The Dreaded “Middle East” Autosomal Result

One of our blog followers, Ron, asked this question:

“My late father and his brother were born and raised on Hatteras Island which was a very isolated community until relatively recent times. Curious about their genetic ancestry, I had my uncle do the Family Tree DNA Family Finder test. His results for the Family (Population) Finder were:

Europe (Western European) – Orcadian 91.37% ±2.82%

Middle East – Palestinian, Bedouin, Bedouin South, Druze, Jewish, Mozabite 8.63% ±2.82%

The 8.63% Middle East was surprising since most if not all of his ancestors, going back 4 or more generations, were born on the OBX (Outer Banks). Most of the original families on Hatteras Island trace their roots back to the British Isles and western Europe.

Since my mother’s parents were immigrants from eastern Europe, I thought it would be interesting to know what contributions my maternal grandparents added to my genetic ancestry, so I submitted my DNA samples for the same test.  The Population Finder test showed that I was Europe Orcadian 100.00% ±0.00%. I was shocked that some other population did not show in the results.

Can you help me understand how the representative populations are determined and why Middle East didn’t show in my sample?”

Yes, indeed, the dreaded “Middle Eastern” result.  I’ve seen this over and over again.  Let’s talk about what this is and why it might happen.  As it happens, the fact that Ray is from Hatteras Island provides us with a wonderful research opportunity, because it’s a population I’m quite familiar with.

Given that Dawn Taylor and I administer the Hatteras Families DNA Projects (Y-line, mtDNA and autosomal), I have a good handle on the genealogy of the Hatteras Island Families.  They are of particular interest because Hatteras Island is where Sir Walter Raleigh’s Lost Colonists are rumored to have gone and amalgamated with the Hatteras Indians.  The Hatteras Indians in turn appear to have partly died off, and partly married into the European Island population.  Both the Lost Colony Project and the Hatteras DNA Projects at  http://www.familytreedna.com/public/HatterasFathers and http://www.rootsweb.ancestry.com/~molcgdrg/hatteras/hifr-index.htm are ongoing and all Hatteras families are included.

As part of the Hatteras families endeavor, Dawn and I have assembled a data base of the Hatteras families with over 5000 early settlers and their descendants to about the year 1900 included.  What Ron says is accurate.  Most of the Hatteras Island families settled on the island quite early, beginning about 1710.  Nearly all of them came from Virginia, some directly and others after having settled on the NC mainland first for a generation or so in surrounding counties.  By 1750, almost all of the families found there in 1900 were present.  So indeed, this isolated island was settled by a group of people from the British Isles and a few of them intermarried with the local population of Hatteras Indians.

Once on the island, it was unusual to marry outside of the island population, so we have the situation known as endogamy, which is where an isolated population marries repeatedly within itself.  Other examples of this are the Amish and Jewish populations.  When this happens, the founding group of people’s DNA gets passed around in circles, so to speak, and no new DNA is introduced.

Typically what happens is that in each generation, 50% “new” DNA is introduced by the other parent.  When the new DNA is from someone nonrelated, it’s relatively easy to sort out using today’s DNA phasing tools.  But when the “new” DNA isn’t new at all, but comes from the same ancestral stock as the other parent, it has the effect of making relationships look “closer” in time.

Let’s look at an example.

You carry the following average percentages of DNA from these relatives:

  • Parents 50% from each parent
  • Grandparents 25%
  • Great-grandparents 12.5%
  • Great-great-grandparents 6.5%

As you can see, the percentage is divided in each generation.  However, if two of your great-grandparents are the same person, then you actually carry 25% of the DNA from that person, not 12.5.  When you’re looking at matches to other people in an endogamous community, nearly everyone looks more closely related than they are on paper due to the cumulative effect of shared ancestors.  In essence, genetically, they are much closer than they look to be on a genealogy pedigree chart.

Ok, back to the question at hand.  Where did the Middle Eastern come from?

Looking at the percentages above, you can see that if Ray’s Uncle was in fact 8% (plus or minus about 2%, so we’ll just call it 8%) Middle Eastern, his Middle Eastern relative would be either a great-grandparent or a great-great-grandparent.  Given that generational length is typically 25 to 30 years, assuming Ray’s birth in 1960 and his uncles in 1940, this means that this Middle Eastern person would have been living on Hatteras Island between 1835 and 1860 using 25 year generations and between 1810 and 1840 using 30 year generations.  Having worked with the original records extensively, I can assure you that there were no Middle Eastern people on Hatteras Island at that time.  Furthermore, there were no Middle Eastern people on Hatteras earlier in the 1800s or in the 1700s that are reflected in the records.  This includes all existent records, deed, marriages, court, tax, census, etc.

What we do find, however, are both Native Americans, slaves and free people of color who may be an admixture of either or both with Europeans.  In fact, we find an entire community adjacent to the Indian village that is admixed.

We published an article in the Lost Colony Research Group Newsletter that discusses this mixed community when we identified the families involved.  It’s titled, “Will the Real Scarborough, Basnett and Whidbee Please Stand Up” and details our findings.

