In recent weeks, as a new administration in Washington has worked diligently to possibly destroy our government (or merely, as Elon Musk claims, to trim its fat), Americans have learned a great deal about how that government works. They’ve also learned about the roles and responsibilities of a few unheralded agencies with obscure acronyms for names. One of these is the Office of Management and Budget, or OMB: a vast office that oversees federal operations and the census, and which is often responsible for determining the data and systems by which our tax dollars are allocated and spent.

One of the OMB’s legacies, which I’ve been thinking about a lot lately, was issued in 1977. That’s when the office adopted Directive No. 15, which established four standard racial categories by which Americans could be grouped: White, Black, Asian/Pacific Islander, and American Indian/Alaska Native. The OMB’s civil servants had identified a need for consistent racial data across the government, per a series of civil rights acts that had been recently signed into law. Yet these categories—racial labels that the OMB argued had no basis in genetic science—became the standard by which government, society, and a great deal of scientific research would come to approach the vexed topic of race, and still do so today.

The four categories comprising “OMB Race” were themselves borrowed from what I call “UNESCO Race”—an understanding of human difference that the United Nations’ scholarly wing, not long after the UN’s founding in the aftermath of World War II, proclaimed as humanity’s gold standard in 1951. UNESCO’s conception of race was based not in skin color but continent of origin; and it explicitly affirmed equality among peoples and rejected White supremacy. It also rebutted the eugenicist ideas of the Nazis and many other governments—including our own—whose approach to public health, immigration, and other realms reflected a tacit embrace of the idea that humanity was in fact comprised of different racial subspecies, identifiable by skin-color but possessed of other intrinsic traits, whose mixing was to be avoided.

Yet UNESCO’s continental framework was an idea as old as race itself, a concept that naturalists in seventeenth-century Europe had invented to sort through the flood of data returning with explorers and colonists traveling the globe. In ensuing centuries, as race was used to justify slavery and conquest and other ills, the bases and contours of racial ideas would shift—from Enlightenment thinkers’ expounding of the notion that only Europeans had a capacity for reason, to Charles Darwin’s affirmation that humans were a single species comprised of several subspecies, and on to modern eugenicists’ belief, riffing on Darwin, in a genetic hierarchy to those subspecies. But nothing underscores the persistent grip of continental “race thinking” on society like the fact that even our most concerted attempts to combat centuries of racism today have often ended up re-affirming race as an idea based on genes specific to discrete continental populations.

That was certainly the case with OMB Race, whose categories remained based on a continent-of-origin understanding of racial difference. And the OMB’s categories, which remain with us still, would have fateful consequences as genetic science rapidly progressed across the last decades of the twentieth century and into this one.

Since Darwin’s time, geneticists had used family pedigree maps of bloodlines to track down health and disease-causing genes in humans. But as scientists found ways to isolate and analyze specific sequences of DNA within genes in the tail end of the twentieth century, and as they learned to manipulate those sequences with new chemistry and informatic technologies, modern genetics was eclipsed by “genomics,” the study of all the DNA sequences in an organism. Genome scientists deployed novel techniques for mapping and cloning DNA to learn a great deal about the basic mechanics of human DNA that apply to every person. They found that DNA—deoxyribonucleic acid—is the machinery in our cells that encodes all the proteins (hormones, enzymes, antibodies) that we need to live. And they confirmed that every one of our cells contains the same DNA that’s passed down to us from our ancestors, via our parents’ egg and sperm, in the twenty-three chromosomes that lie in our cells’ nuclei. (A tiny bit of DNA, known as mtDNA, also resides in our cells’ mitochondria.)

Courtesy of W. W. Norton & Company

In 1986, the famed discoverer of the double helix, James Watson, along with several other leading genome scientists working at the Salk Institute, U.S. National Institutes of Health (NIH), and U.S. Department of Energy (DOE), called for a global initiative to sequence a human genome. The aim of these researchers––many of whom were homing in on sequences associated with cancer and other debilitating illnesses—was to produce a reference that anyone anywhere in the world could use to better understand DNA sequences. In 1989, Francis Collins, a researcher at the Howard Hughes Medical Institute in Maryland, used an innovative mapping technique he had developed to find the gene responsible for cystic fibrosis. Within months, his team and others around the world were able to pinpoint disease genes that had stumped geneticists for over a century. With this proof of concept, the federal government swiftly moved to fund the proposed genome project, allocating three billion dollars to James Watson at the NIH and the molecular biologist David J. Galas at the DOE.

