Rosenberg lab at Stanford University
People are same, but different
Humans can be sorted into five groups based on ancestry, major genetic study finds.
By Robert Lee Hotz, Los Angeles Times, Friday, December 20, 2002


People the world over are almost identical, yet still so different genetically that they can be easily sorted into five major groups based on ancestry, new research shows.

In the largest study so far of human genetic variation, an international research team separated people by the major migrations of ancient humankind, from Africa into Eurasia, East Asia, Oceania and the Americas, in a way that overturns conventional notions of race.

With growing assurance, scientists are overturning deep-seated prejudices over what makes human beings different -- skin color, facial features, physique -- even as they risk creating new prejudices founded in the molecular biology of human genetic variation.

Researchers investigating human variation effectively heighten the importance of genetic differences, even as they discount misconceptions about racial distinctions and emphasize the unity of all humans.

"The core question," said biomedical ethicist Mildred Cho at Stanford University, "is why is it we keep looking at genes and variations and organizing them into categories?"

Everyone has a reason.

Many researchers are eagerly sifting the raw ore of human genetic variation for DNA sequences that could lead to more effective medications or better treatments for disease. Others hope to use racial profiling as a diagnostic tool to identify those at greater risk of chronic diseases or adverse reactions to prescription drugs. Some just want to better understand human evolution.

On the whole, there is less genetic difference between human beings than between any two members of almost any other mammalian species, scientists said.

But advances in biotechnology have made those minor differences more important and more lucrative than ever. Last month, researchers launched a $100-million effort to map as many of those variations as possible.

The latest research, published today in Science, highlights a paradox of modern genetics, several scientists said.

"Everybody is the same; everybody is different," said Mary-Claire King, an expert in human genetics at the University of Washington in Seattle. "That is the paradox."

In the new study, researchers pored through the biochemical scrawl of the human genome like a travel diary that records the migrations, matings, illnesses, wars of conquest and mishaps shaping the human species.

Mapping the patterns of human variation, the research team led by USC geneticist Noah Rosenberg analyzed DNA samples from 1,056 people from 52 populations in five major geographic regions of the world: Africa, Eurasia (Europe, the Middle East, and Central and South Asia), East Asia, Oceania and the Americas.

Looking for patterns of human ancestry, the research team used distinctive segments of DNA called micro-satellites that are passed down from generation to generation. They do not code for genes. Their only known virtue is that they are among the most variable parts of human heredity.

In all, they analyzed 377 of these DNA markers.

Researchers have identified between four and 32 variations of each DNA marker. Some variations occur more frequently in certain parts of the world.

With a powerful statistical computer program, the researchers could accurately identify the ancestral continent of virtually every person based solely on his or her DNA. They could arbitrarily change the number of genetic groupings by altering the parameters of the computer program.

"We found that there were several major genetic clusters, which correspond largely to the major geographic regions," Rosenberg said. The research drew on efforts by scientists in France, Russia, and in U.S. at Stanford, the University of Chicago and Yale University.

The genetic groups they identified represented the slow process of adaptation and mutation as humanity spread across the world.

"This analysis tells you the great swath of human history, more elegantly, more completely and better than anyone has done it before," King said.

Such studies are stirring a fierce debate over the proper place of racial categories in medical research and public health.

Last year, the New England Journal of Medicine concluded in an editorial that " 'race' is biologically meaningless."

Nature Genetics warned of the "confusion and potential harmful effects of using 'race' as a variable in medical research."

But as a matter of public health, racial labels have traditionally been a measure of life and death.

Infant mortality has been about twice as high for blacks as for whites since 1950; Native Americans have substantially higher rates of death from unintentional injuries than any other group; and, compared with whites, native Hawaiians are more likely to die from heart disease, cancer and diabetes.

Blacks, Ashkenazi Jews, Chinese, Asian Indians and other groups in America can have different -- and potentially dangerous -- reactions to common medications such as heart drugs, tranquilizers and painkillers. Many common diseases have a genetic component.

In the absence of any reliable scientific definition of race, such categories can easily be misleading or misconstrued.

Comparing birth and death certificates for 120,000 babies who died in the early 1980s, for example, the U.S. Centers for Disease Control and Prevention found that many had been identified as one race at birth and another at death.

Researchers also may find an inherited, racial linkage where none exists, or overlook the medical effects of what people eat, where they live and how they are treated. Even when groups are closely related genetically, ethnic variations in diet can radically alter reactions to medications and other important medical characteristics, research shows.

"That is the danger," said Arno Motulsky, a geneticist at the University of Washington.

When it comes to the genetics of human variation, humans differ most importantly in ways largely invisible to the eye, researchers discovered.

The physical characteristics so often taken as hallmarks of racial origin -- skin color, hair texture or the shape of nose and lips -- are poor clues to ancestry, according to a new DNA analysis of the people of Brazil, published this week in the Proceedings of the National Academy of Sciences.

Working in one of the world's most multicultural societies, where Africans, American Indians and European settlers have intermarried for centuries, the Brazilian scientists sought to determine how much physical appearance could predict a person's genetic ancestry. They conducted DNA analysis of several hundred Brazilian men who categorized themselves by skin color and other features.

On average, Brazilians with white skin owed almost a third of their genes to African ancestry, while those with dark skin could trace almost half their genes to European origin, the researchers determined.

"Our study makes clear the hazards of equating color or race with geographical ancestry and using interchangeable terms such as white, Caucasian and European on one hand, and black, Negro or African on the other," said geneticist Sergio Pena at the Federal University of Minas Gerais in Brazil.

"It does not make any sense to use someone's physical appearance to guess about their predisposition to hypertension or drug metabolism or anything," Pena said.

As immigration and intermarriage makes the world a more cosmopolitan place, Cho said: "in the future we will all be like Brazil."