By David Brown
Washington Post Staff Writer
Friday, February 22, 2008
We're all pretty much the same except, of course, for the little things that make us different.
Those are the conclusions of three studies published this week that looked at human diversity through the keyhole of the genetic mutations we all carry.
The findings -- the latest dividend from the world's investment in the Human Genome Project in the 1990s -- confirm a broad narrative of human history known from previous biological, archaeological and linguistic studies. But the new research adds an astonishing level of detail and a few new insights that were not previously available.
All three studies support the idea that modern human beings left East Africa, walked into Central Asia and then fanned out east and west to people the entire planet. The studies also confirm earlier research showing that as a group, Africans have more diverse genes than people of other continents. But the new research further shows that genetic diversity declines steadily the farther one's ancestors traveled from Addis Ababa, Ethiopia, which is roughly the site of the exit turnstile for the "out-of-Africa" migration.
The studies also show that many seemingly "purebred" ethnic groups have ancestry traceable to more than one continent.
For example, the Arabian Peninsula's Bedouin -- a culturally distinct group -- are descended not only from longtime Middle Eastern peoples, but also from Europeans and peoples originating from around modern Pakistan. The Yakut people of eastern Siberia share blood with East Asians, Europeans and American Indians, but very little with Central Asians, who are geographically closer to them than two of those populations.
The research may also shed light on the genetic underpinnings of human disease. One study found that Americans of European descent carry a larger number of damaging gene variants than African Americans do -- a byproduct of Caucasians' arduous march westward to the shores of the Atlantic.
The biggest message, though, is that these differences are the details, not the main message, of human diversity.
About 90 percent of the full catalogue of human genetic diversity exists in every human population. Individuals are likely to have almost as many differences with people we consider to be "like us" as with strangers on the other side of the world.
"What this says is that we are all extremely related to each other," said Richard M. Myers, a geneticist at Stanford University School of Medicine, who helped lead one of the studies being published today in the journal Science.
"Most genetic variation is shared worldwide. It is only a small part of human genetic variation that is private to particular continents," said Noah A. Rosenberg, of the University of Michigan. His group's findings were published yesterday in the journal Nature.
All three studies examined single "letter" changes in the 3-billion-letter transcript that makes up each person's genome. Every individual carries tens of thousands of these variations. Some do not change the "words" that are the genes; some change a word but not its meaning; and some change the meaning in a way that can be beneficial or harmful.
Each person's collection of these changes (called "single nucleotide polymorphisms, or SNPs) contributes to his or her individuality. People with a common ancestry, however, tend to have similar collections of SNPs (pronounced "snips").
"There is no single gene, no single DNA marker, that would distinguish one population from another," Myers said. Instead, he said, "it is a pattern, like a bar code with thousands of lines on it," that allows researchers to tease apart the fine points of relatedness among populations.
He and his colleagues looked at 938 individuals from 51 different populations whose DNA is in a repository in France. The group lead by Rosenberg and Andrew B. Singleton, of the University of Virginia, studied 485 people from the same collection. Each person studied had a clear-cut ethnic identity and in most cases came from a family that had lived in the group's homeland for generations.
With such diverse and abundant starting material, the researchers were able to sketch a picture of ethnicity far more detailed than previously known.
For example, Africa's surviving hunter-gatherers -- two groups of pygmies and the San people of southern Africa who were formerly known as Bushmen -- are closely related to one another and quite distinct on a genetic basis from all other black Africans. The Hazara of Afghanistan and Pakistan and the Uygur of northwestern China are also close genetic relatives, despite living far apart. On the other hand, China's dominant ethnicity, the Han, is actually two genetically distinct groups, the northern and the southern Han.
The research shows that populations' genetic footprints on the planet are deep, sharp and not easily covered over by time.
Both research teams using the French DNA collection found geographic distance from East Africa is a major determinant of genetic differences among groups.
"Each group carried only a subset of the genetic variation from its ancestral population. So there is a loss of genetic diversity with the distance from Africa," Rosenberg said.
One of the more interesting consequences of that pattern is the subject of the third study, also published in Nature.
Carlos D. Bustamante of Cornell University and his colleagues measured SNPs in 20 European Americans and 15 African Americans. They found that the average person carries at least 2,000 SNPs that change the meaning of a genetic "word." However, in the European Americans, a larger proportion of those changes were likely to be unhealthy or unfavorable.
The reasons for this curious finding are not fully known, although there are theories.
The chief explanation is that the ancestors of Europeans (and most white Americans) suffered repeated population "bottlenecks" in which their numbers crashed as result of epidemics, environmental catastrophes and genocide. Each time that happened, the population lost a lot of its genetic diversity simply because a lot of people died.
The survivors, like their ancestors, carried a certain random collection of deleterious SNPs -- genes that caused disease or increased the risk of disease. When the population rebounded, those genes were spread widely as the small number of survivors gave rise to all living descendants.
But if they were potentially bad, why weren't they flushed out by natural selection? That is the mystery.
It may be that in the rebound after the bottleneck, the slight hazardousness of these SNPs did not make much difference. New conditions, perhaps territory free of competitors or a new technology for getting food, allowed people who carried them to flourish just as well as people who did not have them.
It is even possible that some deleterious SNPs were "dragged along" into the future because they were physically close on chromosomes to newly arising SNPs that increased a person's biological fitness.
With time, natural selection will tend to flush the deleterious SNPs out, as it has done -- relatively speaking -- in African populations. But in the case of Europeans, not enough time has passed.