By now you have read about the new paper in Science, Greenlandic Inuit show genetic signatures of diet and climate adaptation. Carl Zimmer has an excellent treatment in The New York Times, Inuit Study Adds Twist to Omega-3 Fatty Acids’ Health Story. The backstory here is that for decades people have been told to take fish oils because of their possible protective role against heart disease. Apparently some of these recommendations were based on observing the dietary habits of indigenous peoples of the Arctic and their health outcomes. Unfortunately studies which attempt to gauge the impact of these recommendations on Western populations have come back mixed at best. I myself stopped taking fish oils years ago after a review of the literature and asking around. Well, it turns out that there may have been a confound that the populations of the Arctic are adapted to their particular diet.
The figure to the right gets at the heart of the result. Greenland Inuit (they selected for individuals with less than 5% European ancestry), Europeans, and Chinese, exhibit a particular genome-wide pattern of relatedness, which you can see at the bottom. Looking at their results the authors found that there is gene flow from a Greenland-like population to the Chinese at some point over the last 20,000 years. This seems plausible. Additionally, I recall that Greenland natives and Europeans share Ancestral North Eurasian heritage. This is not a population genomics paper focused on phylogenomics, so these details aren’t too important. The takeaway is that on a set of derived alleles around the fatty acid desaturase genes the populations of Greenland seem fixed for variants which are very different from the major alleles in both Chinese and Europeans. These genes are very extreme in terms of their results on the population branch statistic (PBS), which measures deviations in allele frequency against reference groups.
The details are somewhat gnarly. The authors look at several groups of genes, before zeroing in on the FADS group, and they also look at several variants within FADS, as well as various phenotypes. It turns out some markers make the Inuit differ in height and weight, and the height result also applies to Europeans (the frequency is far lower, so that may be why it wasn’t picked up in earlier GWAS).
But I want to focus on a major top-line result. First, here is Rasmus Nielsen in Carl’s piece, “The same diet may have different effects on different people.” And from the paper itself: “In addition to the associations with height, we also found known associations with low fasting serum levels of insulin, total cholesterol, and LDL cholesterol for European carriers of low-frequency–derived alleles of FADS1 variation, suggesting that there may be a protective effect of these variants on cardiometabolic phenotypes.” The implications of this study are commonsense, but they’re also very deep, as they confirm a deep intuition that the same dietary regime may not have the same outcome in all humans. As the authors note in the piece many of alleles at high frequency in Greenlanders are also at high frequency in American native populations in general. Looking at the time depth of the selection event it seems likely that a lot of change occurred in Beringia or Siberian, so for New World groups this may be an ancestral suite of characteristics. But perhaps even more interesting is that many populations have high minor allele frequencies of these alleles. I looked at one marker in the 1000 Genomes data set, and the range is wide. Many Eurasian populations have the “Greenland” variant at ~10% frequency, so ~1% might be homozygote for that genotype.
What that means is that studies in small populations like the natives of Greenland may still have wide-ranging implications. There are literally hundreds of millions of people with these alleles. Though one might suggestion caution about extrapolating results out-of-population, some of the phenotypes are replicated already in Europeans who have the variant.
This study can’t be understood in isolation. It allows for broader generalizations. Ten years ago I read Why Some Like It Hot: Food, Genes, and Cultural Diversity. This was really a pre-genomic era book, drawing on an older body of work. But it is very interesting, and reports on a wide range of studies and the author’s own experiences. Much of it won’t be a surprise to many, but others would still benefit from its comparative method. Today we know a lot more about population-level variation, and what it might tell us about individual-level variation. It’s going to be fun times ahead, as I suspect that the intersection of diet, nutrition, genomics, and quantified-self is going to be a very big deal in the near future.How and what we eat is important. The diet industry is nearly a one hundred billion dollar market.