Daniel Hwang remembers the moment his team at the University of Queensland realized a gene linked to the scent of onions might also be quietly shaping human health. What began as a curiosity—how our senses guide what we eat—has now become a groundbreaking genetic framework that could transform how scientists untangle the messy relationship between diet and disease. In a study published in BMC Medicine, Dr. Hwang and his colleagues analyzed 325 taste and smell genes across more than 160,000 adults, uncovering not just how biology shapes food preference, but how it can be used to reveal whether certain foods actually cause chronic illness—or merely correlate with it. This distinction, long elusive in nutrition science, is critical: while decades of observational studies have tied diet to conditions like diabetes and heart disease, proving causation has remained a stubborn challenge.
The new framework leverages Mendelian randomization (MR), a method that uses genetic variation to infer cause and effect, but with a novel twist: it focuses on chemosensory genes—those governing taste and smell—as instruments to predict food intake. Because people with certain variants of these genes are naturally drawn to or repelled by specific foods, researchers can use these genetic markers as proxies for dietary habits, cutting through the noise of self-reported eating patterns. For instance, individuals with genetic variants that heighten their sensitivity to the pungent aroma of onions tend to consume more onions—and, the study found, show lower blood pressure and a reduced risk of type 2 diabetes. This isn’t just correlation; it’s a step toward proving onions may actively contribute to better metabolic health.
Using data from the UK Biobank and validating findings in a younger cohort from the Avon Longitudinal Study of Parents and Children, the team demonstrated that their approach could reliably link food preferences to health outcomes. With unhealthy diets responsible for an estimated 11 million premature deaths each year—mostly from cardiovascular disease and cancer—this tool arrives at a critical time. It offers a faster, more cost-effective path to generating robust evidence without large-scale dietary trials. "Our framework gives scientists another way to study food and disease-related relationships without needing to undertake complicated experiments," Dr. Hwang said. Already, the method is being applied to investigate causal links between diet and conditions like cancer and obesity. As global rates of diet-related diseases climb, this biologically grounded approach could help turn the tide—one gene, one flavor, one vegetable at a time.