Over 100 new genes linked with the foods we like

Data from over 2,000 Viking Health Study - Shetland volunteers helped to contribute to this important discovery.

Friends eating pizza

Over a hundred genes can be associated with our love of different foods. Genes for liking Avocadoes, chilli peppers, aniseed, steak, oily fish and curry have all been identified in the first-ever large scale study of the genetics of food liking.

Scientists used the latest statistical analysis to develop a detailed food map. It shows the genetic determinants - and associated neurophysiological factors - linked with the food people like. The identified genes go well beyond taste receptor genes.

Researchers from the University of Edinburgh and the Human Technopole, Milan, tested 187,000 volunteers from the UK Biobank. They wanted to understand how much people like 139 specific foods, using a nine-point scale.

They also used data from over 2,000 of our Viking Health Study - Shetland volunteers. This same questionnaire has been answered by over 3,000 Viking II volunteers. 

The team then used a research approach, known as a Genome Wide Association Study (GWAS), to study the data. It compared the genetics of respondents with their food preference answers. This allowed researchers to search for genetic variants that occur more often in volunteers who like specific foods more than other volunteers.

Three distinct food groups

The resulting food map found three distinct food groups:

  • Highly palatable foods including meat, dairy and desserts
  • Strong-tasting foods defined as ‘acquired’ ranging from alcohol to pungent vegetables
  • Low-calorie foods, such as fruit and vegetables.

No relationship was found between the highly palatable food group and the other two food groups. This suggests there might be independent biological processes underlying the liking of high reward foods compared with the low-calorie and strong-tasting foods.

The three food groups also shared genetics with different health traits. For example, the highly palatable food group shared genetics with obesity and lower levels of physical activity. The low-calorie dimension shared similar genes with higher levels of physical activity.  Finally, the strong-tasting food group shared genetics with a healthier cholesterol profile, lower obesity, and higher physical activity. However, it also shared genetics for a higher likelihood of smoking and higher alcohol intake. 

In addition, the food groups each shared genetics with non-overlapping, distinct networks and areas in the brain. This was measured using MRI scans and suggests separate neural mechanisms underlying the liking of different food groups. For example, the low-calorie and strong-tasting food groups shared genetics with areas involved in sensory responses and decision making.

Over 100 genes

In total, 171 genes were found to influence which foods the volunteers liked. Many genes influenced more than one food-liking trait. For example, variants of the major obesity-risk gene influence 51 food-liking traits, including shellfish and porridge.

However, only a small minority (11) of these were taste or olfactory (smell) receptor genes. The strongest link was an olfactory receptor gene linked to liking onions.

This is a great example of applying complex statistical methods to large genetic datasets in order to reveal new biology, in this case the underlying basis of what we like to eat and how that is structured hierarchically, from individual items up to large groups of foodstuffs.

Prof Jim Flett Wilson
Principal Investigator, Viking Genes Personal Chair of Human Genetics, University of Edinburgh

One of the important messages from this paper is that although taste receptors (and thus taste) is important in determining which foods you like, it is in fact what happens in your brain which is driving what we observe. Another important observation is that the main division of preferences is not between savoury and sweet foods, as might have been expected, but between highly pleasurable (and high calorie) foods and those for which the taste needs to be learned. This difference is reflected in the regions of the brain involved in their liking and it strongly points to an underlying biological mechanism.

Dr Nicola Pirastu
Senior Manager, Biostatiistics Unit, Human Technopole, Milan

The research, published in Nature Communications was funded by the Medical Research Council. The full paper can be found at the link below:

Large-scale GWAS of food liking reveals genetic determinants and genetic correlations with distinct neurophysiological traits