Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 16
Evaluating Human Genetic Diversity 2 Scientific and Medical Value of Research on Human Genetic Variation No person or group can be the measure of humanity. All people—in every part of the world in all our cultural, linguistic, and biologic diversity—are equally human. Comparisons of the genetic variability between and within populations can do much to increase our understanding of the human species. As a consequence, the application of modern molecular techniques in the analysis of genetic variability of specimens collected from human populations throughout the world can contribute substantially to our knowledge of ourselves. The results of such worldwide studies should greatly enlarge our knowledge and understanding of the diversity of the human species. Most research on human genetic variability has been done independently by different investigators. Their work is rarely comparable, however, because different sampling procedures are used. The studies also tend not to have broad geographic coverage, because investigator interests, ease of access, and financial resources drive the selection of the geographic region covered. There is need for better coordination, comprehensive geographic coverage, standardization of sampling and data-processing, and better compliance with basic standards. To address those needs, research on human genetic variability will require international cooperation and collaboration in the acquisition, examination, and sharing of the resources and knowledge essential to our collective understanding and characterization of the human species. For example, understanding human genetic variability can greatly improve our grasp of human history and migration, of the evolution of the human gene pool, and of the basic mechanisms of genetic evolution. It can also provide data relevant to biomedical application. The potential effects of research in human genetic variability are addressed in this chapter.
OCR for page 17
Evaluating Human Genetic Diversity HUMAN HISTORY AND EVOLUTION The technologic advances of the last decade in molecular genetics—which measures variability of the genetic material directly, not indirectly through phenotypes—promise to revolutionize the study of human genetic variation. Many long-standing questions in anthropologic genetics can now be examined with this powerful new set of scientific tools. The field has long focused on the bipartite nature of human genetic variability through studies of variation within and between populations. By examining those 2 levels, researchers have tried to explain the evolutionary processes that have given rise to the current patterns of human variation. For example, the study of the diverse population structures of a large number of human groups has clearly shown that random genetic drift—the accidents of chance that result from the small size of local groups—and the flow of genes from one population to another through migration or mate exchange are responsible for much of the patterning within human populations (Crawford 1976; Harpending and Jenkins 1975; Neel 1978; Williams-Blangero 1989). Much has been learned from such small-scale evolutionary studies, but the usefulness of the genetic data on which they are based is limited by the small number of marker loci that were examined. Recent advances in molecular and statistical techniques might substantially improve our knowledge of the evolutionary dynamics of human populations. For instance, pairwise comparisons of sequence differences among people within a population have shown that such differences retain information regarding the demographic history of the population (Rogers and Harpending 1992; Slatkin and Hudson 1991) and can be used to make inferences regarding the forces that shaped the pattern of intragroup genetic variation. Such approaches provide a specificity and accuracy of information that has not previously been possible. Intensive studies of intragroup variation have often been successfully undertaken by single investigators. However, past efforts to examine human genetic variability between populations on a large geographic scale have been problematic because they require many investigators and worldwide collaboration. The examination of worldwide human genetic variation focuses on between-population comparisons aimed at inferring population history or population migration patterns (Cavalli-Sforza 1994; Sokal and others 1996, 1997). For example, Cavalli-Sforza's broad-scale examination of the variation between human groups has found a major concordance between language similarities and genetic relationships. Similarly, his notions, although controversial, about the spread of agriculture by migration, rather than through the spread of customs and techniques, further illustrate the kinds of questions that could be studied in much-greater detail. In general, such endeavors have been severely compromised by the paucity of comparative information—an unfortunate byproduct of the lack of coordinated collection efforts and the inability of multiple investigators to genotype large
OCR for page 18
Evaluating Human Genetic Diversity numbers of samples for large numbers of genetic loci. Once information becomes available on a large and consistent set of genetic markers for many human populations distributed around the world, much more systematic investigations of the determinants of human variation will be possible. For example, questions regarding the timing of occurrence, the number, and the size of long-distance migration from Siberia to the Americas and the effects of such migration on the genetic variation within and between Native American populations will be easier to answer. Moreover, despite the relatively short time that has elapsed since those migrations, cultural differences between Mayans and US-situated plains Indians are striking, and it would be of interest to know whether these differences are reflected in genetic differences. Similarly, phylogenetic questions regarding events that occurred over a much-longer period could also be more-precisely answered. For example, considerable controversy exists about the contributions of early Homo erectus populations outside Africa to the genetic makeup of modern human populations. One theory is that after Homo erectus spread over most of Eurasia and Africa, geographic races developed and evolved together, with genetic interchange occurring among them as they evolved into modern humans (Wolpoff 1989). Another theory holds that anatomically modern humans emerged in Africa and spread over the Old World, replacing earlier populations without having substantial genetic interchange with them (Cann and others 1987; Stringer 1988; Stringer and Andrews 1988). Evidence from mitochondrial DNA and other genetic material used in support of the latter position is now known to be based on samples too small to test the hypothesis rigorously (Ayala 1995; Spuhler 1988; Templeton 1993). More-sophisticated sampling of more-diverse populations should help to resolve the issue. The study of the determinants of human biologic variation has often focused on readily observable phenotypic characteristics, such as height, weight, facial dimensions, skin color, and body composition. Those traits have complex underlying genetic components, and we know little about the evolutionary processes that have led to the extensive variability