AncestryDNA Ethnicity Results

DNA Ethnicity results aren't an exact science, and in some respects, are mostly for entertainment purposes. A large number of genealogists will tell you they do the DNA tests in order to find living relatives to communicate with, not for ancestral ethnicity. For that knowledge, we research the records.

In 2012 AncestryDNA described my Ethnicity like this. I knew I should be about 75% European Jewish, if I inherited exactly 25% of my DNA from each grandparent, but 53+17 came pretty close.

In October of 2013 they updated their results, and the Uncertain amount disappeared.

The trace amounts of Pacific Islander surprised me. Caucasus can include Russia, so that wasn't too surprising. Though I later learned that the Caucasus was on my maternal line, which meant either some of my Transylvanian Jewish ancestors came from Russia originally, or there were some Caucasus roots elsewhere.

The breakdown has remained pretty consistent at Ancestry. At some point in the past 7 years, they  added their information on Communities, but the overall ethnicity breakdown has remained the same for me. Until a recent update:



No more Caucasus. No more Pacific Islander. And I am 79% European Jewish. (That's actually the high end of a 66%-79% range. So I think it's a pretty good estimate. And illustrates how useless DNA ethnicity charts are for most European Jews. Yes, the community information is nice, but Ancestry is unable to currently tell us how much from each.) The composition of the remaining 21% of my ancestry doesn't divert much from my research. 5% is almost one second great grandparent, so that feels a little high for my Irish/Scottish ancestry, but I know I have some. I don't know who my Finnish ancestors are.

Returning to the question of a Vanevery Surname Non-Paternity Event

Back in 2016 I blogged about the possibility of a Van Every Surname Non-Paternity Event. Several researchers have questioned the parentage of McGregor Van Every (1723-1786). One theory being Martin Van Iveren and an unnamed McGregor mother. One theory being a McGregor father and an unnamed Van Iveren/Vanevery mother.

Either way, there would be Van Every DNA in the descent, so the only way to prove the theories would be through a Y-surname test of a multitude of Van Every male descendants. Which I am not one.

Back in 2016 I noted that in all of my matches, I had 2nd cousins, 3rd cousins, 4th cousins, and 5th cousins, all sharing a Van Every common ancestor. Nothing higher than 5th cousins. All descendants of McGregor's son, David. If there was no Non-Paternity-Event, or even if there was, and one of the prevailing theories was true, one would expect more distant Van Every cousins.

A Shared Ancestor Hint popped up at AncestryDNA recently. (Excerpt below)

This provided a lot of excitement. Though I knew even with the two prevailing theories, I should have a match with someone descended from Martin Van Iveren. So even though our Family Trees matched up, it didn't prove anything. It was nice to see.

And then I checked the individual's Shared Matches.

As I said above, I have DNA matches who are 2nd through 5th cousins, all descended from David Van Every. Not one of them is a shared match with this individual.

Martin Van Iveren is in both of our online family trees. I think I am fairly sure that's not where our shared DNA resides. All of our shared matches have Eastern European Jewish DNA. That's the 75% of my ancestry that doesn't come from my maternal grandmother, Myrtle Van Every Deutsch. [It is fun to see a cousin from that 75% also has Van Iveren/Van Every ancestry.]

Which raises the question again. If neither I, nor any of my Van Every DNA cousins, share any Van Every DNA with this descendant of Martin Van Iveren...is it possible there was a Non-Paternity Event that completely broke the DNA trail. (Like an adoption?) It's certainly possible. Of course, the lack of evidence isn't proof of anything. It just continues to raise the question.  

DNA by the Numbers

My DNA - Ancestry.com (as of April 2017)

• 67% European Jewish
• 11% Western Europe
• 3% Ireland
• 2% Italy/Greece
• 2% Eastern European
• 2% Finland/Russia
• 5% Scandinavia
• 5% Caucasus
• <1 Pacific Islander
• <1 Siberian
• <1 Great Britain

Genetic Communities
• Jews in Central Europe
• Jews in NW Poland, Lithuania, & West Belarus
• Settlers of New York City & Long Island

5 DNA Circles
• Samuel Vanevery (1820-1888) – 2nd great grandfather – 4 members (3 DNA matches + me)
• Abigail Stuart (1825-1866) – 2nd great grandmother – 3 members (2 DNA matches + me)
• Andrew Vanevery (1798-1873) – 3rd great grandfather – 3 members (2 DNA matches + me)
• David Vanevery (1757-1820) – 4th great grandfather – 5 members (3 DNA matches + me + 1 non-match)
• Israel Swayze (1753-1844) – 4th great grandfather – 6 members (1 DNA match + me + 4 non-matches)

Cousin Matches
• 2nd Cousins – 2
• 3rd Cousins – 9
• 4th Cousins or Closer – 1,734

30 Shared Ancestor Hints: 1 second cousin, 1 third cousin, 6 fourth cousins, 22 distant cousins

My wife's DNA - Ancestry.com (as of April 2017)
• 27% Scandinavia
• 21% Great Britain
• 20% Western Europe
• 13% Eastern Europe
• 11% Italy/Greece
• 7% Ireland
• 1% Iberian Peninsula

