Most of the recent studies have used data from the HapMap, a database of common patterns of human genetic variation produced by an international group of researchers. Four populations were selected for the HapMap—ninety Nigerians, ninety Americans of European ancestry, forty-five individuals from Tokyo, and forty-five from Beijing. For some purposes we will group the Japanese and Chinese together as an "East Asian" sample.
The human genome has about 3 billion bases (the four molecular building blocks that make up DNA) organized into forty-six separate bundles of DNA called chromosomes. For the most part, DNA sequences are the same in all humans, but every few hundred bases, a variable site crops up. These are the only sites in which the bases of DNA are likely to vary from one individual to another.
A particular pattern of variation at these sites is called a haplotype. Imagine three successive variable sites—the first can be G or C (representing guanine or cytosine), the second can be A (adenine) or G, and the third can be T (thymine) or C. A particular individual might have C in the first site, A in the second site, and T in the third site—his haplotype would be CAT—while another person has the haplotype cytosine-guanine-thymine, or CGT. A haplotype is like a hand of poker, while the bases in the variable sites are like individual cards.
And just like cards, haplotypes are shuffled. In each generation, a new chromosome is assembled from the inherited parental chromosomes in much the same way that one can cut two decks of cards and assemble a new deck, a process we call recombination. There can be multiple cuts: in humans, an average of one to three per chromosome.
This means that haplotypes are partially broken down every generation: The complete pattern that existed over the whole of the parent's chromosome will no longer be intact after recombination. However, smaller parts of that pattern are likely to remain unchanged, since a chromosome is millions of bases long and the few breaks that occur are likely to be far away.
Over many generations, any haplotype will eventually be completely reshuffled. But if a favorable mutation occurs on a chromosome, people with that mutation will have more children survive than average, so over time, more and more people will bear that mutation. If the advantage is large enough, the mutation can rapidly become common, before recombination completely reshuffles its original haplotype, rapidly enough that people bearing that mutation will also carry the original local haplotype that surrounded it when it first came into existence. The longer the shared haplotype, the younger the mutation. It's as if part of your last hand of cards showed up again in the new deal: You would guess that there hadn't been much shuffling, and you'd be right.
The HapMap studies looked for long haplotypes (long un-shuffled regions) that existed in a number of individuals in the dataset. Any such shared pattern would be a sign of recent strong selection—quite recent, since recombination eventually breaks down all such patterns.
One well-known example is the gene that makes lactase, the enzyme that digests milk sugar. In most humans, and in mammals generally, lactase production stops around the age of weaning, but in many Europeans and some other peoples, production continues throughout life. This adaptation lets adults drink milk. Lactose-tolerant Europeans carry a particular mutation that is only a few thousand years old, and so those Europeans also carry much of the original haplotype. In fact, the shared haplotype around that mutation is over 1 million bases long.
Recent studies have found hundreds of cases of long haplo-types indicating recent selection: Some have almost reached 100 percent frequency, more have intermediate frequencies, and most are regional. Many are very recent: The rate of origination peaks at about 5,500 years ago in the European and Chinese samples, and at about 8,500 years ago in the African sample. Again and again over the past few thousand years, a favorable mutation has occurred in some individual and spread widely, until a significant fraction of the human race now bears that mutated allele. Sometimes almost everyone in a large geographic region, such as Europe or East Asia, shares a trait that goes back to one such allele. The mutation can affect many different things—skin color, metabolism, defense against infectious disease, central nervous system features, and any number of other traits and functions.
Since we have sequenced the chimpanzee genome, we know the size of the genetic difference between chimps and humans. Since we also have decent estimates of the length of time since the two species split, we know the long-term rate of genetic change. The rate of change over the past few thousand years is far greater than this long-term rate over the past few million years, on the order of 100 times greater. If humans had always been evolving this rapidly, the genetic difference between us and chimpanzees would be far larger than it actually is.18
In addition, we see far more recent alleles at moderate frequencies (20 percent to 70 percent) than we do with frequencies close to 100 percent. Since a new favored allele spends a long time at low frequencies (starting with a single copy), a short time at moderate frequencies, and then a long time closing in on 100 percent, the only explanation is that this rush of selection began quite recently, so that few selected genes are in that final phase of increase.
The ultimate cause of this accelerated evolution was the set of genetic changes that led to an increased ability to innovate. Sophisticated language abilities may well have been the key. We would say that the new alleles (the product of mutation and/or genetic introgression) that led to this increase in creativity were gateway mutations because innovations they made possible led to further evolutionary change, just as the development of the first simple insect wings eventually led to bees, butterflies, and an inordinate number of beetles.
Every major innovation led to new selective pressures, which led to more evolutionary change, and the most spectacular of those innovations was the development of agriculture.
This page intentionally left blank
Was this article helpful?