Over the next couple years, 17 nations joined the NIH and DOE to launch the first Human Genome Project. Watson and Galas soon passed leadership to Francis Collins and Ari Patrinos, leader of the DOE’s Biological and Environment Research division. In these first years of the Human Genome Project, none of the project leaders discussed global diversity or race. They merely sourced DNA from the most convenient repositories on hand—those in northwestern Europe and the United States. At the same time, the Surgeon General and officials across America’s chief public health departments such as the Department of Health and Human Services (HHS), Centers for Disease Control and Prevention (CDC), and Food and Drug Administration (FDA) were working hard to figure out how to implement OMB race in their operations. The Surgeon General and HHS had published reports about the long-overdue need for racial minorities’ inclusion in health research and greater research into racial health disparities. To ensure that U.S. health science was fulfilling these briefs, the NIH mandated in 1992 that all publicly funded researchers tabulate the race of their study participants. Thus was the Human Genome Project indelibly shaped by existent discussions of a faux-scientific concept, fiercely lodged in our imaginations and our societies alike. Our genome’s mapping, which should have been predicated on helping bury race forever, was primed instead to perpetuate old lies about human difference.

*

Once scientists were required to report the race of their study’s participants according to the OMB taxonomy—White, Black, American Indian and Alaska Native, or Asian and Pacific Islander—they had to decide just how they were going to sample DNA and affix racial data to it. And this wasn’t only so in the United States. The Human Genome Project relied on collaboration by scientists and governments from around the world. But as the U.S. Health Department extended OMB race to more U.S. federal agencies that handled research abroad and the Project entered its final phase in the mid-1990s, it was the NIH that initiated much of its work. The U.S. government and leaders in American public health often administered its projects. And in 1997, the FDA also extended the use of OMB categories to drug research worldwide by requiring all new drug applications to tabulate safety and efficacy by OMB Race. This meant that drug makers interested in marketing drugs on an international scale had to report dosage recommendations by the racial scheme that the OMB had concocted for social purposes.

Seeing that the project’s own collection had no OMB data, that same year the NIH created a repository of samples that it gathered and stored using OMB Race. This racially stratified “Polymorphism Discovery Resource” would become the basis for nucleotide mapping projects and research throughout government, academic, and industry science for years to come. As the NIH worked to make the United States a leader in genomic science and diverse and equitable public health, Francis Collins and Craig Ventner, among other leaders of the Human Genome Project, began formulating a new understanding of race that matched the way sociologists and other scholars had come to view it: as a social construction formed in part by racial bias and racism, and constantly reinforced and enacted by extant inequalities and institutions in society. But these scientific leaders also insisted on examining how this social identity could interact with our genetics. They appeared in the international news media to exclaim that, without the social component, race was “a bogus idea” and that “ranking people” was something we needed to “get . . . out of our heads”; but they also called for using federal guidelines to bring racial minorities into genomics to better understand the diseases from which they suffered.

At a White House ceremony held in the East Room to commemorate the Genome Project’s first draft of the human genome, President Bill Clinton remarked on the project’s step forward for humankind, drawing attention to race: “In genetic terms, all human beings, regardless of race, are more than 99.9 percent the same.” Collins and Venter joined Clinton, Britain’s Prime Minister Tony Blair, and world leaders from Japan, China, Germany, and France in celebrating unity and denying inaccurate and harmful definitions of race. Collins delighted in sharing the news that “the only race we are talking about is the human race.” Venter added, “The concept of race has no genetic or scientific basis.” Yet within months, at a press conference, Collins issued a new statement on race that articulated the more nuanced genomic view implying that racial health disparities could be caused by legacies of gene-environment interactions accumulating disease disparities in different races over time.

With the release of the Human Genome Project’s first draft map in 2000, a new version of race dawned—“Genomic Race.” Like OMB Race and its immediate predecessors, Genomic Race revolves around a fictitious continental-genetic scheme (White, Black, Asian, Pacific Islander, Indigenous American) and views all races as being human, descended from a single origin, and of a single species. But unlike UNESCO’s conception upon which OMB Race is based, Genomic Race argues for research that combines social­, environmental, and genomic data to explain the amalgam of factors that make race real. It inadvertently positions its own science of gene­-environment interactions—how DNA impacts and is impacted by social experiences—as what will reveal the varying physiologies and pathologies we see in today’s races.