Genetic Communities: None
DNA Circles: None

Cousin Matches
• Second Cousin: 1
• Third Cousin: 4
• Fourth Cousin or Closer: 304

19 Shared Ancestor Hints: 1 third cousin, 10 fourth cousins, 8 Distant Cousins

Notes

1) I discussed several months ago the possibility of a Van Every Surname Non-Paternity Event. One piece of evidence I used was that I appear to have no genetic cousin matches with a known-shared-ancestor more distant than my 4th great grandfather. I granted that it might be because it gets difficult to research that far back (though research that far back has been published for the Vanevery family.) Another possibility I considered was that the amount of shared DNA that far back becomes smaller and smaller.

Looking at my “Ancestor Circles” – the cousins in my 2nd great grandparent circles, should also be in my 3rd great and 4th great. But they’re not all there. I’m sure this isn’t due to an NPE, but instead, we share less DNA from our 4th great ancestors. We can share DNA from our shared 2nd greats, but still share none from our shared 4th greats.

2) I find the non-DNA matches in my Ancestor Circles interesting, but obviously, it is possible for two people to have different non-matching DNA from a shared ancestor.

3) I’ve also discussed the significant difference in cousin matches for my wife and me. I still think this is due primarily to a higher rate of false-positive cousin matches for Jewish testers. I’m not sure the reason behind this, but I do know Ancestry stated that they changed their algorithm to reduce it. (And they did reduce it significantly) But it is still there.

Still, I notice that my wife has 33% less Shared Ancestor Hints than I do. Those aren’t false positives. It appears she has a smaller percentage of relatives interested in their genealogy. (Or a smaller percentage who have done enough research for the Shared Ancestor Hints to kick in.)

4) As I mentioned when discussing my FamilyTreeDNA results, I am curious about my alleged 1% Iberian and Pacific Islander DNA. It may be a false-positive, but both AncestryDNA and FamilyTreeDNA claim it is there.

New FTDNA ethnicities

FamilyTreeDNA updated their DNA ethnicity guesses. (Yes, guesses, the science behind them is nowhere near exact)

My current breakdown is on the left.

I am aware that the trace results are so small that they can easily be false-positives.

However, is that Siberian indicative of my alleged maternal Native American ancestors who may have come over on the land bridge?

My mother's current DNA breakdown is on the right.

She doesn't have the trace Siberian. Is it possible that it is from my Paternal side? What about the Oceanian and Iberian?

I haven't encouraged my father to be tested as I assumed it would be 100% Jewish Diaspora/Eastern European, and the high number of false-positive cousin matches for Jewish DNA makes it less useful as a tool for finding cousins.

The odds are that my trace results are false positives, but still, perhaps I should get someone on my paternal side tested to be certain.

My Middle Eastern ancestry is clearly maternal. Ancestry (see below) still defines it as "Caucasus" and indicates it could also be Romanian, which suggests that it comes from my Transylvanian maternal grandfather.

Both Ancestry and FamilyTreeDNA suggest a trace of Pacific Islander/Oceanian. That's rather unlikely, but it does appear in both places, and is a very intriguing possibility.

The Possibility of a Van Every Surname Non-Paternity Event (NPE)

A Non-Paternity Event in genetic genealogy is

"any event which has caused a break in the link between an hereditary surname and the Y-chromosome resulting in a son using a different surname from that of his biological father" (Source)

When Selig and Julius, the sons of my ancestor, Samuel Dudelczak, immigrated to America and changed their surnames to Feinstein and Odelsohn, these were Non-Paternity Events. But that's not usually what is thought of when one hears the term.

In her Records of the Van Every Family, 1947, Mary Blackadar Piersol theorized that McGregory Van Every might be the son of Martin Van Iveren and an unidentified McGregory. Later researchers have wondered if it is equally possible that he could have been the son of a McGregory male and an unidentified Van Every female.

First, here is a transcription of relevant passages from Piersol's work indicating her sources (and lack thereof).

No entry has been found of the names of McGregory's parents due, probably, to gaps in the records of several of the churches of the Hudson Vallley. Indirect evidence, however suggests that his father was Martin Van Every and his mother possibly a member of the MacGregory family. His first children were named Francis and Patrick, the former a name from the Jaycocks family, the latter a favourite in that of the Colonel. Moreover with McGregory a significant change occurs in Van Every names. Previously they had all been Dutch but his is of Scotch derivation and among his sons we find such Scotch names as David, William, Peter and Andrew. Moreover one line of the Van Every family in Canada holds strongly the tradition that it has in its ancestry "a Scottish officer by the name of MacGregory." The wording is that of Anne Laurence Van Every, great-granddaughter of McGregory Van Every. (Piersol, p.98) 

