Scientific American - Do races exist?

If the definition says that races are genetic groups that differ from each other, the answer to the question is no. But researchers can use genetic information to group people into groups in a medically meaningful distribution

Michael J. Bemshad and Steve A. Olson

The article is taken from issue 10 of the Israeli edition of the journal "Scientific American", published by Ort

Look around you in the streets of any large city and you will see examples of the external difference between the members of the human race: skin tones from milky whitish to dark brown, hair textures from thin and straight as a bar to thick and curly. People often use these characteristics - along with geographic origin and common culture - to associate themselves and others with "races". How valid is the concept of race from a biological point of view? Do a person's physical characteristics say anything about their genetic makeup, other than the fact that some person has a gene that dictates blue eyes or curly hair?
The problem stems in part from the fact that the definition of a person's racial affiliation is different in different regions of the world. A person defined as "black" in the US, for example, may be considered "white" in Brazil and "colored" in South Africa (where this definition differs from both "white" and "black").
However, sometimes the conventional definitions of race are effective when dividing the population into genetic groups according to certain diseases. Sickle cell anemia, for example, often occurs in people of Mediterranean or African descent, while cystic fibrosis (Leift) disease is more common among people of European descent. Furthermore, some studies (which are still controversial) suggest that African Americans respond less well to certain heart disease medications than other American groups.

In recent years, scientists have been collecting information on the genetic makeup of different populations on Earth. They did this to decipher the link between the origin of humans and different disease patterns. This information now provides answers to infuriating questions that carry a lot of emotional charge: Can genetic information be used to differentiate between groups of people with a common heritage or to assign a person to a particular group? Do such groups fit today's accepted descriptions of race? And the more useful question: is dividing people according to racial definitions or according to genetic similarity useful when coming to test the reaction of group members to disease or drug treatment.
We would usually answer the first question "yes", the second "no" and the third a qualified "yes". Our answers are based on some generalizations about race and genetics. Some groups are indeed different from each other genetically, but the way we distinguish between the groups depends on the genes we test: simply put, you may belong to one group based on the genes for your skin tone and to another group based on another characteristic. Many studies have shown that about ninety percent of the genetic variation between humans appears within the population of each continent, while only ten percent of it differentiates the population of one continent from another. In other words, the difference between two people belonging to different population groups is greater but slightly, on average, than the difference between two people belonging to the same population group. Human populations are very similar, yet often distinguishable.

Classification of humans
The first step on the way to identifying the link between the social definitions of race and the genetic heritage is a reliable division, according to the origin of different groups of people. Humans whose anatomy is modern have migrated over the last 100,000 years from Africa to the rest of the world. The number of our kind has also increased dramatically. This expansion has left its mark on our DNA.
In determining the degree of genetic kinship between different groups, geneticists rely on small changes in DNA (a phenomenon called polymorphism). These changes occur in the sequence of the base pairs - the building blocks of DNA. Most of these changes (polymorphs) do not appear within the genes, the sections of DNA that contain the information code for building proteins (proteins make up most of our bodies and carry out the chemical reactions necessary for life). Therefore, the common polymorphs are neutral, meaning they do not directly affect any traits. However, there are polymorphs that appear within the genes and then they contribute to differences in traits and genetic diseases.
When the scientists sequenced the human genome (the complete sequence of the DNA found in the cell nuclei), they identified millions of polymorphs. The distribution of these polymorphs in different populations reflects the history of these populations and the influence of natural selection. The ideal polymorph used to distinguish between human groups would be the one that appears in all members of the group and does not appear in any other group. However, most human groups separated from each other too late and mixed with each other too much, so such differences do not exist.
And yet polymorphs that appear with different frequencies around the world can allow a rough division of humans into groups. One useful type is polymorphs called "Alu". These are short DNA segments that are similar to each other. These polymorphs are sometimes duplicated, and the copies randomly integrate into new locations on the original chromosome or another chromosome. They usually integrate in a location that does not affect the function of the neighboring genes. Each integration is a unique event. The integrated Alu sequence can remain in place forever and be passed on to offspring. Hence, if two humans have the same Alu polymorph at the same point in the genome, it can be concluded that they are the offspring of a common parent who gave them that particular DNA segment.
One of us (at Meshed) is conducting joint research with scientists from the University of Utah, Lynn B. Jordan, Stephen Wedding and Scott Watkins, and Mark A. Betzer from the University of Louisiana. We tested 100 Alu polymorphisms in 565 sub-Saharan Africans, from Asia and Europe. We first tested the presence or absence of these 100 genetic changes in each of them. Then we removed all identification labels from the information (such as geographic origin and ethnic affiliation), and sorted the subjects according to the genetic results.
This sorting led to the creation of four groups. We put the labels back in place to see if each individual's grouping matched the predefined definitions of race or ethnic group. We came to know that in two of the groups there were almost exclusively people from groups south of the Sahara. One of these groups was almost entirely pygmies from the Mbuti tribe (Mbuti Pygmies). The other two groups consisted of Europeans and East Asians respectively. We found that 60 polymorphs were required to assign an individual to its continent of origin with 90% accuracy. To get close to 100% accuracy, we needed 100 Alu polymorphs.
Other studies have produced similar results. Noah A. Rosenberg and Jonathan K. Pritchard, geneticists who previously worked in Marcus W. Feldman's laboratory at Stanford University, studied about 375 polymorphs, called "short tandem repeats". They examined more than 1000 people belonging to 52 ethnic groups from Africa, Asia, Europe and North and South America. The researchers observed the frequencies in which these polymorphs appear and were able to distinguish five distinct groups of humans whose ancestors were separated by oceans, deserts, and mountains. The groups are: Africans south of the Sahara, Europeans, Asians west of the Himalayas, East Asians, Papuans and Melanesians and the original natives of America. They were also able to identify subgroups in each region and these generally corresponded to the ethnic group that the person claimed to belong to.
The results of these studies show that genetic analysis can distinguish between groups of people according to their geographic origin. However, caution must be exercised here. The groups that were easiest to distinguish between were those that were far apart geographically. These examples demonstrate the maximum genetic variation. When Meshed and his colleagues used the 100 Alu polymorphism method to classify a sample of people from South India into separate groups, it became clear that the commonality between the Indians and Europeans or other Asians was greater than the commonality between the group members. In other words, India was subject to many genetic influences from Europe and Asia and therefore people from the subcontinent did not cluster into a unique cluster. We therefore concluded that we may need hundreds, and perhaps even thousands of polymorphs to distinguish between groups whose ancestors mixed throughout history with many other populations.