Scientists working in this paradigm believe in using genomics to illuminate the interaction of racial bias with genetic processes in the body—how race gets “under the skin,” and sickens us—making us more aware of the simultaneous social and biological nature of race. But like the paradigms of yesteryear, Genomic Race fails to dismantle the genetically deterministic view of race. In using continental and quasi-continental schemes of analysis like OMB Race, it reinforces the belief that racial differences are due to our genetics.

*

After the Human Genome Project first draft map appeared, the NIH donated its OMB-classified DNA samples to a new project comprising the Human Genome Project’s British sponsor, the Wellcome Trust, and 13 pharmaceutical and biotech companies. These companies established The SNP Consortium, named for single nucleotide polymorphisms, often called SNPs (pronounced “snips”), which are the most common type of genetic variation among people. Their goal was to develop an applied map of the human genome that could be used to enhance our understanding of disease and help facilitate research. In little more than a year, The SNP Consortium located over a million genomic variants and showed that humans passed down segments of nucleotides in chunks: what researchers called “haplotypes.” Genomic scientists rejoiced that they now had a clear path to understanding human diversity—to map these chunks of DNA in our cells.

Haplotype mapping promised to shorten the process of finding genetic culprits of disease, requiring a fraction of the time and cost of other forms of genome mapping. It also promised to generate important information about race: to make it easier to compare DNA in people and groups all over the world, revealing the history of human evolution and migration. Haplotype mapping would finally tell us what global diversity meant in terms of disease disparities—which OMB races had which diseases, and why.

In 2002, the NIH and The SNP Consortium, along with the Wellcome Trust and governmental agencies in Canada, China, and Japan, launched the International HapMap Project (what I refer to as “HapMap”). Following OMB guidelines, HapMap attempted to collect DNA from people believed to be indigenous to various continents.

Not every community embraced that mission of DNA collection. Some Native American representatives from the Arhuaco, Chippewa, Lakota, Navajo, and Paiute nations protested genetic analysis of indigenous American DNA because they did not agree with prior genome project interpretations of DNA lineages. These groups rejected the idea that Native American peoples were part of a single genetic lineage or that a DNA test could prove a person’s tribal membership. Especially given the government’s use of blood quantum laws, which limited claims to citizenship or land based on a person’s percentage of “native blood,” genetic testing was and remains a fraught topic in many native communities. Ultimately, HapMap decided to sample only DNA from Africa, Asia, and Europe, and to group those samples into continentally distinct collections. HapMap soon became the largest DNA repository in the world, one that scientists from nearly every country used in their research. This racialized database enabled Genomic Race to become the gold standard in health science worldwide.

HapMap’s leadership—scientists like Aravinda Chakravarti, a co-director with Francis Collins of the Polymorphism Discovery Resource and David Altshuler, leader of The SNP Consortium—were extremely sensitive to the quandary of addressing race. They saw how eugenics had produced a health science that served only Whites, and how eugenics had structured modern medicine to exclude racial minorities from research and healthcare. As leaders in American public health, they determined to make genomics a diverse, equitable, and inclusive science; they understood that without a strong inclusionary program, the effects of drugs and treatments on communities with minority ancestries would remain understudied, as would their basic genetics.

Project leaders compelled external researchers using HapMap samples to report any analyses with labels that either noted the population origins from which the DNA had been sampled or an acronym of their preferred ethnic terms. In the case of DNA that the project grouped as “African,” samples had to be referred to as “Bantu-speaking populations in Africa,” for example, or “YRI” for “Yoruba of Ibadan, Nigeria.” DNA characterized as “Caucasian” was to be referred to as “populations in Europe” or “CEU” for “Utah residents with Northern and Western European ancestry from the CEPH collection.” The DNA from individuals grouped as “Asian” was to be referred to as “populations in East Asia” or “CHB/JPT” for “Han Chinese/Japanese.” So, though project leaders attempted to communicate geographical and cultural specificity by grouping collections by continent, they prompted continental analysis and comparison according to OMB Race. Genomic Race made race and genetics seem inextricably linked, while obscuring the social reality of race that it purportedly desired to expose.

*

HapMap’s DNA collection and labeling, in the main, perpetuated the longstanding misconception that racial groups were genetically real. More in-depth haploypte mapping of groups living in Africa, though, showed once and for all that all humans evolved from a single ancestor in Sub-Saharan Africa, and that humans are indeed one unified species. Tracing the lineages of specific haplotypes, researchers showed that there are very few human groups that correspond to what scientists would call a distinct genetic population, or a “population isolate.” Groups overlapped and intermixed, not remaining genetically distinct, mating only with people in their own culturally or geographically distinct group (what scientists call “endogamous”), as the term implies. Though a population isolate could potentially arise if a group has remained apart from the rest of the world—for example, a group living on a remote island with no inflow of outsiders, or one residing in a mountainous territory without migration—very few groups have ever remained so isolated.

