Taste Preferences and Genetics: How We Tend to Choose Food

Some people find the bitter taste of coffee, tea, or dark chocolate pleasant, while for others it can be too sharp or even unpleasant. Where do these differences come from, and what contribution does genetics make?

Taste preferences are shaped by genetic, cultural, social, psychological, and socioeconomic factors (Rondini et al., Epidemiologia, 2026).

The role of genetics in shaping taste preferences is studied at the level of individual taste receptor genes, primarily those for bitter taste, as well as at the level of numerous genetic variants associated with food preferences. However, how these differences show up in everyday life, namely in food choices, is also influenced by taste experience: family habits, food culture, repeated tasting of foods, and so on (Rondini et al., Epidemiologia, 2026).

How Genetic Variants Affect Bitter Taste Perception

The best-studied genetic contribution to taste perception concerns bitter taste. The TAS2R38 gene (taste 2 receptor member 38) encodes a bitter taste receptor that responds, among other things, to bitter compounds called glucosinolates, found in plants of the cruciferous family (Rondini et al., Epidemiologia, 2026).

Two common haplotypes (sets of variants inherited together) are well described for TAS2R38: PAV and AVI. PAV is associated with higher, and AVI with lower, sensitivity to the model bitter compounds PTC (phenylthiocarbamide) and PROP (6-n-propylthiouracil). These synthetic substances are not found in ordinary food, but they share the same chemical group (N–C=S) with the glucosinolates of cruciferous vegetables (Keller & Adise, Annu Rev Nutr, 2016). That is why PTC and PROP are used as standard substances for testing bitter taste perception. Modern studies more often choose PROP, because it lacks the sulfurous smell and toxicity characteristic of PTC (Rondini et al., Epidemiologia, 2026).

The letters in the names PAV and AVI denote amino acids at three positions of the receptor protein. Each of the three amino acid substitutions corresponds to a separate single nucleotide polymorphism (SNP). One of them, rs713598, causes the substitution of alanine with proline at position 49 of the protein (A49P). The three SNPs are tightly linked, meaning they are usually inherited together, so they form the common PAV and AVI haplotypes. rs713598 is used as a marker of these haplotypes, although a single SNP does not reflect the full combination of variants (Rondini et al., Epidemiologia, 2026).

How strongly these genetic differences relate to actual perception of bitterness is shown by a meta-analysis of PTC and PROP studies. Carrying the PAV haplotype was significantly associated with the sensitive phenotype for these compounds: odds ratio 5.88 (95% confidence interval 3.87–8.95). The AVI/AVI genotype was associated with the non-sensitive phenotype: odds ratio 6.73 (95% confidence interval 4.57–9.90) (Shivam, Exp Clin Psychopharmacol, 2024).

At the same time, genotype does not fully predict individual sensitivity. In a sample of 176 participants, the correlation between rs713598 and PROP sensitivity was moderate (ρ = 0.404). This is consistent with the fact that bitter taste perception depends not on a single genetic variant: humans have 25 functional bitter taste receptor genes, and TAS2R38 variants reflect only part of the genetic variability of this trait (Rondini et al., Epidemiologia, 2026).

A TAS2R38 variant describes how intensely a person may perceive the bitterness of certain compounds, in particular glucosinolates in cruciferous vegetables (broccoli, white cabbage, mustard, horseradish, radish). This is a characteristic of perception. Whether this difference translates into liking or disliking these vegetables is a separate question.

For more on genetic differences in response to caffeine, lactose, and gluten, see the article Lactose, Gluten, Caffeine: Genetic Features of Food Sensitivity.

Research on the Genetics of Food Preferences

A large-scale study showed that genetic associations are not limited to individual taste receptors or specific foods. The researchers used data from the UK Biobank, a large British database of genetic and medical data, selecting 161,625 participants of European ancestry, each of whom rated on a nine-point scale how much they like 139 foods and beverages. These ratings were compared with genetic variants across the entire genome. The analysis identified 1,401 associations covering 173 genetic loci (May-Wilson et al., Nat Commun, 2022).

Preferences for different foods did not form a single genetic pattern. The authors identified three main dimensions: highly palatable foods, acquired tastes, and low-calorie foods. The last dimension included, in particular, vegetables, fruits, and whole-grain products. The genetic correlation between highly palatable foods and the other two dimensions was low (0.05 and 0.16), whereas between acquired tastes and low-calorie preferences it was 0.59. This indicates that different groups of food preferences have a partly different genetic basis (May-Wilson et al., Nat Commun, 2022).

The same study compared liking and consumption frequency of the same foods. The genetic correlation between them, that is, a measure of how much both traits are linked to the same genetic variants, was high. However, the share of differences between people that can be attributed to common genetic variants (SNP heritability) averaged about 8% for liking and about 4% for consumption (May-Wilson et al., Nat Commun, 2022).

Some of the associations found were tested in 11 independent groups of participants with a combined size of up to 26,154 people. It was possible to test 235 associations, and 61 of them were statistically confirmed. However, in 82.5% of the tested associations the relationship had the same direction as in the main study, and the authors explain the smaller number of confirmations by these groups being too small for such a test (May-Wilson et al., Nat Commun, 2022). The effect of each individual variant is moderate (May-Wilson et al., Nat Commun, 2022), so such associations describe patterns at the population level and do not make it possible to predict the preferences of a specific person.