in them among current human populations. Attempts to relate the regularities of observed spatial variation in such phenotypes with environmental factors have been problematic. This stems largely from the difficulty in taking into proper account the correlation between the measurements, such as height and weight or weight and blood pressure, used to characterize variation. A common goal of such studies is to test whether observed variation requires some nonrandom force, such as natural selection, so researchers must be able to predict what would be observed if variation were random. Random variation is a quantitatively definable null hypothesis from which to infer selection. With a large set of independent, neutral genetic markers, scientists can predict the effect of random variation from estimates of kinship between the populations on which phenotypic information exists. Such estimates serve as records of a population's evolutionary history. By controlling for that history, researchers can determine whether the pattern and magnitude of phenotypic variation is consistent with random evolutionary forces or whether a more-
OCR for page 19
Evaluating Human Genetic Diversity complex explanation is required. The availability of environmental measures will make possible powerful tests of the association between phenotype and environment in which the statistically confounding effects of shared genetic factors are minimized. BASIC MECHANISMS OF GENOME EVOLUTION Surveys of genetic variation within and among human populations can be used to address a number of basic questions in evolutionary genetics. In very few instances (in any organism) do we know the relative importance of various evolutionary forces in determining the amount and character of genetic variation within and among populations. Examples of those evolutionary forces include mutation, types of natural selection that can vary temporally and spatially, genetic drift, recombination, and migration. In particular, we are largely ignorant of the answer to a question of fundamental importance to evolutionary biologists: How much genetic variation is directly subject to natural selection—that is, how much genetic variation has direct phenotypic or functional effects that influence the survival and reproduction of individuals? Studies of genetic variation have the potential to answer such questions. Such studies undertaken on nonhuman organisms have shown that one of the most powerful means of demonstrating and characterizing the effects of natural selection is to contrast the patterns of variation in different regions of a chromosome, at different loci, and at different sites within a gene (Aguade and Langley 1994; Akashi 1995; Begun and Aquadro 1992; Eanes and others 1996; Hudson and others 1987; Langley and others 1993; McDonald and Kreitman 1991; Sawyer and others 1987; Stephan and Mitchell 1992). Kreitman and Akashi (1995) provide an excellent review of many of the methods and their applications. Unusually high variation at a locus can result from a form of natural selection called balancing selection (Hudson and Kaplan 1988; Kreitman and Aguade 1986) that maintains polymorphisms at such loci as the HLA. Unusually low variation at a locus can be caused by a recent fixation of a mutation that is favored by selection (Kaplan and others 1989; Maynard-Smith and Haigh 1974). If a particular kind of variation (such as amino-acid changing variation) has fewer low-frequency variants than other kinds of variation, that can indicate that natural selection is acting against the low-frequency variants (Sawyer and others 1987). Contrasting the amount of variation within species with the amount of divergence between species for different classes of variation can also indicate some forms of selection (Akashi 1995; Hartl and others 1994; Sawyer and Hartl 1992). Methods that contrast different types of variation are powerful because they control for demographic and phylogenetic factors, which are necessarily the same for different loci. Studies of human genetic variation have the potential to be particularly informative with respect to those evolutionary issues. If such studies were done on humans, they could also elucidate genomic features peculiar to our species. Many studies suggest that expansions in triplet
OCR for page 20
Evaluating Human Genetic Diversity repeats within some genes that underlie neurologic disorders (fragile X, myotonic dystrophy, Huntington disease, and so on) might be found only in human populations. Whether that type of genomic expansion is due to the unique history of the human species or to unique mutational or other genomic mechanisms is unknown. Recent studies have shown that the triplet repeats are polymorphic in all human populations tested, but populations with higher numbers of repeats (alleles) have higher incidences of the disorders. The incidence of disease can depend strongly on the biologic origins of a human population. Medical examinations of people around the world can also bring to light new mechanisms that can be difficult to identify in other organisms. For example, the insertion of putative genetic elements known as retrotranspositions (specifically, the short interspersed repeated DNA sequence Alu and the long interspersed repeated sequence LINE1) into actively transcribed genes, such as the antihemophilic factor VIII and neurofibrin, observed as new mutations in patients, has revealed that retrotranspositions are a feature of the human genome. Because some Alu elements are polymorphic, they are useful genetic markers for the study of recent human population differentiation (Batzer and others 1996). Another recently discovered genomic feature is that some human chromosomal telomeric segments–segments at the extremities of a chromosome–show greater similarity, in sequence, to telomeric segments on nonhomologous than on homologous chromosomes. Whether that feature occurs on some or all chromosomes or in some but not other human populations is unknown. Worldwide studies of genetic variation should contribute substantially to answering this question. BIOMEDICAL APPLICATIONS The kinds of anthropologic, evolutionary, and migratory and other historic information to be derived from worldwide studies would involve intergroup comparisons based on sampling procedures that do not identify individual persons. They do not require intensive studies of variation within human groups, especially those using family-level differences and extensive phenotypic information, including medical, environmental, and occupational data on each individual. Epidemiologic information can be derived directly from survey data that does not identify individuals. For example, noninsulin dependent diabetes mellitus appears to be emerging as a global public health problem yet there is surprisingly little information on its frequency in many of the world's populations. The specimens from which DNA is to be derived could readily be tested for such parameters as glycosylated hemoglobin. These tests are inexpensive and do not require a high degree of technical skill, and would provide useful, albeit crude, estimates of the frequency of noninsulin dependent diabetes mellitus in all of the populations sampled. Such information could be useful to public health authorities responsible for the health of such populations and could provide the basis for more comprehensive studies of the disease's etiology.