Haplotype analyses have shown that even ethnic groups that scientists once assumed met the criteria for endogamy have less shared genetic ancestry than was previously believed. For example, a group once deemed a population isolate is the Parsi community. Parsis, who number in the 20,000s, comprise a group of Persians who migrated from ancient Iran to what is today India and Pakistan to practice their religion, Zoroastrianism. Another population isolate is the Sherpa community of Nepal and Tibet who, living in remote regions of the Himalayas, have adapted to the highest altitudes on the planet.

As technology has advanced, scientists have gotten a better look into the genomes of these groups, and some have retreated from identifying them as “isolates.” Genomic analysis has shown that Parsis, for example, have been more cross-cultural than was previously known, and have been endogamous only periodically. Over half of their gene pool today can be attributed to ancestral DNA lineages common in other South Asian groups (as compared to Iranians who attribute less than 10 percent of their ancestral DNA to such lineages). Likewise, Sherpas display patterns of genetic ancestry similar to those of other Tibetan, Burmese, and Nepalese groups that are not considered isolates (blends of South and East Asian lineages).

In addition to debunking myths of isolation, haplotype analysis has shown that traits and diseases that were previously thought to only benefit or harm one group hailing from one specific continent are shared by others in faraway parts of the world. Rare traits, such as the ability to breathe at high altitude, exist in populations worldwide—Europe, Asia, and the Americas—as do rare conditions: Black people aren’t the only group who get sickle cell disease and Jewish people aren’t the only group who get Tay-Sachs. The genomic prevalence of these rarities in so many populations shows how similarly humans have evolved in different parts of the world, how far we have traveled, and how much we have “mixed” with the humans we have encountered.

Rare traits are sometimes shared by very distant and different peoples. In the case of high-altitude breathing, we see evolutionary similarities among Sherpas in the Himalayas and dwellers in the Andes. We can trace lactose tolerance in a number of ancient farming communities in Sweden, South India, and Sudan. Sickle cell disease is common among the ancestral populations of western and central Africa, India, the Arab Peninsula, the Mediterranean, and parts of South America, Central America, and the Caribbean. Tay-Sachs is prevalent among the Amish, Ashkenazi Jews, Cajun, and French Canadians, but has also been on the rise in Chinese populations for several decades.

All the high-resolution haplotype mapping has led genome scientists to formulate a new nomenclature of haplogroups, grouping similar haplotypes that have been shown to come from a common ancestor, and labeling them with letters and numbers (“U5,” “M14,” etc.). These codes refer to real sequences of DNA, not anything beyond that such as race or ethnicity, nationality or language, or even traits or diseases. Haplogroup M21a, associated with my mother’s native northern Indonesian ancestry, for example, is but a small fragment of my larger genome. M21a does not encode specific genetic functions in me that would make me the same as everyone who has inherited DNA from Indonesia, and it most certainly does not make me the same as all others with Asian heritage (that is, all who share my race). By the same token, haplogroup EM2 has been associated with African American ancestry. But the EM2 haplogroup is one of many short sequences of DNA that is shared by some Black people’s genomes but isn’t present in the genome of all Black people. In other words, there is no common, universal “Black” DNA.

The neutral coding of haplogroups in alphanumeric sequences helps us avoid confusing haplogroups, which refer to genetic ancestry from very specific pinpoints on the globe, with continental race. No haplogroup is shared by all Indonesians or all African Americans, which challenges the notion that our genomes hold “Asian” or “Black” DNA. The diversity and admixture of haplogroups show that scientists should not group DNA samples or populations into discrete races as global genome projects have done, and that genomic research should not deploy the language of race as Genomic Race has done.

No matter how well-meant, using OMB Race as a proxy for genetic variation is faulty because it inaccurately draws a genetic distinction between groups of people based on the continents their ancestors came from. These categories simply don’t correspond to biological differences. From a genetic perspective, there is no logic to grouping people based on their identity as White, Black, American Indian, Alaskan Native, Asian or Pacific Islander. There is a wide range of diversity within groups local to a continent, and there can be a large amount of similarity between groups across different continents. The inaccurate fixation on continental groupings of race then leads to a social and political reality that remains incredibly harmful for every human being in the world.