the human race
If so, it is possible to roughly sort humans using genetic information. Do the accepted concepts about race correspond to the genetic differences underlying the differences between populations? In some cases yes, but in other cases no. Skin color or facial features, for example - traits affected by natural selection - are usually used to separate humans into races. But groups with similar physical characteristics, as a result of natural selection, may be very different genetically. Sub-Saharan Africans and Australian Aborigines have similar skin color (due to adaptation to strong sun) but genetically they are quite different.
Conversely, two genetically similar groups may be exposed to different selective forces. In this case, natural selection can highlight the differences between the groups, so that outwardly they will appear more different than the basic difference between them. Traits such as skin color have been strongly influenced by natural selection, so they do not necessarily reflect the population processes that shaped the distribution in the world of polymorphs such as Alu or short tandem sequences. Therefore, traits or polymorphs affected by natural selection do not predict group membership well. They may imply genetic kinship even in cases where its existence is in doubt. Another example of the difficulty in classifying people is the US population. The ancestors of most Americans who consider themselves of African descent came not long ago from West Africa. People of such ancestry have polymorphs whose frequencies differ from those of Europeans, Asians, and Native Americans. However, the proportion of genetic variation that African Americans share with West Africans is not uniform. Throughout the generations the Africans who came to America mingled to a great extent with Africans from other regions and even with people from other continents.
Mark D. Schreiber from the University of Pennsylvania and Rick A. Kittles from Howard University have in recent years defined a group of polymorphs by means of which they estimate the proportion of genes that originate from some region on the continent. They found that the contribution of West Africans to the genes of an African-American person averages 80%, although it can range from 20% to 100%. Mixing of ancestry groups is also evident in many people who believe that their ancestors are only Europeans. Schreiber's analyzes show that for about 30% of Americans who define themselves as "white" it was found that less than 90% of their lineage members are indeed of European origin. Thus, a person's reporting of their ancestry does not predict well the genetic makeup of many Americans. Hence, the accepted concepts about race do not always reflect the genetic background of humans.

