Some people immediately recognize cinnamon when they smell it, while others hesitate even when given a choice of names. Could genetic differences, as well as experience and memory, contribute to this variation?
In 2025, Franz Förster and colleagues combined data from up to 21,495 people to investigate genetic differences associated with correctly identifying odors. Their aim was to explore the genetic mechanisms supporting odor identification, whether these associations differ between women and men, and how they relate to sex hormones and neurodegenerative diseases.
Choosing among four names for each of 12 odors
The participants had European genetic ancestry and came from four studies in Central Europe and the United States. They smelled odors from pen-like devices and selected one matching name from four alternatives. The analysis used 12 odors: orange, leather, cinnamon, peppermint, banana, lemon, licorice, coffee, clove, pineapple, rose, and fish.
The outcome measured here was correct identification: recognizing an odor and linking it to a name. The lowest concentration someone can detect, how intense an odor feels, and whether it is liked are different questions. The test also used standardized stimuli representing the odors, not the actual fruits or other objects.
The total-score analysis for the 12 items included 21,495 people; analyses of correct or incorrect responses to individual odors included 18,895. Accounting for factors such as age and smoking, the team compared associations between performance and many differences in DNA. This was a study seeking associations across large numbers of people, not an experiment manipulating genes.
Not a single gene that determines a good nose
At the statistical threshold for each trait, ten independent genetic regions were associated with performance, seven of them newly discovered. Seven regions remained at a stricter threshold that also accounted for testing multiple odors, including four new ones. Finding a region does not establish that a particular gene within it is the cause.
The associations were mainly odor-specific. Even the region associated with the total score owed much of that result to its association with correctly identifying pineapple. The findings do not fit a single measure in which one genetic difference means being better at recognizing every scent.
Candidate genes included not only olfactory receptors that receive odor molecules, but also genes involved in transmitting those signals within cells and in brain development. The authors suggest that multiple stages, from receptors at the entry point to neural processing, contribute to identification ability. However, the study did not experimentally test the function of each candidate.
The same genetic difference can have different associations in women and men
A distinctive feature was the separate analysis of women and men, together with tests of whether association strength differed between them. For orange and cinnamon, one region each showed an association in women but not men. For pineapple, one region was associated in both sexes, with a stronger association in men.
These are not comparisons of average performance showing that women are better at orange or men at pineapple. They show that the relationship between a DNA difference and the probability of a correct answer differed between women and men. The region associated with cinnamon in women was significant at the trait-specific threshold but did not meet the stricter study-wide threshold.
Do sex hormones explain these differences? Computational analyses identified DNA sites near associated regions that might respond to male sex hormones. These are clues for future testing of hormone-related regulation of gene activity. The researchers did not administer hormones or experimentally verify the activity of those DNA sites.
In a separate analysis, the authors used genetic information to investigate potential causal relationships between sex hormones, including testosterone, and identification ability. No significant relationships remained after accounting for multiple comparisons. Because data limitations prevented assessment of estrogen in women, the results cannot establish that sex hormones as a whole are unrelated.
Which direction does the relationship with disease run?
The study also examined relationships between odor-identification ability and Alzheimer’s and Parkinson’s diseases. It used genetic differences associated with disease susceptibility to estimate the direction of potential causal relationships.
For Alzheimer’s disease, the results suggested a direction from higher genetic disease risk to a lower total identification score. However, this result depended strongly on one variant in a gene called TOMM40 and was no longer significant when that variant was excluded. A single genetic difference could affect multiple pathways, so the finding cannot be confined to an explanation in which the disease directly reduces smell function.
TOMM40 relates to the function of mitochondria, which support cells’ energy needs. The authors discuss the possibility that disruptions in this function may be shared contributors to disease and changes in smell. Conversely, the reverse direction, from identifying individual odors to changes in disease risk, was not supported under the study’s criteria. These results do not establish that this odor-identification task can diagnose disease in an individual.
From 12 odors to a broader understanding of smell
Identifying an odor requires not just signals reaching the nose, but previous experience and associations with words. Many of these 12 odors were food-related, and familiarity varies across cultures. The results cannot be generalized directly to everyone worldwide or to every scent.
The formulations of the test odors have changed over time, and information about their ingredients is incomplete. An association with correctly identifying pineapple therefore does not identify a response to a single molecule responsible for pineapple character. Studies with more diverse participants and odors, along with tests of how candidate genes actually function, are needed.
The research received funding from organizations including the DFG, BMBF, and Wellcome. One author reported receiving funding from Owkin for an unrelated project; the other authors declared no competing interests.
Summary
The ability to connect odors with names showed genetic associations specific to particular odors, some of which differed between women and men. An important extension of the findings is that the candidates involved not just receptors but also signal-transmission mechanisms and the nervous system.
Regulation by sex hormones and cellular mechanisms shared with neurodegenerative diseases remain hypotheses for explaining these differences. The next challenge is to connect the newly identified genetic regions to the molecules and cellular processes that actually function there. At which stage of smell processing do differences in recognizing the same scent arise? This study adds clues for tracing that pathway.
References
Förster, F., Emmert, D., Horn, K., Pott, J., et al. (2025). Genome-wide association meta-analysis of human olfactory identification discovers sex-specific and sex-differential genetic variants. Nature Communications, 16, 5434. DOI: 10.1038/s41467-025-61330-y. Full original article. The original article is published under CC BY 4.0.