There is evidence of intimate acquaintance in later years between the descendants of the Colonel and those of Burger Van Iveren … Not only did they live in the same general locality, Ulster and Orange Counties, NY., but their names appear in the records of the same churches and are linked by the same documents. When the Colonel’s son, Patrick, made his will, Francis Jaycocks signed as witness. The records of the Zion Lutheran Church at Lunenburg, now Athens, show that the Colonel’s grandson, also named Patrick, witnessed the baptism of Maria Jaycocks, and the banns for her marriage to McGregory Van Every were recorded by the Dutch Reformed Church at Poughkeepsie, N.Y. (Piersol, p. 97-98) 

As there is at present no documentary evidence as to the exact relationship I should hesitate to affirm it. But, as it was common knowledge of my Mother, and her Father, I would equally hesitate to deny it. The widow of Wm. Van Every lived with her son Joseph until her death, and my Mother, born in 1839, remembered her well. The legend is not without foundation as Dr. Piersol’s research shows. There was intimacy more than usual, which strongly suggests relationship, such as witnessing the will of the Colonel’s son, etc. The period to which it relates is not so remote as to be beyond the age of credible tradition. – J.J. Charteris-Thomson (Piersol, p.98-99)

As can be seen, her conclusions weren't based on solid documents. I have had my DNA tested at both Ancestry and FamilyTreeDNA. Of all matches that have Van Every ancestors in their known tree (that is, a tree they have made available), all are descendants of David Van Every and Sarah Showers, David being one of McGregory Van Every's sons. Cousins of my generation, with whom David and Sarah are my most recent common ancestors, are my fifth cousins. Both websites indicate 5th-8th degree cousins. There are matches that I have confirmed are 2nd, 3rd, 4th and 5th cousins through my Van Every lines. And then it stops.

One could attribute it to the difficulty of tracing one's lineage that far back. David Van Every served in the American Revolution. Except, Piersol's work has been available since 1947 to genealogists researching the surname, and her research documented several other branches.

The absence of a 6th cousin Van Every match, or greater, doesn't mean one doesn't exist. And it certainly doesn't prove an NPE. Perhaps I don't have enough DNA from McGregory. Even if there is an NPE, since I don't have any matches with descendants of David's siblings, the NPE could easily be with David and not McGregory. (Or both.)

Unless another descendant of David and Sarah [Showers] Van Every has had a different experience with DNA cousin-matching, the only way to resolve the question is for a group of Van Every males of varying descent to take the Y Surname test. I have no ability to make this happen beyond bringing the issue up.

DNA Cousin Matching

Below is the AncestryDNA summary for my wife's DNA matches

188 4th-6th cousin matches

Below is the AncestryDNA summary for my DNA matches

1025 4th-6th cousin matches

I suspect a number of readers who have had their DNA tested will look at my wife's numbers, and feel they are similar to their own. And then look at mine and be surprised. "You're so lucky!" they'll think.

Except...I'm sure there's a large number of false positive matches. Back in 2014 Ancestry changed their formula to reduce the number of false positives in general, but I think it's still high for Ashkenazic Jews. It's definitely still high for me.

• I have 437 pages of matches, 50 matches each, or a total of 21,850 matches. This obviously includes a lot of ancestors Ancestry categorizes as "Distant" or 5th-8th cousins.
• My wife has 163 pages, or 8,150 matches

The situation is similar at FamilyTreeDNA. I have a total of 5,728 matches, my wife has 876, and my mother has 3,374.

So if I assume the number of matches everyone gets, in a purely random environment, should be approximately similar...It's possible between 4/5 and 5/6 of my "4th Cousins or closer" matches are wrong.

So if the match doesn't have a family tree, and their Ethnicity information suggests we connect on my Jewish ancestry, why should I bother emailing them, when there's over an 80% chance of a false positive? [Unfortunately, even those with family trees don't include generations beyond the 'immigrant ancestor' which would be necessary to make a connection.]

Another situation I have -- several years ago, I came into contact with another researcher on one of my paternal lines who had done a lot of research in Polish archives, and his research took back that surname for me a couple generations providing many cousins. Yay! However, the name he had for my second great grandmother was wrong, the date of birth for my second great grandfather didn't match family records, and the birth record for my great-grandmother hasn't been discovered yet, leading me to wonder if there wasn't a second person with my second great grandfather's name. Well, this person has been declared a cousin of mine by FamilyTreeDNA, and the distance corresponds to the research I was given. However, is this proof?

Well, no. I can assume that I am related to all people of that surname from that Polish town. In some manner. However, the DNA match in no way confirms what generation we really have a shared ancestor. And the less accurate the matching is, the more this is the case. [But it is still nice to see his name pop up in my matches at FamilyTreeDNA.]

AncestryDNA Shared Ancestor Hints

AncestryDNA provides what they call "Shared Ancestor Hints." That is, if both you and a DNA match have linked to an Ancestry Family Tree, and there is a shared ancestor on those trees, they will provide you with the information. So instead of guessing how closely you are related by the amount of shared DNA, you know exactly how close. (Assuming both sets of research is accurate.)

