Tuesday, December 20, 2016

Interpreting Your Ethnicity Admixture Results (Amerindian)

Yes, we know that our ethnicity admixture estimates should be used casually, not causally, and is conversation-starter fodder best reserved for social events. We're reminded the best use of our DNA results (autosomal DNA) is synergistically utilizing them with our genealogies to build our family pedigrees, to connect with our genetic relatives and to trace our roots. But the truth is some of us are seriously hung over on our ethnicity admixture estimates and have found them useful.

 you're one of those people who loves to imbibe ethnicity admixture cocktails, like any other intoxicator you should at least enjoy them responsibly. In this blog I'm going to explain how you should be interpreting your ethnicity admixture results when trying to determine what they could mean. Essentially my goal is to get beginner and intermediate genetic genealogists to think more like population geneticists ("5% of my DNA is similar to British Isles populations...") rather than genealogy astrologers ("I have 5% Irish in me..."). Therefore I've adapted these 10 lessons from my inaugural Admixture Centrifuge blog and your questions over the years for quick reference.(NOTE: Although these lessons focus on Native American admixture you can substitute any ethnicity here as the principles of interpretation remain the same.) Enjoy and SHARE:

(1) You can't use your ethnicity admixture results for tribal enrollment or recognition, to obtain a tribal card or make any beneficiary claims to a Native American ethnic group or tribe. In fact Native Americans in the US and Canada have never used genealogical DNA tests for enrollment criteria or ethnic identity. Even the Bureau of Indian Affairs does not use genealogical DNA testing for its federal recognition process. Instead Native American tribes rely on genealogical paper trails and with US-government recognized tribes, valid documentation and their own "adoption" procedures. Sometimes organized tribes will use traditional paternity tests if the issue of parentage comes up with a child who may belong to a tribal member. However one useful way to use genealogical DNA testing is to search for genetic relative matches with proven ties to a specific Native American tribe and with whom you share Native American admixture (see my blog on Ethnicity Chromosome Mapping). You may be able to trace your Native American ancestry through these genetic relatives. However you would still have to provide reasonable documentation, and other genealogical or acceptable  proof. Of course this doesn't help if tribal enrollment is closed or if you don't meet certain Blood Quantum laws or other enrollment requirements, including ancestral, ethnic, cultural, genealogical linguistic, national and social ties to the tribe. 

(2) Your ethnicity admixture results can’t identify a specific tribe or ethnic group from which you descend or originate. Your ethnicity admixture results are just estimates of your genomic similarity to specially chosen reference populations. These estimates are derived by comparing your selected Ancestry Informative Markers (known as SNPs) to those found in the specially chosen reference populations. DNA testing companies essentially organize these reference populations into clusters and assign them an ethnicity label, which may not fully reflect the reference population's inherited genetic admixture or how they ethnically identify. And more often the same or similar reference populations may be organized and categorized differently from one DNA company to the next = what may be Amazonian at one company may be labelled Native American at another. Further your Native American percentages may be affected by the admixture calculator's Native American categories and reference samples utilized (or lack thereof) for those categories. Since most major DNA companies use a handful of available Amerindian populations — mostly from Central and South America — to represent their Native American admixture categories, there is no way possible for the DNA test to show what tribe you actually descend from based on such limited data. Some of the chosen reference populations also share substantial ancestry with reference populations labelled something else. Another point is your genetic similarity to these reference populations is determined by looking at very small amounts of SNPs that could be prevalent in many Native American and other ethnic groups, thus making it difficult to identify the specific common genetic ancestor, ancestral couple or ethnic group. The hard truth here is your ethnicity admixture results won't represent what's in you nor will they identify your specific ethnic origins, but they will show how similar your ancestry informative markers are to reference population datasets. If you take time to actually read the "fine print" about ethnicity admixture estimates that DNA companies provide, they will tell you as much;  for example AncestryDNA says: "Please note that genetic ethnicity estimates are based on individuals living in this region today. While a prediction of genetic ethnicity from this region suggests a connection to the groups occupying this location, it is not conclusive evidence of membership to any particular tribe or ethnic group." So infer if you must but never conclude. 