OCR for page 21
Evaluating Human Genetic Diversity Another example of a biomedical issue to whose understanding a large set of clearly defined population samples could contribute is linkage disequilibrium. Linkage disequilibrium refers to the statistical nonindependence of 2 gene loci. The strength of such disequilibrium is a function of population structure and history and of the physical distance between the 2 loci. Disequilibrium decays every generation, and in large populations it is found only between very tightly linked loci. However, depending on the antiquity of genetically important events in a population's history, disequilibrium can be observed in some populations to span much-larger chromosomal regions. For example, in recently admixed populations, disequilibrium can be observed over areas as large as 5-10 cm, or some 2-5% of the estimated linkage length of the average human chromosome (Morton 1991). But in small isolated populations, linkage disequilibrium might be extensive. Disequilibrium can be exploited to help map loci that are involved in disease or other complex phenotypes. Ultimately, association studies that test whether a genetic marker is statistically correlated with a disease-related phenotype can be conducted (Risch and Merikangas 1996). Because random events and the peculiarities of population structure are so important in the generation and maintenance of linkage disequilibrium, it will be useful to screen large numbers of populations for the presence and magnitude of linkage disequilibrium. This will entail selecting some loci to reflect various linkage distances, including some that are very tightly linked. Populations with more-extensive regions of linkage disequilibrium can then be identified and studied more completely for the complex inherited diseases that they experience. As the foregoing suggests, most biomedical applications require more-specifically targeted studies involving sample sizes and sampling procedures under sampling levels 4 and 5, as described in chapter 3. Potential biomedical applications employing these more complex sampling strategies might include Specific gene-disease relationships (variations both within and between groups in disease susceptibility or resistance according to genotype), natural selection, and genetic drift; Gene-environment interactions—phenotypic variation and pharmacogenetic and toxicologic implications; Linkage disequilibrium associated with relatively common disorders, such as diabetes, asthma, and bipolar disease; Complex genetic traits associated with specific aberrant behavioral characteristics, such as aggressivity and alcoholism; Genetic factors linked to multivariate non-disease-related characteristics, such as height, intelligence, and aging; Genotype associations—predisposition to particular cancers, autoimmune disease, and other disorders;
OCR for page 22
Evaluating Human Genetic Diversity HLA and other immune-response-related genotypes in diverse populations—their implications for transplantation, vaccine development, and related therapies. Unquestionably, detailed information concerning genetic diversity in widespread human populations theoretically might be applied, in various ways, to any or all of the above biomedical issues. The specifics of how such applications could be made more scientifically rigorous pose some dilemmas in the design and conduct of a survey of genetic variation. It is possible that a few of the above applications might be addressed directly, with narrowly confined information from each individual sampled and with small to moderate samples, but most biomedical investigations will require considerably larger samples and substantially more information on each person sampled than the committee deems practical on a global scale. Accordingly, population-based surveys of genetic variability might best be viewed not as a mechanism to answer such critical biomedical and population-genetics questions directly, but rather as a resource from which, once established, control samples could be derived for specific biomedical projects. Moreover, the availability of a survey database and sample repository could serve as points of departure for pilot studies with biomedical applications from which more-extensive and more-comprehensive investigations might be derived. Population-based data on genomic variation could best serve as the foundation for more-expansive, conclusive research studies in the future. Such studies might be directed to subpopulations identified through pilot studies derived from the initial population-based genomic-variability survey samples. These derivative investigations would be carried out more with the populations under study and would therefore be less likely to create social and ethical dilemmas for both the investigators and the participants involved. In summary, although biomedical applications are clearly important ultimate goals of population-based surveys of genomic variability, it appears more realistic at this stage of planning for biomedical investigations to be viewed as secondary targets. The committee appreciates that this view will be controversial and that it could have some negative consequences, such as a lesser willingness to participate in a study that has no immediate health benefits for potential subjects. CONCLUSIONS A comprehensive survey of human genetic variability both between and within populations could map such variability and place it in social and environmental context. Careful variability sampling in conjunction with the Human Genome Project could contribute fundamentally to a new era of modern molecular medicine and transform scientific understanding of human evolution and the course of human prehistory.
Representative terms from entire chapter: