The 0.1% Portrait of Human History

Stored in the human genome, perhaps, is the record of human evolution and existence on this planet. Many say, however, that this history and the benefits it may unfold for human health cannot be found in the single, essentially complete human sequence--99.9% similar to any other human sequence. It's the 0.1% difference that should tell the tale--not only of migration, war, technological achievement, and conquest--but also of the differences that confer susceptibility to complex, multigenic diseases. In piecing together the travails of diverse human populations, geneticists use a growing number of tools and technologies. But, what would arguably be the most valuable resource to population geneticists still does not exist. First envisioned a decade ago, a project to sample diverse populations on a global scale faltered in its planning stages amid controversy and criticism that its proponents were not sensitive to the concerns of groups it proposed to study. Its legacy lives on, however, as proponents work to piece together the remnants of a once grand scheme, and as funding agency officials, namely National Human Genome Research Institute head, Francis Collins, confront the need and the sensitivity of understanding genomic diversity across ethnically diverse populations. Such sensitivities surround the disputed concept of race and the idea of placing it in a genetic context, but health disparities between different populations suggest that a better understanding of the differences may contribute to health research that will benefit everybody. Proxy identifiers of human populations, such as skin color or language, can't replace a genetic understanding of world population structure, the distribution of human variation. "It really needs to be understood properly if the benefits of genetic research are going to apply to everyone," says Kenneth Weiss at Pennsylvania State University. STRUCTURED LEARNING Key findings have recently demonstrated that it is possible to determine population structure by investigating neutral variation--the random polymorphisms that don't appear to affect phenotype. "Although the differences between populations are small, if you look at enough loci, you begin to accumulate enough differences so that individuals can be assigned," says Lynn B. Jorde, Eccles Institute of Human Genetics, University of Utah. His group recently demonstrated that it was possible to correctly assign roughly 560 individuals to their continents of origin using about 100 markers.1 The loci they used included mostly Alu insertion polymorphisms and some microsatellite data. The resolving power of the markers broke down, however, when they tried to define groups at a local level. They focused on individuals in South India, where Jorde's group has conducted intensive study and where numerous migrations from Eurasia had stirred the gene pool. "We were able to categorize 87% of those individuals into a different cluster, but it illustrates the fact that with some populations, classification is not going to be as easy," Jorde says. As migrations can shuffle the mixture, deeper anthropological questions about those movements could be answered with a thorough investigation of variation. Noah Rosenberg, a postdoc at the University of Southern California, Los Angeles, worked with Stanford's Marcus Feldman and others to compare variation at 377 microsatellite markers across 1056 individuals from 52 populations worldwide.2 This glimpse of one of the most comprehensive sample sets around underscored the idea that humans are largely alike. Genetic variation between groups accounts for only 3% to 5% of all variation, the researchers estimate. The rest presumably existed as populations first expanded out of Africa. Using the same markers, Rosenberg says that he hopes to build models of recent history that include migration. New tools are being applied. Sarah Tishkoff of the University of Maryland studies multiple types of variation in numerous African populations. In addition to studying microsatellites (short tandem repeat polymorphisms), she's begun to investigate SNPs (single nucleotide polymorphisms), which have some advantages over microsatellites. "They're not as variable or have as high a mutation rate as the microsatellites," says Tishkoff. They also enable more throughput. Tishkoff has been working with Penn State's Mark Shriver to type upwards of 10,000 SNPs per individual. The cost is limiting, she says: They can only hope to type perhaps three populations using this technology. A MISSING LINK Many claim a big limitation is the resource that doesn't yet exist. A decade ago, Stanford population geneticist L. Luca Cavalli-Sforza and others proposed the Human Genome Diversity Project (HDGP), which would sample some 10,000 individuals. Researchers across the globe would collect samples from 200 to 400 distinct population groups and create a central repository of cell lines for population geneticists and anthropologists of future generations. In the planning stages, organizers proposed the inclusion of small indigenous groups. Says Weiss, who was involved, "When Science or other news media called, we were vulnerable and we sent pictures of these tribal groups." Mentioning some by name, they raised the ire of nongovernmental organizations representing indigenous people. Stanford Law professor Hank Greely says, "I think we were a very useful demon for a number of groups who were able to say, 'Look, now they're trying to steal your genes.'" Inevitably, the project wasn't