(3) Native American admixture won't be represented by Middle Eastern, Jewish, West Asian, South Asian, African or European admixture on a reputable admixture test with an adequate reference population dataset and genomic maker coverage. Harvard chemist Dr. Doug McDonald says, "If people see that they have this Middle Eastern percentage they are sometimes trying to find explanations in their recent ancestry. They think that the Middle Eastern component might represent Jewish ancestry, Native American ancestry, Moorish ancestry, etc, whereas in reality this is mostly not the case at all, if the rest is Orcadian/Irish." Dr. McDonald continues, "Native American is listed as just that. It is quite uncommon for it to be listed in error … except for genuine people from Siberia and Saami. Mideast won’t represent American! But it does mean something!"… But it certainly won't be an indication of a Native American ancestor. Since these DNA tests only look back about 500 years, if your test shows similarity to Middle Eastern, Jewish, West Asian, South Asian, African or European  admixture it most likely comes from a post-Christopher Columbus event or may represent another component of an ancestor who was not of Native American descent. 

(4) Blood Quantum laws do not statistically correspond to inherited DNA from our direct ancestors. Because of random genetic recombination, we inherit about 50% of our DNA from each parent. But you don’t inherit each parent’s genetic admixture evenly —  for example, if your mother has 10% Native American DNA, this does not mean you will receive 5%; you and your siblings could get anywhere from 0% to 10%. And with your grandparents and preceding generations, the DNA you inherit from them [via your parents] is uneven so you most likely will not receive 25% of DNA equally from all four grandparents as suggested by Blood Quantum laws. In other words, you may get 30% of any part of grandparent #1 ...  20% of any part of grandparent #2 ... 23% of any part of grandparent #3 ... and 27% of any part of grandparent #4. So if your paternal grandmother is 1/4 Cherokee by Blood Quantum this does not mean she will show 25% Native American admixture if she were to test. And even then she may only pass 10% of her Native American admixture to your parent even if she was 30% by admixture. This also means if 1/4 Cherokee grandma has other admixture — ie African or European   then she may pass more or less of the African or European to your parent.

(5) You probably share
 NO genetic relationship with ancient or modern Native American reference populations utilized by DNA companies, Gedmatch's admixture calculators and accompanying Oracles population-fitting programs. So if you see "Miwok" or "Lumbee" showing for you on Gedmatch Oracles, it doesn’t mean you are related to the Miwok or Lumbee tribes in any way unless purely coincidental. This also includes ancient DNA samples like Anzik Clovis child and Kennewick Man (see Estes Ancient DNA Matches -- what do they mean). And even if you were by rare chance related to the reference population sample, the matching ancestry markers may be too small to conclude a genetic relationship without additional proof. (I will note some DNA companies are exploring using our genetic relatives for reference populations but this is not in wide use yet.) In most cases certain Native American reference samples (usually academic) utilized on that particular test could be the most similar to your Native American ancestry even if not directly related. Another example, if you get a Native American percentage on a DNA test and see that "Mayan" is used as reference population, it does NOT mean that  you or your ancestors are Mayan or from southern Mexico and Central America. In fact Mayan is often used as an umbrella term to describe indigenous populations from Mexico and Central America, and in terms of ethnicity admixture tests as a generic proxy for all Native Americans. Dr. Doug McDonald says, “Mayan is the usual listing for any Native American north of Panama, through all of Mexico, and east of the Rockies in the USA and Canada." As a final related point, just because you match someone with a significant amount of Native American admixture, or that has a Native American haplogroup or whom identifies with a tribe does NOT mean you share Native American ancestors with them even if they are from the same region as your ancestors or have similar genealogical information.

(6) On ethnicity admixture calculators with sub-regional Native American categories (ie Mesoamerican, North Amerindian, South Amerindian), your DNA percentages may show in some or all of those sub-regional categories depending on the admixture test, reference population samples and methodologies utilized. The reason why you might show percentages to several Native American sub-categories, including such related categories as Siberian or Asian, is because on a macro-level indigenous peoples of the Americas are generally more similar to each other than to non-indigenous populations (ie Europeans, Africans). To note when comparing Amerindians to Siberians, the North Amerindians tend to show higher similarity to Siberians than Central/South Amerindians. [See the composition range of ancestral population components of Native Americans in this DNA Tribes Digest Article.] In most cases a showing of Native American DNA percentages in these more specific or related Native American sub-categories means there is good possibility of a Native American ancestor in your past. Nothing more.

