90s study from Bristol, UK), checked that the same genes also influenced how much of each food people actually ate, and ruled out genetic variants that might be confounded by wealth or social factors.
After this filtering, 25 robust genetic instruments for 20 different foods remained. One stand-out was an olfactory receptor gene, OR2T6, which is linked to how much a person likes onions. This variant predicted onion liking in both older and younger adults, it predicted onion consumption, and it showed no associations with social deprivation or unrelated health conditions.
Using the OR2T6 variant as a genetic proxy for onion preference, the researchers tested whether a genetically driven tendency to eat more onions affects health outcomes. They found that people with the genetic variant linked to greater onion liking had, on average, lower systolic blood pressure (approximately 1.3 mmHg per point on the liking scale) and lower diastolic blood pressure (approximately 0.7 mmHg).
They also had a roughly 14% reduced risk of developing type 2 diabetes.
There was no effect on body mass index, blood fats, or blood sugar levels, suggesting this isn't simply a case of onion eaters being generally healthier people.
Onions are rich in quercetin and other compounds with known anti-inflammatory and cardiovascular properties, providing a plausible biological explanation for these findings, Reed said. Previous laboratory and small human studies have hinted at onions' potential health benefits; this new evidence leveraging genetics provides much stronger grounds for taking those effects seriously.
By anchoring genetic instruments in the biology of taste and smell, the approach is more resistant to the confounding and reverse causation that have plagued nutritional epidemiology for decades.
"What we've built is essentially a new way of asking whether a food is genuinely good for you," Reed said. "Previous methods often used statistical signals that turned out to be picking up on the fact that sick people change their diets, which sends the analysis in the wrong direction. Grounding the instruments in chemosensory biology helps avoid that trap."
Co-authors include Monell Senior Research Associate Cailu Lin, PhD; Daniel Liang-Dar Hwang, PhD, Institute for Molecular Bioscience, The University of Brisbane; David M. Evans, PhD, University of Queensland; Nicholas G. Martin, PhD, QIMR Berghofer Medical Research Institute; and Paule V. Joseph, CRNP, PhD, MBA, National Institutes of Health.
The work was supported by Australian Research Council Discovery Early Career Researcher Award DE240100014; NHMRC Leadership Fellowship 2017942; National Institute of Alcohol Abuse and Alcoholism Z01AA000135; the Intramural Research Program of the National Institutes of Health; National Institute on Deafness and Other Communication Disorders; the Rockefeller University Heilbrunn Nurse Scholar Award; the National Institutes of Health Distinguished Scholars Program; the UK Medical Research Council and Wellcome grant MR/Z505924/1; the University of Bristol; and the Monell Chemical Senses Center.