funded. Collins says that their failure to first consult with the populations being considered was problematic. "The resistance to the project had reached such a high level that I think to start funding it anyway would have only inspired a greater sense that NIH was not concerned about community criticisms." Cavalli-Sforza and others persevered, however, amassing more than 1000 cell lines from the collections of several population geneticists to provide, essentially free to researchers, an infinite amount of DNA from 51 diverse populations.3 Stored in Paris at the Centre d'Etude du Polymorphisme Humain (CEPH), it was the resource Rosenberg and others used in their broad study of microsatellites (their data suggested the presence of a 52nd population). The diversity cell line panel is incomplete, lacking full representation of some groups and missing others entirely: India and Australia are not represented at all. Yet, it's garnered some interest. Howard Cann, who manages the cell line panel in Paris, says he's sent DNA to 46 research groups to date. Cavalli-Sforza says he's hopeful for the future: "It has taken 10 years to make that little beginning. ... Essentially, if we want to use the genome, we have to persevere and extend this enormously." If the resource is expanded, he says, it could benefit human health. Public funding agencies have invested in understanding human variation. The $100 million public-private effort, known as the International HapMap Project, has as its primary goal the characterization of common blocks of SNP variation. These haplotype roadmaps will allow researchers to predict a large portion of a person's inherited genotype based on a few SNPs, and should direct faster discovery of genes that contribute to disease-susceptibility and drug-response. To some, however, the initial focus on just a few world populations raises questions as to the project's wider applicability. "Doing Chinese and Japanese, mixed Europeans, and one West African population is just not going to give a reasonable picture," says geneticist Kenneth Kidd of Yale University School of Medicine. "The ability to extrapolate from that to another unstudied population is just going to be very poor." A DIVERSE VIEW: Geographic location of the 52 population samples studied by Neil A. Rosenberg et al. (See Reference 2) The numbered barriers correspond to sequential partitioning of the sample into genetic clusters. Click for larger version (19K) A BROADER APPLICATION Particularly unnerving, for some, is the sampling of only a single population in Africa, where variation thrives. "The idea that all Africans are the same is just nonsense," says Kidd, "Compared to any two African tribes, the whole rest of the world shows no variation." Maryland's Tishkoff says that many of the SNP variants being used to create haplotype blocks were identified largely in non-Africans, "So, people have argued, 'You're introducing a bias.'" Collins says data support the ability to extrapolate based on a few population groups, but adds that a pilot project is underway to determine how variation from an expanded number of groups (up to a dozen including populations in South Africa, East Africa, and other continents) will affect haplotype construction. "We may be OK without sampling very broadly throughout the world. The similarities are substantial enough that maybe three [geographic] areas will be sufficient to produce a tool that you can use anywhere," says Collins The strict guidelines used today by the HapMap project and others4 evolved from the legacy of the HGDP. In addition to individual consent, the group being studied must also grant permission, which involves a greater level of community involvement, a departure from the types of research conducted decades ago. Greely says that the HGDP used sophisticated requirements for informed consent from individuals and groups. "I think the project put forth what is still, to this date, the most pro-research-population set of ethical guidelines that exist." But those guidelines did little to stem the tide of criticisms. Collins says they were ahead of their time, but still below today's standards, "[The HGDP] had the unfortunate circumstance of being caught in the middle of a transition in what the community-agreed-upon ethical standards might be. In many ways they catalyzed that transition, so they both inspired it and were victimized by it." Brendan A. Maher can be contacted at bmaher@the-scientist.com. References 1. M.J. Bamshad et al., "Human population genetic structure and inference of group membership," Am J Human Genet, 72:578-89, March 2003. 2. N.A. Rosenberg et al., "Genetic structure of human populations," Science, 298:2381-5, 2001. 3. H.M. Cann et al., "A human genome diversity cell line panel," Science, 296:261-2, 2002. 4. S. Jaffe, "Groundwork for genetic studies," The Scientist, November 25, 2002. function sendData() { document.frm.pathName.value = location.pathname; result = false if (document.frm.score[0].checked) result = true; if (document.frm.score[1].checked) result = true; if (document.frm.score[2].checked) result = true; if (document.frm.score[3].checked) result = true; if (document.frm.score[4].checked) result = true; if (!result) alert("Please select") return result; } .myradio{background-color : #94bee5} Please indicate on a 1 - 5 scale how strongly you would recommend this article to your colleagues? Not recommended 1 2 3 4 5 Highly recommended Please register your vote

The 0.1% Portrait of Human History

Interested in reading more?

Become a Member of