(7) The reason why a DNA test using only minimum Native American reference datasets can pick up your Native American admixture is because Amerindian populations show a lower genetic diversity to each other than populations from other continental regions. In other words on a continental level,
 Native American populations are (a) more alike to each other than to non-Native American populations, and (b) are distinctive enough to be identified as a indigenous American from non-Native American populations. This is probably best explained with ancient genomes of the Kennewick Man and Clovis Anzik-1 (aka Clovis child), both of which purports to have more "ethnic purity" than modern Native populations.  According to Rasmussen et al,“When we compare Kennewick Man with the worldwide panel of populations, a clear genetic similarity to Native Americans is observed both in principal components analysis (PCA) and using f3-outgroup statistics....In particular, we can reject the hypothesis that Kennewick Man is more closely related to Ainu or Polynesians than he is to Native Americans....Model-based clustering using ADMIXTURE24 shows that Kennewick Man has ancestry proportions most similar to those of other Northern Native Americans, especially the Colville, Ojibwa, and Algonquin. Considering the Americas only, f3-outgroup and D-statistic based analyses show that Kennewick Man, like the Anzick-1 child, shares a high degree of ancestry with Native Americans from Central and South America, and that Kennewick Man also groups with geographically close tribes including the Colville ...".

Contrarily just because Native Americans are similar to each other on a continental or genome-wide level does not mean that all Native Americans are genetically alike on more granular genomic level. Per Rasmussen et al: “Despite this similarity, Anzick-1 and Kennewick Man have dissimilar genetic affinities to contemporary Native Americans. In particular, we find that Anzick-1 is more closely related to Central/Southern Native Americans than is Kennewick Man....The pattern observed in Kennewick Man is mirrored in the Colville, who also show a high affinity with Southern populations...but are most closely related to a neighbouring population in the data set.... This is in contrast to other populations such as the Chipewyan, who are more closely related to Northern Native Americans rather than to Central/Southern Native Americans in all comparisons.” With modern indigenous American populations, according to Bolnik et al, "Although populations from the same geographic region usually exhibit similar haplogroup frequency distributions ... those from the Southeast instead exhibit haplogroup frequency distributions that differ significantly from one another. Such divergent haplogroup frequency distributions are unexpected for the Muskogean-speaking southeastern populations, which share many sociocultural traits, speak closely related languages, and have experienced extensive admixture both with each other and with other eastern North American populations. Independent origins, genetic isolation from other Native American populations due to matrilocality, differential admixture, or a genetic bottleneck could be responsible for this heterogeneous distribution of haplogroup frequencies." 

(9) Modern Native American populations from different ethnic groups can be admixed with each other, as well as with modern Europeans, Africans and Asians. Prior to Christopher Columbus's arrival some Native American tribes were nomadic and may have moved around because of climate, environment and the availability of food and shelter. Native American tribes also had conflicts with each other, captured and enslaved each other (though not to extent of chattel slavery introduced by Europeans), as well as cohabited with and married outside of the tribe like most other human populations. There were also large cities like Cahokia, which was located in present-day Missouri. According to Wikipedia, it was the largest and most influential urban settlement of the Mississippian culture which existed more than 1000 years before European contact. Since Cohokia was described as a cosmopolitan city, we can reason it was a melting pot of Native American diversity and most likely they mixed with each other. After European colonizers arrived, Native Americans were forced from their lands and killed off, often causing different tribes to absorb other tribes, in addition to mixing with Europeans and Africans. Rasmussen et al explains: “Due to high levels of recent admixture in many Native American populations, we masked European ancestry from the Native Americans.  No masking was done on the Kennewick Man [because there was no need to]."