A recent email exchange with one of these matches sparked my curiosity enough to categorize these matches by shared ancestor. I have 22 Shared Ancestor Hints. Breaking the hints down into the generational categories Ancestry provides as estimates:

2nd Cousin – 3rd Cousin
Shared ancestor
1. Melvin Vanevery (2nd cousin)

4th Cousin – 6th Cousin
Shared ancestor
1. Andrew Vanevery (4th Cousin)
2. William Denyer (4th Cousin)
3. Samuel Vanevery (3rd Cousin)
4. Israel Swayze (Sr) (6th cousin)

Distant Cousin (5th to 8th Cousin)
Shared ancestor
1. George W Hartley (4th cousin)
2. Samuel Swayze (7th cousin)
3. Samuel Vanevery (3rd cousin)
4. Israel Swayze (Jr) (5th cousin)
5. Israel Swayze (jr) (5th cousin)
6. Barbara Oberholzer (7th cousin)
7. Samuel Swayze (7th cousin)
8. Samuel Swayze (7th cousin)
9. Israel Swayze (Sr) (6th Cousin)
10. Samuel Swayze (7th cousin)
11. Adam Johann Schauers (7th cousin)
12. Adam Johann Schauers (7th cousin)
13. Joseph Swayze (8th cousin)
14. Adam Johann Schauers (7th cousin)
15. Israel Swayze (Jr) (5th Cousin)
16. Michael Schauers (6th cousin)
17. Barnabas Horton (8th cousin)

Totals by Surname
• Denyer: 1
• Hartley: 1
• Horton: 1
• Oberholtzer: 1
• Schauers: 4
• Vanevery: 4
• Swayze: 10

What traits do all these ancestors share:

1) They are all relatives through my maternal grandmother. I have many matches that I know are through my paternal grandparents’ and maternal grandfather’s lines. However, these matches haven’t posted their family trees back far enough for the Shared Ancestor Hints. Mostly because it is difficult to trace ancestry in Eastern Europe. Especially if your ancestors are Jewish.

2) My Swayze, Vanevery, and Schauers/Showers ancestors during the Revolutionary War were all Loyalists. It is possible that having ancestors in North America at the time of the revolution increases the likelihood that you will have traced your family tree back that far in order to identify Patriot or Loyalist lineage. (Or it's just a matter of having ancestors that far back in North America, where the research is relatively easy compared to other countries.)

Barbara Oberholtzer was part of a Mennonite family, who were mostly Conscientious Objectors. There were Patriot Hortons, but by the time of the Revolution they weren’t my ancestors.

3) I have a lot of Swayze matches. Does the celebrity nature of the Swayze name also have an influence on the likelihood of research?

Three out of the 21 matches (14%) are actually more closely related than Ancestry's estimate based on shared DNA. An 86% correct rate seems to me to be pretty good.

DNA Testing Without Relative Matching

Image Source: Office of Biological and Environmental Research 
of the U.S. Department of Energy Office of Science.

I've seen several people on social networks and on blogs question why, when they email one of their potential cousin matches on a DNA testing site, they don't get a response.

There can be reasons one wants the test, but doesn't want to communicate with relatives.

For example:

Let's say you've recently adopted a child. You either know the names of the birth parents, or you don't; you've either traced their ancestry back a few generations, or you haven't. You'd like to test their DNA to see the composition of their ethnicity, however, you have no interest whatsoever in communicating with close family members of the birth parents, and furthermore, you have no clue who the birth parents told about the existence of the child and their decision to give the child up. You feel it's certainly not your place to get involved in that. You can imagine the shock of seeing an unknown "match" that indicates a close Parent/Child or Aunt/Nephew relationship, and don't want to put a stranger through that shock

I've been reading the Privacy statements for 23andMe, FamilyTreeDNA and Ancestry seeing what provisions there are to have one's DNA tested, get an ethnicity breakdown, but not participate in the relative matching services. (Italics indicate direct quotes from their privacy guidelines.)

• For 23andMe their "DNA Relatives" feature is Opt-In.

23andMe gives you the ability to share information with other individuals who have 23andMe accounts through (i) our community forums, (ii) relative finding features (e.g., “DNA Relatives”), and (iii) other sharing features (such information is “User Content”). Please refer to your settings. You may be required to opt-in to some of this sharing, but some features require an opt-out. For example, we provide the ability to opt-in to our ancestry DNA Relatives Database where your information will be shared with potential relative matches. Alternatively, if you were participating in the DNA Relatives Database you may opt-out or change the visibility of your profile data by visiting your Settings. Also, please note that certain types of your User Content may be viewable by other 23andMe users and once posted, you may not be able to delete or modify such content.

• For FamilyTreeDNA the privacy policy states that they only share Contact Information with matches if both sign a release statement.

If the matching program finds a genetic match between you and another person in the database and you have each signed the release form Family Tree DNA will notify you via e-mail.
If a genetic match is found between you and another individual who enters the library at some time in the future, both will be given the information that a potential match is in the database provided that BOTH of you have signed the release form. Only where both parties have signed the release form will we release contact information concerning the separate parties to the other party. In this way, all persons in the database will have the right to decide if they want to contact their genetic match(es).