Belonging has rights
Understanding the relationship between race and genetic variation has important practical implications. Some of the polymorphs whose frequencies vary from group to group have certain health effects. The mutations responsible for sickle cell disease and some cases of cystic fibrosis, for example, result from genetic changes. Attention to the frequencies of these genetic changes has risen, probably because they contribute to protection against diseases common in West Africa and Europe respectively. People who have inherited only one copy of the gene with the polymorph that causes sickle cell anemia have partial resistance to malaria. People with one copy of the gene whose damage causes cystic fibrosis are less susceptible to dehydration in cholera. The symptoms of both diseases appear only in those people who are lucky enough to inherit both copies of the mutations.
Genetic differences also play a role in personal susceptibility to the scourge of our generation, AIDS. Some people are missing a small section in their genome in both copies of the gene that codes for a receptor that appears on the surface of cells, called chemokine receptor 5 (CCR5). As a result, their body does not produce the receptor. Most strains of HIV-1, the virus that causes AIDS, bind to 5CCR receptors, and thus they enter cells. People who do not have this receptor on the surface of their cells will therefore be resistant to contracting AIDS. The polymorphism in the 5CCR gene appears almost exclusively in population groups from northeastern Europe.
Some of the 5CCR polymorphs do not prevent HIV infection but affect the rate at which HIV-1 infection leads to AIDS and death. Some of these polymorphs have similar effects on different populations, and others have an effect on the speed at which the disease progresses only in certain groups. One polymorph is associated, for example, with a slower development of the disease in Americans of European descent, but with an accelerated development in Americans of African descent. Scientists can study these population-dependent effects, and use the knowledge for healing purposes. But they can only do this if they know how to sort people into groups.
In these examples, and others like them, we see that polymorphisms have a relatively large effect in certain diseases. If genetic scanning were a cheap and efficient process, it would be possible to scan all humans for the presence of all disease-related mutations. However, genetic testing has always been a very expensive test. The more significant concern that genetic scanning raises is related to privacy and consent: some people may refuse to know the genetic factors that may increase the risk of developing a certain disease. The personal report of origin will therefore continue to be used as an effective diagnostic tool by doctors, until these issues are resolved.
The matter of the origin of humans can also be significant regarding some diseases that are common among a certain population. Most common diseases, such as hypertension and diabetes, are the cumulative result of polymorphisms in several genes, each of which contributes its small effect. Recently some studies have suggested that polymorphs, which have a certain effect in one population group may have a different effect in another group. This complexity makes it difficult to use known polymorphs to direct medical treatment. Until more studies are conducted regarding the genetic contribution and the contribution of the environment to complex diseases, doctors will have to rely on information about the patient's origin in order to better treat his disease.

race and medicine
However, in recent years the controversy regarding the importance of belonging to a population group, and the connection between this belonging to medical treatment, has greatly increased. In January 2003, the US Food and Drug Administration (FDA) published guidelines recommending that information about race and ethnicity be collected in every drug trial. Some researchers have argued that it is better to avoid referring to ethnicity in medical studies because the differences between the population groups are very small and in contrast the misuse of dividing people into races throughout history has been extremely extreme. They vehemently state that the FDA should abandon its recommendations and instead ask researchers involved in drug trials to collect genetic information on each and every individual. Others say that only using group affiliation, which includes commonly accepted definitions of race based on skin color, will lead to an understanding of how genetic and environmental differences between groups contribute to the development of disease. This controversy will only be resolved through further research examining the validity of race as a scientific factor.
In the March 20, 2003 issue of the New England Journal of Medicine, a series of articles debated the medical implications of race. The authors of one paper, Richard S. Cooper of the Loyola University Stritch School of Medicine, J.S. Kaufman of the University of North Carolina at Chapel Hill and Rick Ward of the University of Oxford, argue that race is not an appropriate criterion when doctors choose a particular drug. for a patient They mention two findings of racial differences that are now considered questionable: one, that a certain combination of vasodilator drugs is more effective in treating heart failure in patients of African descent and the other, that certain angiotensin converting enzyme inhibitors (ACE inhibitors) are not effective in treating these patients In contrast, a group of researchers led by Neil Risch of Stanford University claimed, in another article, that racial or ethnic groups can differ from each other genetically and that the differences between them may have medical importance. They cited a study that showed that the frequency of complications resulting from type 2 diabetes varied accordingly. for race, even after offsetting the effect of differences in education and income.
The intensity of these debates reflects both scientific and social factors. In many biomedical studies, belonging to population groups is not strictly defined, relying instead on assumptions based on racial types. The controversy surrounding the importance of belonging to a population group demonstrates how much the perception of race is shaped by social and political views.

Belonging to a population group, defined geographically or culturally, has a correlation with genetic traits related to health. In such cases, information about the person's belonging to a certain group may be important to the doctor. Different human groups live in different environments and have different experiences, which can have a health impact. In these cases, belonging to a group may also reflect non-genetic factors that affect health.
The research findings are exciting in themselves, even without regard to the medical implications of the breed's genetics. For hundreds of years people have wondered where the different human groups came from and how they are related to each other. Hypotheses have been raised about the reasons for the different physical appearance of human populations and many have wondered if the differences are deeper than the skin layer. New genetic information and new research methods allow us, finally, to approach these questions. The result will be a deeper understanding of our biological nature and our human connection.

Human genetic diversity
Researchers sometimes use short pieces of DNA called Alu polymorphisms to determine whether different populations are related. These polymorphs have no known function, yet they are randomly copied and integrated throughout the genome. After Alu type polymorphs are fixed in their place, they are not removed and removed from the genome. Therefore, they can be used as a criterion for the genetic kinship between two people, or between two populations. For example, the sample revealed that the Alu polymorph on chromosome 1 appears in about 95% of the Africans of sub-Saharan origin that were tested, in 75% of the Europeans and North Africans and in 60% of the Asians. Another Alu polymorph, on chromosome 7, is found in about 5% of sub-Saharan Africans, 50% of Europeans and North Africans, and 50% of Asians. And there are people in whom both polymorphs appear. No single polymorph can, by itself, distinguish all members of one population group from all members of another population group. However, by examining hundreds of polymorphs, scientists can group people from different places based on their genetic profile.