(10) Admixture percentages below 2% should NOT be dismissed outright as statistical noise. Notably the thresholds created for admixture percentages below 2% are essentially someone’s opinion based on what percentage is high enough to be genetically relevant. And let's face it most of you are fascinated with your trace admixture. Of course, smaller admixture percentages invariably means your Native American ancestry is very distant or virtually washed out by your generation. Scientific studies certainly discuss admixture percentages below 2%. For example a Bryc et al study using 23andMe customer data finds that African-Americans have an average of 0.8% Native American admixture and “more than 5% of African Americans are estimated to carry at least 2% Native American ancestry genome-wide … With a lower threshold of 1% Native American ancestry, we estimate that about 22% of African Americans carry some Native American ancestry ….” The point here is the lower percentages of Native American admixture discussed in the study are legitimate, and in this instance influenced by historic events (ie Indian Removals, which disrupted continuous gene-flow between Native Americans and African-Americans). In other instances a small admixture percentage may indicate one single DNA segment of your chromosome, which could be quite lengthy depending on location and chromosome. For example my 23andMe Ancestry Composition shows 0.6%(+/-) ethnicity similarity to Ashkenazi Jewish — this admixture is represented by one long segment on my chromosome 9. Yet I will warn that smaller admixture percentages should not be accepted lightly as legitimate, and therefore you MUST apply additional tests to determine legitimacy. This is because such trace admixture has a higher chance of being incorrectly assigned or misattributed an ethnicity label; OR it’s the closest fit for a population missing from the reference population samples offered by the test, OR it shows because you have ancestry from a population sharing genetic ancestral linkages with another population. As a general precaution you must make sure your trace Native American percentage shows in a consistent range — and at most conservative confidence levels — with a number of reliable personal genome services (ie 23andMe, AncestryDNA, FamilyTreeDNA) and admixture utilities (Gedmatch.com, DNA.Land). 
  • One legitimacy test is TRIANGULATION  — comparing yourself, your parent(s), other close relatives’ DNA segment(s) assigned “Native American” TO other genetic matches and their parents DNA segments assigned “Native American” —  if ALL OF YOU match each other on that particular Native American segment(s) then it’s likely you all share a common Native American ancestor /ancestral couple, and thus making it more likely the Native American admixture is real. [See Blaine Bettinger's Triangulation Intervention and Visual Phasing]. 
  • Another legitimacy test is PHASING — if you and your parent(s) take a DNA test (preferably with the same service), then you’ll be able confirm whether you inherited Native American ancestry from one or both parents, which also increases the odds your trace admixture is legit [see Roberta Estes's Parental Phasing].  
Furthermore, testing your grandparents, aunts/uncles, etc. may reveal higher amounts of Native American admixture and uncover Haplogroups exclusively found in indigenous populations, all of which may help legitimize your trace amount of Native American admixture. If you were expecting admixture that wasn't there, you may have to accept the fact that it is absence (see my blog Native American Admixture Is Just Not Into You). Ultimately only you can decide if your trace ancestries are worthy of pursuit.



  1. Glad I found your blog! Looking forward to reading more.

  2. Let's say for research sake, that my father's mother/ my grand mother is based on blood quantum rules is the product of a full blood native American woman & a full blooded African conceived in in 1890's. Thus one would assume that Grandmother and her nine other siblings are 50% Native American & 50% African. But, not so much based on the aforementioned narrative. So, perhaps some of nine siblings and her self included will inherit varying percentages of Native American & African DNA. So, now let's say my father's mother married a mixed man who possessed tri-racial admixture. Then, what would her off springs inherit as far as Native American DNA admixture. It seeming could get 0to be like Chinese math.

  3. . I was born in Denmark with a Danish mother and Polish father. Yet my Ancient DNA
    seams impossible unless those Siberian/Vikings went everywhere.
    Clovis Anzick-1 17.64 % North America 12,500 old Clovis boy Montanan
    Ust.lshim Siberia 15.64 45,000-year-old modern human male from Siberia.
    Br2 Hungry 20.94 Hungarian Neolithic, Copper, Bronze and Iron Age 5,000 yo
    LBK Luxingburg 19.23 Stuttgart in Germany 7,500 old farmer
    Loschbour Luxingburg 15.03 rock shelter in Heffingen, Luxembourg. 6220–5990 BC.
    NE1 Hungry 19.32
    BR2 “ “ 20.95
    Kosteeki 14 W. Russia 3.65
    Bot 15+17 Reo Brazil 2. Botocudo Indians and Polynesian