‘Vegetarian’ gene variant, more common in India, makes up for fatty acids in meat

A genetic variant that helps the body make long chain polyunsaturated fatty acids, found in meat and fish, is more common among traditionally vegetarian populations according to research published in Molecular Biology and Evolution. The variant was found in 68% of Indians from Pune compared to 18% of Kansans. The authors caution that as modern diets change, people with the ‘vegetarian’ variant may produce a higher ratio of omega-6 to omega-3 fatty acids, which is considered a risk factor for lifestyle-related diseases.


Dr. Marcus Feldman, Burnet C. and Mildred Finley Wohlford Professor of Biological Sciences, Stanford University (webpage):

Expertise: Human genetic and cultural evolution, mathematical biology.

“Trying to find out whether there is a difference in the evolutionary history associated with vegetarianism and non-vegetarianism is a fascinating problem. Pune, where they got the Indian sample from, has been a majority Brahmin city for a very long time and the Brahmin caste has been identifiably vegetarian for between 2,000 to 2,500 years.  What that leaves for us to understand is whether the selection detected by the various statistical methods employed in the paper is strong enough to have made the observed difference within 2,500 years. The data are suggestive but not conclusive.

“One of the metrics used to indicate that selection may have occurred is called the F statistic, and the authors show the authors show a significant differences between South Asians and Europeans via this statistic. It would be really interesting to know what the F statistic was between the vegetarians, who made up 38% of the 234 people in the Pune sample, and the non-vegetarians, who made up 62%.  If there was a significant difference within the Pune sample, it would be difficult to say that the high proportion of the ‘vegetarian’ allele is characteristic of the whole of South Asia.

“I think we’re a long way from extrapolating from population type analyses of this kind to suggesting how an individual should make a decision about diet. The paper gives powerful evidence of some kind of selection over the past 2,500 years but it doesn’t help an individual to decide whether being a vegetarian is healthier than being a non-vegetarian.

“We know from other studies that diet can be associated with differences in gene expression, and gene expression can vary with body mass index (for example with obesity). What we don’t get from the data in this paper is any associations that are covariant with risks of chronic disease, such as associations with BMI, and the risks of chronic diseases could vary significantly among the populations that they looked at.”


Dr. David Cutler, Associate Professor, Department of Human Genetics, Emory University (webpage):

Expertise: Population genetics applications to human disease studies, analyzing whole genome data sets to discover genetic variants associated with disease.

This is a nice, simple paper, not to be overblown or overstated.

“Certain types of fats are essential for human life. One particular type (long chain polyunsaturated fatty acids, LCPUFA) can be either eaten or created by the human body. People who regularly eat meat generally get enough LCPUFA to live healthy lives. People who eat little meat must make LCPUFA from other building blocks.  There are different versions of a gene that appear to influence how efficiently people make LCPUFA from building blocks.  One version appears to be somewhat more efficient at making LCPUFA than the other version.   The more efficient version is shown by these authors to be at very high frequency in a South Asian community that eats very little meat. The authors further show that this version likely increased to its current frequency relatively recently (hundreds or thousands of years ago, not tens of thousands of years), suggesting that the increase in frequency of the efficient gene was very likely due to natural selection in these largely vegetarian communities.

“The effect sizes here might be incredibly small.  The data is consistent with something like a 1 in 1,000 chance of dangerous LCPUFA deficiencies among vegans with the less efficient gene (and is thus a really small risk in absolute terms).  While the results are interesting, and very likely to be correct, it is far  from clear that they have any important health implications.”


Declared interests (see GENeS register of interests policy):

No interests declared.



Positive Selection on a Regulatory Insertion–Deletion Polymorphism in FADS2 Influences Apparent Endogenous Synthesis of Arachidonic Acid‘ by Kothapalli et al, published in Molecular Biology and Evolution on Tuesday 29 March, 2016.

Please feel free to leave your comments below, but be aware that by doing so you agree to our Terms & Conditions.