Tuesday, September 16, 2008

Is there such a thing as French blood?

The question in my title is deliberately rhetorical and provocative, merely to draw attention. It's like a newspaper heading such as: Must man who bit dog wear muzzle? A more rigorous down-to-earth title for the present post would have been: Are there correlations between DNA and the geographical origins of Europeans? It would appear that the answer to that intriguing question is yes. In any case, what I want to do in this post is to summarize what I've understood—if anything at all—about this question, and about the answers provided by research assisted by the GlaxoSmithKline pharmaceutical company. Maybe readers who are better versed in genetics than me might correct possible blunders in what I have to say... or they might consider that this subject is so fuzzy that it's better not to say anything at all.

Let's start at the beginning. We all know that the basic stuff of life, DNA, can be imagined as a lengthy "word" written by means of only four "letters". In the following fragment of DNA, I've represented the four "letters" by arbitrary colors:

Now, let me drop the inverted commas around "letter": a metaphor for nucleotide. From one human being to another, throughout the planet, 99% of DNA sequences are identical. But every now and again, for such-and-such a fragment of DNA, one of the letters might be different, as illustrated here:

As you can see, in the normal fragment of DNA, the third letter is green, whereas in the case of the individual we've just encountered, the third letter is red. If this kind of variation occurs for at least one in every hundred new individuals they examine, geneticists refer to the changed letter as an SNP, pronounced "snip". In the case of humans, potential snips are commonplace. They probably number around 3 million. But, as I said, for any particular snip, only a small proportion of humans will possess the changed letter. Concerning the vast majority of snips, geneticists have no idea whatsoever of the consequences upon an individual, if any, of the changed letter. On the other hand, certain snips have been identified as sources of possible health problems, meaning that they can be used as medical indicators... which is why snips are of interest to pharmaceutical companies such as GlaxoSmithKline.

Let's get back to the question of European geography. The research project was headed by Manfred Kayser, a geneticist at Erasmus University at Rotterdam in the Netherlands. Researchers were able to use a vast collection of European DNA samples that had been obtained by GlaxoSmithKline in the context of their constant hunt for genes responsible for side effects brought about by certain pharmaceutical products. Within the DNA sample for each European studied by Kayser's teams (including researchers at the University of California, Los Angeles), half a million snips were examined. When I say "examined", that merely means that the researchers noted whether each snip letter, for that individual, was normal or anomalous. The result of this analysis was a huge collection of yes/no snip data for each person being studied. Using conventional number-crunching methods, all this data was reduced in such a way that the individual's snip profile could be represented as a point on a two-dimensional graph. And the researchers added an elementary item of information to each point: namely, the place where that individual happened to be born.

Well, the results were astounding. All the points corresponding to individuals born in France formed a cluster, which was located alongside another cluster of the points corresponding to individuals born in Italy, and so on... In other words, the geneticists' graph of snip profiles was equivalent to a geographical map of Europe! Consequently, it's a fact that, if a new human candidate were to be examined, and his snip profile happened to fall inside the French cluster, there's a good chance that he's a Frenchman.

In fact, there's very little genetic diversity within Europe, because people have remained largely within their territorial borders. Not surprisingly, the researchers found that the greatest diversity existed in Mediterranean Europe, whereas Scandinavian, British and Irish data was more uniform. Noah Rosenberg, a geneticist at the University of Michigan in Ann Arbor, concluded: "A pattern in which genes mirror geography is essentially what you would expect from a history in which people moved slowly and mated mainly with their close neighbors."

It's important to understand that this research has little to do with chromosomes, genes and inheritance. It's simply a matter of the statistical analysis of snip data, correlated with geography. It would be crazy to imagine that the researchers are suggesting, for example, that there's a "French gene" that might be injected into an Englishman (Heaven forbid!) to transform him into a Parisian. That would be just as crazy as the idea of a "lipstick gene" for pigs.


  1. I read something similar a few days ago, unfortunately I didn't bookmark it, I don't even remember if it was in French, German or English - so would be a bit difficult to find it again...

    It was about AIDS. The article said that people living in Southern Europe (say countries covered by the Roman Empire) are more vulnerable to AIDS, because they miss a gene (or a sequence of a gene), while people of Northern Europe have got it. Explanation: people of the North and the South didn't mix before recently.

    I shall try to find this article again.

  2. Europeans have played such a central role in the recent history of humanity, and there's so much data of all kinds about them, that there'll probably be many reports about findings of the type you mention. I like the idea that Mediterraneans might be more prone to AIDS in that they have a metaphorical and genetic "screw loose" [slang, pun intended]. As far as mixing is concerned, I have the impression that brown-skinned Mediterranean males and Scandinavian blonds encountered one another quite some time ago. Naively, I would hope that such encounters have given rise, not merely to AIDS victims, but to... children!

  3. Mea culpa. I suddenly realize, with shame, that I'm sexist. In the above comment, why didn't I think of evoking the perfectly plausible idea that fair-skinned Scandinavian males might have exerted their charm and prowess with bronzed Mediterranean maidens?

  4. You are not sexist, just realistic. It does not work both ways. Usually one sees Scandinavian blond females with brown-skinned Mediterranean males...