These families were present on the island and were recorded as being “of color” before 1790, so the intermarriage occurred early in the history of the island.

Furthermore, these families continued to intermarry and they continued to live in the same community as before.  In fact, in May and June of 2012, we visited with a woman who still owns the Indian land sold by the Indians to her family members in 1788!  And yes, Ray’s surname is one of the surnames who intermarried with these families.  In fact, it was someone with his family surname who bought the land that included the Indian village in 1788 from a Hatteras Indian woman.

So what does this tell us?

Having worked with the autosomal results of people who are looking for small amounts of Native American ancestry, I often see this “Middle Eastern” admixture.  I’ve actually come to expect it.  I don’t believe it’s accurate.  I believe, for some reason, tri-racial admixture is being measured as “Middle Eastern.”  If you look at the non-Jewish Middle East, this actually makes some sense.  There is no other place in the world as highly admixed with a combination of African, European (Caucasian) and Asian.  I’m not surprised that early admixture in the US that includes white, African and Native American looks somewhat the same as Middle Eastern in terms of the population as a whole.  Regardless of why, this is what we are seeing on a regular basis.

New technology is on the horizon which will, hopefully, resolve some of this ambiguous minority admixture identification.  As new discoveries are made, as we discussed when we talked about “Ethnicity Finders” in the blog a few days ago, we learn more and will be able to more acutely refine these minority amounts of trace admixture.

If Ray’s ancestor in 1750 was a Hatteras Indian, and if there was no Lost Colonist European admixture already in the genetic mix, then using a 25 year generation, we would see the following percentages of ethnicity in subsequent generations, assuming marriage to a 100% Caucasian in each generation, as follows:

  • 1750 – 100% Indian
  • 1775 – next generation, married white settler – 50% Indian
  • 1800 – 25% Indian
  • 1825 – 13.5% Indian
  • 1850 – 6.25% Indian
  • 1875 – 3.12% Indian
  • 1900 – 1.56% Indian
  • 1925 – 0.78% Indian
  • 1950 – 0.39% Indian

Remember, however, about endogamy.  This group of people were neighbors and lived in a relatively isolated community.  They married each other.  Every time they married someone else who descended from someone who was a Hatteras Indian in 1750, their percentage of Native Heritage in the subsequent generation doubled as compared to what it would have been without double inheritance.  So if Ray’s Uncle is descended several times from Hatteras Indians due to intermarriage within that community, it’s certainly possible that he would carry 6-10% Native admixture.  There are also records that suggest possible African admixture early in the Native community.

So now to answer Ray’s last question about inheritance.

Ray wanted to know why he didn’t show any “Middle Eastern” admixture when his uncle did.

Remember that Ray’s Uncle has two “genetic transmission events” that differ from Ray’s line.  Ray’s Uncle, even though he had the same parents as Ray’s father, inherited differently from his parents.  Children inherit half of their DNA from each parents, but not necessarily the same half.  Maybe Ray’s father inherited little or none of the Native admixture.  In the next generation, Ray inherited half of his father’s DNA and half of his mother’s.  We have no way of knowing in which of these two transmission events Ray lost the Native admixture, or whether it’s there, but in such small pieces that the technology today can’t detect it.

Hopefully the new technology on the horizon will improve all aspects of autosomal admixture analysis and ethnicity detection.  But for today, if you see the dreaded “Middle East” result appear as one of your autosomal geographic locations and your family isn’t Jewish and has been in the states since colonial times, think to yourself ‘racial admixture’ and revisit this topic as the technology improves.  In other words, as far as I’m concerned, the jury is still out!

131 thoughts on “The Dreaded “Middle East” Autosomal Result

  1. I assume that the 7.29% middle eastern Palestinian, Adygei, Bedouin, Bedouin South, Druze, Iranian, Jewish, Mozabite portion is related to the “St Luke Motif” in my mtdna.

  2. Hi Roberta! Will the NEW Population Finder that FTDNA JUST sent an email about help us with the “Dreaded Middle Eastern” results?

    • I sure hope so. I have 10% in the Middle Eastern and a definite genetic tie to the Lumbee. But so far nothing has changed on population finder.

      • I’m hoping there’s some North American Native American this time.

        Sent from my Etch A Sketch

        >

      • Hopefully. My Grandmother and her father from that line have straight black hair, very dark eyes, with a noticeable lack of body hair. My grandma said she wore her black hair in braids when she was young and it was so heavy it gave her headaches. This in spite of the fact that her mother was first generation German with very light blonde hair and grey eyes.

      • My dad came in at 16 +|- 3 which is right in line with the family lore. Like I said earlier, hopefully they have some North American designations this time around rather than just South American. Also looking forward to seeing what else there is besides Orcadian. cheers.

        Sent from my Etch A Sketch

        >

      • I have some South American, and Mesoamerican when I use Gedmatch. But my Middle Eastern goes to 1/3 on Gedmatch.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s