For users that signed the release form, Family Tree DNA may show your ethnicity breakdown to your genetic matches. However, you can opt-out of sharing your ethnicity and keep your results private with only you and the administrators of projects to which you belong by modifying your privacy settings here.

I participated in the Autosomal Transfer of my Ancestry.com DNA test results. I don't recall filling out the release form that they link to in their Privacy Statement. I suspect I checked boxes stating that I was consenting, or there was language that by performing the transfer I was consenting, as I am able to email matches, and they are able to email me. It is possible I filled out an online form without remembering doing so.

However, in the account settings, under "Family Finder Matches & Email Notifications" the option is provided:

Make the following DNA matches available on my Family Finder matches page and show my profile/contact information to those matches. Also notify me about new matches by email (Immediate, Close and Distant only). "Yes" enables all of the above. "No" disables all of the above.

It sounds like by selecting "No" I would disappear as a match from those relatives. And the option lets you choose different settings for Immediate, Close, Distant, and Speculative. These labels are defined here.

Immediate means Parent/Child, Siblings, Aunt/Uncle
Close means 1st and 2nd Cousins

• At Ancestry.com it appears you can't remove yourself from the matches list. Their privacy statement indicates

Throughout your AncestryDNA™ experience, we want to ensure that you are comfortable with your settings and how much content you wish to make public. Below are some ways that you can control the privacy settings of your AncestryDNA account and DNA results pages.

1. Username: You can choose how your name appears to your DNA matches...
2. Ethnicity profile: You can decide how much of your complete genetic ethnicity profile you want to make viewable to your DNA matches...
3. Family tree: You can choose whether or not to link your DNA test to an Ancestry.com family tree...
4. Deleting your test results: If you desire, you can choose to delete your DNA test results...

While this provides adequate privacy and security since you can use an anonymous username, not share your ethnicity profile, or your family tree, and refuse to respond to any messages that matches send you through the Ancestry message system, it doesn't seem to provide a way to hide you as a match completely so that you don't get any of those messages.

***

It appears that among the three major DNA services, 23andMe or FamilyTreeDNA are the choices for someone who wants solely to find out their ethnicity breakdown without the Relative-Matching services.

But none of them provide a complete 'cloaking' mechanism that would allow you to see all matches, but wouldn't let those matches see you. FamilyTreeDNA does offer "partial cloaking." One could say "no" to "Immediate" and perhaps "close" relatives and "Yes" to "distant" and "speculative." This would allow the person in the above speculated situation to communicate with more distant relatives, but remain hidden from the closer ones. However, it's not difficult to imagine a 3rd or 4th cousin telling the immediate or close match you're hiding from that you appear on their list as sharing the same chromosomes. So to truly prevent that from happening the person in the hypothetical situation mentioned would have to hide from all relatives.

This is just one example of why someone might not be responding to your messages. (Even if they are at a service where they could hide their results, they might not realize it.) There are probably other reasons, too.

Two DNA Circles or a Venn Diagram?

Ancestry updated their DNA interface by eliminating a large number of false positive matches, and adding a new feature called DNA Circles.

You can read more here and here on Ancestry’s Blog

With some initial reviews at

• The Legal Genealogist 
• Genea-Musings 
• The Genetic Genealogist

What are DNA circles? According to Ancestry:

Ancestry told me that I am part of two DNA Circles
One has 8 descendants of Israel Swayze, Jr
One has 5 descendants of Abigail (Coleman) Swayze

Ancestry calculates the likelihood of the members in the circle sharing the same shared ancestor: Strong, Good, Some, Emerging or Weak. In all of my cases, they're 'Emerging.' This suggests to me that we're all in these circles based mostly on our Family Trees, and less on our DNA, or else it would probably be stronger. However, their description of DNA circles above states all members must share some DNA with at least one other member of the circle. 

Israel and Abigail were husband and wife, and I am unaware of any other spouse for either of them.

All five matches in Abigail's circle are in Israel's circle.
According to our Online Trees, my most recent common ancestor for the three in Israel's Circle that aren't in Abigail's is actually Israel and Abigail's daughter, Johanna Swayze.

I have no clue why these two circles aren't identical - or how Ancestry would be able to distinguish whether particular DNA for anyone in these two circles was from Israel or Abigail, since we all, if the research that went into our online trees is accurate for all of us, are descendants of both of them -- unless Ancestry had information about DNA circles involving other Swayzes or Colemans. If this is the case, I would like to know this information.

Another conundrum is that a known second cousin, a descendant of Melvin Elijah Van Every through a different daughter, isn't in either of these circles, even though her tree is public. I wonder if this implies she doesn't share any of the Swayze-Coleman DNA. Perhaps she does have DNA from Israel and Abigail, but it's not the same DNA that the eight in these circles inherited.