*

Some years ago, I interviewed genome scientists around the world on what they think about this chaos in categorization. I was struck repeatedly by the extent to which they not only embraced the disarray but thrived on it. Elite scientists, who wielded a great deal of power and routinely engaged in public outreach, wanted as much leeway as possible to articulate their own labels. They wanted to be able to use racial and quasi-racial labels in subject recruitment, and to use other kinds of labels, like ethnic or linguistic, when referring to DNA sequences. They also wanted to use racial,quasi-racial, or even nationalistic labels in reporting if their research subjects wanted them to do so. This freedom enabled them to align their reporting with their subjects’ self-determined political labels, for example, using the participants’ term “African American” instead of “East” or “West African.” They needed labeling to be ambiguous and flexible, from the earliest stages of recruitment to the latest stages of publication, so they could maintain their authority over the research process. As a result, they allowed for slippages in labeling even when those slippages seemed contradictory.

Talking with these scientists more recently, I have come to sympathize with their struggles as researchers walking a tightrope plaited with strands of population categorization, political representation, and interpersonal respect. But I have also concluded that label slippage confuses the public about the meaning of race and can introduce misinformation that fuels existing racism. There are many horrifying examples of the harm caused by misinterpreting genomic science. In 2022 a 19 year old in upstate New York, named Payton Gendron, cited genomic research in the manifesto he posted to Discord and Google, before he live-streamed himself killing 10 and wounding three Black shoppers at a Tops Friendly supermarket in Buffalo. He had misinterpreted study findings to mean that races had different intellectual aptitudes. But there are also many more subtle examples of misinformation that endangers people of color, such as when a doctor misreads a genomic study report that suggests a racial finding where there isn’t one and then mistreats a patient accordingly. Research into physicians’ knowledge about genetics and race, as well as their prescription practices by race, shows that physicians’ misunderstandings about genetics and race hurt their ability to diagnose and treat patients of color, and poor medical outcomes widen racial health gaps.

Today’s leading genome scientists have attempted to divert our attention and resources away from old-school notions of race, and many of them have tried to get us thinking in more complicated terms. But decades out from HapMap and the 1000 Genomes Project, we are still steeped in a genetics of race. We still see research and reporting awash in what sociologist Paul Giroy calls “race-thinking” in ways that harm the very groups we intend to help.

There have been other grave but perhaps less obvious barriers to scientists’ inclusionary work. Many genomic researchers have been vocal about “Eurocentrism” and have criticized past scientists for studying Whites only. But giving people of color the chance to donate their own DNA to genomic studies, or to take part in cutting edge science, does not increase access to basic healthcare, not to mention quality healthcare. And it does nothing to prevent the daily onslaught of racism that threatens people of color and leads to poorer health outcomes.

There is harm in equating race-based research inclusion with racial equity. As scientists of color have appealed to laypeople of color to get involved with DNA research, they have inadvertently contributed to racially siloed databases on a par with the siloed collections of the major international genome projects. This reservoir of data has only enabled the systemic practice of comparing the DNA of races as if they were genetically distinct populations—Africans versus Europeans versus Asians and so on.

When the geneticist Rick Kittles, who founded the hugely successful DNA-testing company African Ancestry, says, “When I go into West Africa, people see themselves in what I do and I make sure of that,” or when HapMap’s Charles Rotimi, an eminent Nigerian-born researcher at NIH, says you need “somebody that [people] can relate to, to participate in a study like this,” we must question the endpoint of research inclusion. The scholar Alondra Nelson, who ran the White House Office of Science and Technology Policy during the Biden Administration, warned of the danger of “authentic expertise”: such expertise can be a double-edged sword if it’s used to bring the racially oppressed into an even more genetically deterministic system that does nothing to challenge the social structures that cause oppression in the first place.

Good intentions abound in genomic science, and a complex understanding of what is at stake rules the field, whose leaders are committed to racial justice and to dispelling deterministic misinformation about race. But good intentions are not enough. Complex understandings are also not enough. Scientists can “speak truth to power,” as the political rallying cry says. But as long as genetic sampling and reporting are tagged to OMB Race and common categories, bracketing and therefore lacking sociopolitical analysis, we will continue to see race in terms of DNA differences that don’t exist—and continue to miss how and why race, though it has no basis in our genes, remains painfully real. ♦

Adapted from What's Real About Race?: Untangling Science, Genetics, and Society by Rina Bliss. Reprinted by permission of W.W. Norton & Company. Copyright © 2025 by Rina Bliss.