Genetic EthnoNullification: The Indigenous Ancestors Left Outside the Model

An African American ancestry study explicitly excludes Indigenous sources. Ancestry cites a similar method for its simulations. What does that connection reveal? By Ishmael A. Bey | First Tribe Historical Research Institute

A model cannot separately count an ancestral source it has not included. That simple observation sits at the center of a question deserving public attention: how adequately do the simulations used to evaluate ancestry estimates represent African American families with American Indian lineage?

The evidence begins with a published limitation, rather than an allegation about manufactured DNA.

The sentence that warrants attention

In their 2023 study, Jazlyn A. Mooney, Lily Agranat-Tamir, Jonathan K. Pritchard, and Noah A. Rosenberg state:

“we have not considered Native-American or other sources.”

The statement appears on printed page 13, under “Limitations.” Their African American application models African and European sources. The authors cite genomic estimates of relatively small Indigenous contributions, acknowledge measurement difficulties, and explain that including Indigenous sources would replace some modeled African and European ancestors. They also acknowledge that their application does not account for regional variation across the United States. [1]

This is a disclosed simplification. It is neither evidence that Indigenous ancestors did not exist nor, by itself, proof of an intention to erase them however they were NOT INCLUDED.

For descendants researching particular families, the question is whether a population-level simplification leaves their histories insufficiently represented. An average is not an individual genealogy.

Ancestry.com Ethnicity 2026 white paper https://www.ancestrycdn.com/support/images/Ethnicity2026_white_paper.pdf https://rosenberglab.stanford.edu/papers/MooneyEtAl2023-Genetics.pdf

Where Ancestry enters the story

Ancestry’s October 2026 white paper describes African American performance simulations using a derived method similar to Mooney and colleagues’ study. It explains that historical information guides simulations where pedigrees do not capture admixture patterns. These simulations begin with test-set reference-panel members. European simulation sets likewise use reference-panel people as founders. [2]

This establishes a methodological connection, not an identity between the two implementations. The white paper does not specify in that passage whether Ancestry preserved the cited study’s exclusion of Indigenous sources.

That distinction determines what an investigation can responsibly conclude. A documented omission in the cited model warrants disclosure about Ancestry’s implementation. It does not demonstrate that a customer’s Indigenous DNA has been assigned as African.

What a simulation can establish

Consider a practice examination that leaves out an entire subject. A high score cannot establish competence in the missing subject. Likewise, if ancestry test cases exclude a relevant source, performance on those cases alone cannot establish accuracy for that source.

This is a conditional methodological criticism. To establish actual customer harm, researchers would need to examine the implemented simulations, independent validation evidence, and assignment errors.

A historical reconstruction, not a recovered family tree

Mooney’s framework uses 14 generations of 25 years, ending with the 1960–1965 birth cohort. It reconstructs genealogies under assumptions; it does not directly verify every ancestor across those generations. [1]

Historical assumptions therefore deserve examination alongside statistical performance. Which histories are represented? Which are simplified? How are families with different histories evaluated?

Genetic EthnoNullification as a question of representation

Ishmael A. Bey uses “Genetic EthnoNullification” as an interpretive term for Indigenous ancestry disappearing from analytical representation through exclusion or classification. It is not an established scientific diagnosis.

The term identifies a concern that can be investigated without treating its consequences as already proven. Exclusion from a model must never be confused with demonstrated absence from a family. The disclosure the public needs

Ancestry should explain:

• Whether its African American simulations include Indigenous American sources.

• Which historical assumptions and ancestry proportions those simulations use.

• Whether regional and family-history variation are represented.

• What independent tests evaluate descendants with documented Indigenous lineage.

• How performance changes when Indigenous contributions are added or varied.

The verified finding is specific: a study Ancestry cites excludes Indigenous sources from its African American application. The unanswered question is whether Ancestry carries that exclusion into its own simulations, and with what consequences.

Public confidence benefits from an answer that can be checked.

Sources and verification

[1] Mooney JA, Agranat-Tamir L, Pritchard JK, Rosenberg NA. “On the number of genealogical ancestors tracing to the source groups of an admixed population.” GENETICS. 2023;224(3):iyad079. DOI: 10.1093/genetics/iyad079. See printed pages 5–6 for chronology and page 13 for limitations.Published PDF hosted by Stanford | Journal record | PubMed record, PMID 37410594

[2] AncestryDNA. Ancestral Regions 2026 White Paper. Updated October 8, 2026. See Section 4.1, PDF pages 17–18, for simulation methods.Official white paper

Evidence note: Source statements and the exclusion are verified. Ancestry’s retention of that exclusion, customer-level misassignment, and an intention to erase Indigenous identity have not been established in this investigation. No claim is made that African American DNA or all ancestry results are synthetic.