Finding Your Roots: Jewish DNA

In Ashkenazi Jewish DNA and the Potential to Piece Together Shattered Family Branches Cece Moore, the Genetic Genealogy consultant for the television series, Finding Your Roots, discusses the DNA results they left out of the episode on Dershowitz, King and Kushner.

The episode that aired last week with Carole King, Alan Dershowitz and Tony Kushner did not include any DNA research, but that doesn’t mean that I hesitated to delve into their genetic genealogy. In fact, a short segment featuring Alan is included in the special DNA-themed last episode scheduled to air on November 25.

The article shares the results for all three, and further discusses the difficulties in using Jewish DNA to find cousins. However, Moore indicates she was able to use the results to find cousins for at least two of the three. She also shares a heartwarming story of a family reunification.

Comparing Ethnic Percentages for Ancestry and FamilyTreeDNA

As I mentioned a couple weeks ago, I transferred my raw DNA from Ancestry to FamilyTreeDNA.

I activated my account last Sunday - a process which took two business days. I didn't have access to any functions additional to what I received from the free transfer until Tuesday. This bothered me a little, since I felt I should at least be able to instantly email the 20 free matches they had provided me.

Anyway, once the account was activated, I was able to view FamilyTreeDNA's MyOrigins page, and see how their ethnic breakdown of my DNA compared to Ancestry's. Overall, they're very similar. Which is good to see. Though there are some differences.

FamilyTreeDNA's chart is on the left, Ancestry's is on the right.



FamilyTreeDNA tags a slightly smaller percentage of my European DNA as Jewish. However, when I add Jewish Diaspora + Eastern Europe + Eastern Middle East, I get exactly 75%, corresponding well to what I know of my three Jewish grandparents. On Ancestry, adding European Jewish + Europe East + Caucasus, I got 74%. [Ancestry's definition of 'Caucasus' includes Romania in addition to the Middle East as a possible source of the DNA. FamilyTreeDNA doesn't mention this, but I suspect my Transylvanian ancestry may be responsible, though ultimately, most of my Jewish ancestry should track back to the Middle East.]

The biggest difference might be FamilyTreeDNA says I have 10% Southern Europe DNA - while Ancestry suggested only 2% for Italy/Greece, and less than 1% for the Iberian Peninsula. I haven't traced any of my ancestry back yet to Southern Europe.

FamilyTreeDNA Autosomal Transfers

I read at The Genetic Genealogist that FamilyTreeDNA had dropped their DNA Autosomal Transfer to $39 (from $69), and that they allowed you to see the first 20 matches for free. (Though not all of the FamilyTreeDNA functionality is provided for free. What is provided I illustrate below.)

An Autosomal Transfer means that if you download your raw DNA data from 23andMe (the V3 Test) or Ancestry, you can upload the data to FamilyTreeDNA. This way you can find matches with additional cousins.

It's hard to ignore a free opportunity - which of course is what FamilyTreeDNA is betting on. Get lots of people to upload for free, and see what they can 'unlock' for $39. (You can also unlock it by getting 4 others to upload for free.)

If you follow this link, you can upload your data for free (and help me unlock my data for free.)
But you may want to see what you're getting first...

This is the screen you see first

Enter your name, email address, gender, and click that you accept their ToS. Then click Try it Free!

Click "Upload Raw Data" unless you need to click the help link for downloading data from 23andMe or Ancestry. I had already downloaded my data from Ancestry when they made that option available.

When you click a green circle will begin to swirl. About a minute later (at least for me) you will progress to this screen, where they tell you the upload is complete, and they will email you when they are finished processing it. That can take around an hour. Yesterday morning I uploaded my data at 5:43 before leaving for work, and received the email that everything was processed at 6:50.

Once processed, this is part of the screen you will see after logging in. MyOrigins (detailing FamilyTreeDNA's analysis of your world origins) is one of the options you need to unlock.

But you can click on Matches.

Here's what I see with my first two matches

As you can see:

• It tells you how many additional matches there are. (Interestingly, I've uploaded two sets of raw data, and for the other set, I am told there are only over 402 additional matches. 402 is nice, but a fraction of 3,158.)
• Indicates you won't be able to contact them without unlocking (unless you can find them with an internet search based on their names.)

When I mouseover the FamilyTree icon for the matches I see this response for all 20

Blaine Bettinger said that none of his matches had a family tree available either. The advert at the top of the match screen suggests even if the Family Tree were available, you'd have to unlock to see it. Since FamilyTreeDNA is a DNA-based website, and not a Family Tree-based website like Ancestry, I wonder if not a lot of people upload their trees. However, many do provide some surnames, which is somewhat helpful.

For example, one match below listed a lot of surnames, and their geographical locations. I know he is descended from the Van Everys who settled in Nebraska. I suspect this makes us third cousins, as our earliest shared ancestor is likely my 2nd great grandfather, Samuel Van Every.

If we click on the arrow below the profile image, you see more options. (Most of my initial matches haven't uploaded images. I suspect this is common)

The 'common matches' function needs to be unlocked. I suspect this is exactly what it implies, it will show all your matches in common. As far as I've seen, Ancestry doesn't provide this functionality, and that sounds great.

