For most of human prehistory, Homo sapiens made a living by hunting and gathering. Then about 10,000 years ago, the gatherers turned to farming and the hunters to domesticating animals.

"Homo had never been able to do that before," observed genomicist Eric Lander, director of the Whitehead Institute/MIT Center for Genome Research, in Cambridge, Mass. "Human civilization the last 10,000 years has been under all sorts of new evolutionary pressures - new environments, new diets, infectious diseases.

"And they've shaped to the human genome," he added, "by selecting for different variants. But we haven't been able to see what they are. This now gives us a way to scan across the human genome and ask, What variations in the genome have been selected in recent times - the last 10,000 years or so?' So it's a way for infectious diseases to find all sorts of genes that have been under selection, to protect us from contagions - and I think that has important medical consequences.

"Throughout human civilization," Lander pointed out, "human beings have been subject to a variety of selective pressures - new food sources, cultural environments and infectious diseases. The last 10,000 years," he opined, "have therefore seen some of the most interesting times in human biological history, and may be when disease resistance arose." Lander is senior author of an article in Nature, dated Oct. 9, 2002, titled: "Detecting recent positive selection in the human genome from haplotype structure." (Haplotypes are parentally bequeathed gene variations.)

"In this paper," he told BioWorld Today, "we report having designed a powerful new approach to scan the entire human genome and pick out the genetic variations that have undergone natural selection in recent evolution. The approach," he added, "relies on the relationship between the ancestral age of a genetic variation and its frequency in the human population."

Lander elaborated: "Say some random DNA spelling occurs in the human population. It might be that over a long time it becomes frequent just by slow genetic drift. And the way you know it takes a long time is by looking at how far away you still see correlations with other spelling differences. In every generation," he continued, "recombinations can occur, and they begin to break down any correlation at long distances."

Long-Distance Scans Of DNA Base-Pair Spelling

"The smoking gun - the evidence - is that if a particular spelling variant rises to high frequency in the population quickly, it will be obvious because it shows long-range correlations to distant spelling differences. Maybe an A or a T nucleotide here, and a million bases away there's a G or a C somewhere. Usually, there won't be an association between these two base pairs.

"The original A arose maybe the first day on a chromosome that had a G," Lander went on. "But as time goes on, it breaks down. However, if something was under strong positive selection, it could rise up to very high frequencies very quickly, and then you'd see that at long distances the smoking gun is there. In other words, the correlation between variants provides a kind of clock, almost like radioactive decay provides a clock to measure the age of samples. The breakdown of base correlations at long distances provides a clock for how long a particular DNA spelling difference has been around in the population. If the clock says it's recent but very frequent, that is positive selection.

"And that's all there is to the finding in our Nature paper," Lander said. "It's exceedingly simple. All we're trying to do is look for things that have become too frequent too quickly.

"What we did," he recounted, "was look at the DNA in each of two malaria-related genes, and measure the correlation with spelling differences hundreds of thousands and millions of bases away. To do it, we examined a couple of hundred people in Africa, to see what variants they had. We asked how often there was an A in one person, and a G a million bases away in another.

"The malaria-associated G6PD gene resides on the human X chromosome, and encodes a metabolic enzyme," Lander noted. "Its function in malaria resistance is not fully understood. But for 30 years, people have known that G6PD deficiency classically correlates with resistance to malaria. The malaria-resistance version of that G6PD gene sits in effect in very large historical chunks of DNA in chromosomal regions that haven't recombined. All the other flavors of the gene sit in short blocks, the result of much longer periods of recombination. The second gene we studied is newly linked to malaria. CD40 ligand is an immunological locus not yet confirmed as being a possible malaria-resistance factor, but it makes a little more sense because it's involved in immunological activation.

"The proximate medical application just now," Lander suggested, "is to find a general way of scanning the genome populations that have large burdens of infectious disease, to discover novel resistance factors. I think this has major implications for the Third World, where people are under exposure to many, many infectious diseases. Africa and India are typical examples. And we know the tip of the iceberg. We know G6PD and a few other loci. But I suspect there are hundreds more."

Out Of Africa: Data On Malaria-Related Genes

"In Africa," Lander said, "our team sampled a large population of individuals, analyzed a large number of their genes and measured long-range correlations. Then all they had to do was genotype. It's a completely unbiased way to be able to ask, What parts of the human genome have been strongly shaped by evolution in the last 10,000 years - basically by the forces of population density and civilization?'

"There's so much information to be learned by the human genome's own defenses against disease," Lander summed up, "that while we're all trying to invent such protective strategies by thinking up good ideas, the genome has been at it for longer than we have. Perhaps because it's more patient than we are, it has a longer time horizon, a lower hurdle rate, and probably has been able to experiment a lot more ways to resist disease. It would be a good idea," he concluded, "to ask what it thinks."