Kevin Lamarque / Reuters

After receiving a direct challenge from President Donald Trump to prove her Native American ancestry by submitting to a DNA test, Senator Elizabeth Warren released her results Monday morning. As an enrolled member of the Diné (Navajo) Nation, I obviously have some thoughts on this development. As a geneticist-bioethicist, I have even more. Whether or not Warren has genetic ties to indigenous peoples, she is not Native American, and her public gambit could dangerously obscure that distinction.

Warren did not use a direct-to-consumer test kit, such as 23andMe or AncestryDNA. Instead, she consulted the expertise of a leading population geneticist, Carlos Bustamante.

Statistical genetics rely on well-established facts about DNA inheritance. A quick recap in case you’ve forgotten your high-school biology: People inherit half of their genetic material from each parent and, with every generation, roughly half of that genetic information is lost. Even meiotic-recombination events—the crossing over of chromosomes when two gametes meet, which can explain diversity of traits within the same family—originally thought to be random, can be mapped to statistically predictable regions of the genome.

If you compare your genetic information with that of members from a reference population, you can estimate how long ago an ancestor belonged to that population based on statistical patterns of inheritance.

But your estimate is only as good as your reference.

Bustamante could not use U.S. tribal groups as a reference, because many have declined to participate in genomics studies and have refused to deposit DNA in publicly available databases. (More on that later.)

In the case of Warren’s individual DNA test, Bustamante resorted to what he considered the next best reference: 37 individuals from Mexico, Peru, and Colombia. According to Western scientists—who prefer migration theories that are culturally incongruent with indigenous origin stories—indigenous individuals from these countries represent the closest “sampled” biological relatives to Native Americans (that is, to those who originally inhabited what is now the United States).

This approach leaves a lot to be desired.

There is a great deal of diversity among the 573 federally recognized tribes in the contiguous U.S., all of which have unique cultural, linguistic, and historical traditions, as well as genetic differences. Of course, there exists even greater genetic heterogeneity among indigenous populations across the Americas. The terms Native American and indigenous are not interchangeable. While indigenous is ascribed globally, Native American is usually specific to the tribal peoples of the U.S.

Bustamante concluded that Warren had a “Native American” ancestor in her pedigree six to 10 generations ago, but he should have used the more general indigenous in his report.

In relation to the larger issue here, Native American U.S. tribes have the sovereign right to establish their own enrollment rules, and to my knowledge, they all consider DNA evidence to be insufficient. Indeed, the Cherokee Nation has already denounced the use of DNA tests to claim ancestry as “inappropriate and wrong.”

Who we are—the languages we speak, the traditions we practice, our broader cultures—inform our identities far more than DNA. Culture is not the same as a few genetic biomarkers, especially when those markers have statistical commonality to indigenous people with vastly different histories than Native Americans of the U.S.

Warren seems to get that. She stated, “I’m not enrolled in a tribe and only tribes determine tribal citizenship.” She is not seeking tribal enrollment and she does not (at least currently) claim to identify as culturally Native American.

Warren was born and raised in Oklahoma, which is also home to the descendants of those who were forcibly relocated from the southeastern states by President Andrew Jackson under the Indian Removal Act of 1830. She says she was told of her Native American heritage from childhood, and it’s understandable that she would want to check family lore against the seemingly hard evidence provided by a DNA test.

But whatever Warren intended, the conversation about her test results predictably degenerated into a back-and-forth about whether she is or is not Native American—and the answer to that is clear. She is not. Even if she could show a direct lineage through tribal census rolls and genealogical records, Warren has not experienced the traditional and cultural ways of Native American life.

To ascribe any power to a DNA-test result disempowers those Native Americans who do live according to their traditions. Native American identity is not one of biology, but of culture. And, crucially, “Native American” is a political designation that confers rights. If that designation becomes tied to a DNA test, it could threaten those rights.

I mentioned earlier that most tribal groups refuse to deposit their DNA in public databases. They are concerned that genomic information might be used to cast doubt on and threaten the sovereign status that allows members to access resources—such as water, education, or health care—as stipulated by treaties with the federal government. They want to make sure that tribal communities, not scientists and statistics, remain in control of the system.

Native Americans could use political support for our identities, but Elizabeth Warren’s public DNA test hardly fits the bill.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.