I can use the chromosome browser. If I click the + sign, I add the match to the Chromosome Browser. (You can compare up to 5 matches at a time.)

Then, if I click compare, I get to see exactly where in our chromosomes we matched.

This is also additional functionality not available at Ancestry, and looks like it could be very useful. If multiple matches match in the same area of a chromosome, it probably suggests a shared ancestor. I haven't done much research into this as it's not available at Ancestry.

One of the options on the Chromosome Browser is to "hide 3rd party matches." As the tutorial explains:

I, of course, am a "Third Party Match." They don't tell you up front that the matches are less reliable.
Of course, I won't hide 3rd party matches, since if they used Ancestry the comparison should be more reliable. Right?

So, will I pay $39?

Well....If I can convince 4 people to follow this link, and upload their data for free, I won't have to.
(Actually 3 people, since I followed it to upload a second file of DNA data I had. Of course, I'd like to unlock that as well. But I'll be happy if I only have to pay for one.)

Ancestry Updates their DNA Results

Ancestry last week updated the ethnicity breakdowns on their DNA tests. Some received their update a month ago in a small 'beta preview,' but I was one of the many who had to wait for them to roll it out to everyone.

Here are my original results, which I discussed back in August of 2012, followed by the new results.

Notes

• There is no longer an 'uncertain' category, dashing all my hopes of being part alien.
• While each percentage is actually within a range one sees after clicking for more information, using the approximate numbers given, my "European Jewish" percentage has increased from 53% to 67%, and my Eastern European percentage has dropped from 17% to 2%. This isn't really much of a change - just putting a different label meaning the same thing to me on the DNA. I am pretty certain all of the Eastern European DNA comes from my three Jewish grandparents. Ancestry now provides a detailed list of countries where the regional DNA has been found. For Eastern Europe they say: Primarily located in: Poland, Slovakia, Czech Republic, Austria, Hungary, Slovenia, Romania, Bulgaria, Serbia, Ukraine, Belarus, Moldova, Lithuania, Latvia, Estonia, Kosovo, Croatia, Bosnia and Herzegovina. Also located in: Germany, Russia, Montenegro, Macedonia. I do have some German ancestry on my maternal grandmother's side, but I am still willing to bet that the 2% remaining "Europe East" is from one of my other three grandparents. I have known ancestors for these three from Poland, Hungary, Romania, Ukraine, and Lithuania.
• There are a few new regions that provided a little bit of a surprise: 

Caucasus

Primarily located in: Iran, Georgia, Armenia, Iraq, Syria, Turkey, Azerbijan
Also found in: Turkmenistan, Kuwait, Bulgaria, Romania, Greece, Italy, Jordan, Palestine

The 5% could be entirely from my maternal grandfather's Transylvanian ancestry. Combined with the East Europe and European Jewish, that would bring the representation of those three grandparents up to 74%. As I noted in my original post last year, I know that the only thing I can be certain about is that my mother provided 50% of my DNA, and my father provided the other 50%.  To assume I have exactly 25% of my DNA from each grandparent would be foolish. There are no countries, however, on that list that I currently have discovered in my maternal grandmother's ancestry, and Romania is the country that makes the most sense otherwise. (Of course, If I trace my Jewish ancestry back far enough, I'm sure all three branches reach Palestine.)

Iberian Peninsula

Primarily found in: Spain, Portugal
May also be found in: France, Morocco, Algeria, Italy

While the approximation is less than 1%, if I really have Iberian DNA, I wonder if it represents Spanish Jews who fled the inquisition in the late 15th century? I know my maternal grandmother had some ancestors from Alsace-Lorraine, so there is a chance this DNA comes from them.

Melanesia

Primarily located in: Papua New Guinea, Bougainville, Fiji, Aboriginal Australia
Also located in: Solomon Islands, New Caledonia

While less than 1%, if I really have some Melanesian DNA, I have no clue which line it comes from. British or Dutch ancestors who were among the original explorers of the Pacific Islands, met up with natives, and returned? Australia was discovered by the Dutch in 1606, and my 8th great grandfather, Myndert Fredericksen, was allegedly born in Holland in 1636. Nothing is known about Myndert's mother. The British didn't arrive in Australia until 1780 which I feel is a bit late to intersect with my known British ancestors, and European explorers don't appear to have made it to the islands in the Pacific until the late 18th century either.

• Cousin Matches: The old results provided a caveat with the cousin matches that there were a high number of false-positives for European Jewish DNA. I am unable to find this caveat anymore with the new results. Does this mean that the update removed all the false-positives? I'm unsure, as for the most part I ignored any match for which I didn't see a shared surname, which was most of them. I still have a large number of matches. Over 5500 matches in total, including those that Ancestry indicates they have 'low confidence' in. I'm unsure if this is large compared to others.

Making Sense About Making Sense

Image Source: Office of Biological and Environmental Research 
of the U.S. Department of Energy Office of Science.