From Taste Perception to Eating Habits

Genetic variability in bitter taste perception does not always show up in food choices. In a study of Italian adults, the rs713598 genotype and PROP sensitivity were not significantly associated with preferences for bitter foods, whereas liking for such foods increased with age (Rondini et al., Epidemiologia, 2026).

A systematic review of 13 studies in adolescents showed mixed results. In some studies, bitter sensitivity was associated with lower liking for bitter food and higher liking for sweet food. However, a meta-analysis of three of the 13 studies found no difference in liking for bitter vegetables between bitter-sensitive and non-sensitive individuals: their mean scores were practically the same (Bawajeeh et al., Nutrients, 2020).

Food preferences can change with taste experience. An umbrella review of 11 systematic reviews showed that after 2–10 repeated exposures to the same vegetable, children up to five years old rated it as tastier and ate slightly more of it (Bell et al., Am J Clin Nutr, 2021).

Genetic differences related to bitter sensitivity help explain why the same food can be perceived differently by different people. At the same time, how this sensory perception translates into preferences and everyday choices is also linked to taste experience, age, and lifestyle.

For more on the difference between food allergy and intolerance, see the article Food Allergy and Intolerance: How to Tell Them Apart and What to Do.

Genetic Data and Your Own Taste Experience

Your own observations show which foods a person enjoys now, but they do not explain what may underlie these preferences: features of sensory perception, previous experience, or established eating habits. A genetic result adds what cannot be seen from observation: which variants of taste receptor genes were inherited and which differences in bitter sensitivity they are associated with.

This information helps you read your own taste reactions differently. For example, if cabbage or broccoli seems too bitter, and the person has a variant associated with higher bitter sensitivity, this reaction may have not only a habitual but also a sensory explanation.

These genetic features can be explored in the Apixmed Prism genetic report on the Ultima panel, where results related to taste and food preferences are part of a broader set of genetic data.

Frequently Asked Questions About Taste Genetics

How Is Bitter Sensitivity Tested?

One common method is a paper disc impregnated with a solution of PROP, a synthetic bitter substance whose perception is linked to TAS2R38 variants. The disc is placed on the tongue, and the person rates how bitter the taste feels (Rondini et al., Epidemiologia, 2026). The result describes how a person perceives bitterness now, while genetic analysis shows the variants associated with differences in this sensitivity.

Do Men and Women Perceive Bitter Taste the Same Way?

Not necessarily. In an Italian study, women on average perceived the bitterness of PROP more strongly than men: 2.51 versus 1.77 points on the scale, and the highest rating was chosen by 17% of women and 6.2% of men (Rondini et al., Epidemiologia, 2026). Whether this is a general pattern remains for other studies to show.

Is Bitter Sensitivity Linked to Body Weight?

A meta-analysis of 27 studies found no significant association between bitter sensitivity and overall body weight. Results for the waist-to-hip ratio differed from those for body weight, but this does not support a conclusion that sensitivity affects fat distribution (Shareef et al., Nutr Rev, 2026).

Explore your genetic traits with Apixmed Prism. 

Genetic test results are not a diagnosis and not a substitute for a doctor's consultation. The Apixmed Prism report provides genetic context that complements examination results and helps you make decisions together with your doctor.

Sources

1. May-Wilson, S., Matoba, N., Wade, K. H., Hottenga, J.-J., Concas, M. P., Mangino, M., Grzeszkowiak, E. J., Menni, C., Gasparini, P., Timpson, N. J., Veldhuizen, M. G., de Geus, E., Wilson, J. F., & Pirastu, N. (2022). Large-scale GWAS of food liking reveals genetic determinants and genetic correlations with distinct neurophysiological traits. Nature Communications, 13(1), 2743. https://doi.org/10.1038/s41467-022-30187-w

2. Shivam, V. (2024). A meta-analysis on polymorphic trait of taste perception mediated by TAS2R38 genotype. Experimental and Clinical Psychopharmacology, 32(5), 497–505. https://doi.org/10.1037/pha0000728

3. Rondini, T., Parri, L., Petricciuolo, M., Fatigoni, C., Villarini, M., Federici, E., Moretti, M., & Acito, M. (2026). Correlation between the TAS2R38 genetic polymorphism, PROP perception and eating habits: A cross-sectional molecular epidemiological study. Epidemiologia, 7(5), 131. https://doi.org/10.3390/epidemiologia7050131

4. Bawajeeh, A. O., Albar, S. A., Zhang, H., Zulyniak, M. A., Evans, C. E. L., & Cade, J. E. (2020). Impact of taste on food choices in adolescence: Systematic review and meta-analysis. Nutrients, 12(7), 1985. https://doi.org/10.3390/nu12071985

5. Bell, L. K., Gardner, C., Tian, E. J., Cochet-Broch, M. O., Poelman, A. A. M., Cox, D. N., Nicklaus, S., Matvienko-Sikar, K., Daniels, L. A., Kumar, S., & Golley, R. K. (2021). Supporting strategies for enhancing vegetable liking in the early years of life: An umbrella review of systematic reviews. The American Journal of Clinical Nutrition, 113(5), 1282–1300. https://doi.org/10.1093/ajcn/nqaa384

6. Shareef, A., Mulaw, G. F., Rathore, K., Irwin, C., Vincze, L., Rigby, R., & Jani, R. (2026). Association between bitter taste sensitivity and weight status in adults: A systematic review and meta-analysis. Nutrition Reviews, 84(1), 84–98. https://doi.org/10.1093/nutrit/nuaf160