Sense about Science, a British charitable trust, has released a report entitled, Sense About Genetic Ancestry. The entire report is available as a free PDF download. Already several newspapers, such as the Telegraph, are stating that the report indicates DNA Tests are Meaningless. This is because the scientists refer to DNA testing as 'genetic astrology,' even though they are careful to say:

"There are credible ways to use the genetic data from mtDNA or Y chromosomes in individual ancestry testing, such as to supplement independent, historical studies of genealogy. If, for example, two men have identified – through historical research, possibly involving surnames – a common maleline ancestor in the sixteenth century, it would be reasonable to use their Y chromosome data to test this. There are some ancestry testing companies that offer this service.  

To answer a specific question about individual ancestry with any degree of confidence requires a combination of historical records and genetic information."

No genealogist is going to dispute that advice, but there is no credible scientific way to use astrology, which is where their metaphor breaks down, and leads to bad media reporting. It could be argued the language of the report is as sensationalistic as the language of the DNA companies which the scientists are complaining about.

The points the scientists raise are good ones. There probably are people taking the DNA tests without conducting historical record research, thinking the DNA test is all they need to do. And the report is meant for them, not for the genealogist who is taking the test with their eyes wide open about what the tests mean, and don't mean.

If you're unsure what the tests means, read the report. I'm no expert, but I have read a lot about DNA testing, and I don't think I disagree with any of their premises, though I do find fault with some of the conclusions.

Ancestry's New DNA test: A Jewish Perspective

This entry is not a religious perspective on the ethics of DNA testing.

This entry is a perspective on Ancestry.com's new DNA test from someone of European Jewish descent. This may apply to similar tests from other companies, but not having taken those tests, I am unable to say.

I believe Jews of mostly European descent will find the test less useful for them than for others. I found the results mostly useful for my 25% non-Jewish ancestry.

This is the test that determines one's DNA ethnicity breakdown. This of course can differ significantly from the breakdown of ethnicity of your ancestors in your genealogy database.  Each child receives 50% of their DNA from each parent, but it's a roll of the dice which 50% is passed on.

Theoretically, someone could have none of the DNA from one of their four grandparents.  For example, the 50% from your father might be just the DNA your paternal grandfather passed to him.This is unlikely, but it is also probably unlikely that one has exactly 25% of their DNA from each of their four grandparents. And it's even more unlikely that one has exactly 12.5% of their DNA from each of their eight great grandparents. As one goes further back on their genealogy chart, it becomes more likely that significant portions of your ancestors aren't represented in your DNA.

With that in mind, here are the results I received after submitting my saliva sample recently

From the perspective of my genealogy database, I am 75% Eastern European Jewish and 25% a mixture of European descent (Mostly British, German, and Dutch) and possibly some Native American.  So learning that my DNA is 70% Eastern European/European Jewish isn't much of a surprise.  However, I wasn't aware that that was all the DNA test was going to reveal about 3/4 of my ancestry. It certainly does reinforce the notion of Judaism as a Tribe, though it appears there is enough genetic variance to distinguish between European and non-Euopean ancestry. Looking at the charts of some other people, it appears those of Middle Eastern (Mizrahi) descent aren't 'religiously' tagged.

There are no Scandinavians in my family tree, but that's probably because I haven't gone back far enough.  My mother's Mitochondrial DNA test categorized my matrilineal line as Clan Ursula - which certainly has Scandinavian roots.

Ancestry does note:

Your genetic ethnicity reveals where your ancestors lived hundreds—perhaps even thousands—of years ago. This may update over time as new genetic signatures are discovered.

and

The rise of the Viking culture spread Scandinavian ancestry far throughout Europe. Their earliest coastal voyages took them to Scotland, northeastern England and established the settlement of Dublin, Ireland. As their power continued to grow, the Vikings spread farther afield, down the Volga River in Russia, to the coast of France and Spain.

So 17% of my DNA comes from some Scandinavian ancestors who migrated to either Britain, the Netherlands, or Germany. The 11% uncertain might contain what little Native American DNA I have, and perhaps some of the Alien DNA some of my friends suspect.

In their FAQ Ancestry explains:

Some people may have a percentage with ‘uncertain’ in their genetic ethnicity results. This means that small traces of a specific genetic population have been found in your DNA, but the probability levels were too low to pinpoint it to a specific ethnicity. This is not uncommon, and as more genetic signatures are discovered with a higher confidence level, we may be able to update this ‘uncertain’ percentage of your ethnicity over time.

This may mean that the 11% is a combination of different ethnicities, none of which are large enough to identify with certainty.

So the breakdown for me wasn't very helpful.  Others have had significantly different results. Judy Russell of The Legal Genealogist was very pleased with the specificity of her results.

Cousin Matches

One of the key selling points of the AncestryDNA test is that they will connect you with others who are potential matches.  And Ancestry provides a LOT of matches for me...

Unfortunately, there is a high number of false positives for those of European Jewish descent.

Despite their encouragement, at least for now, with the number of likely false-positives, it doesn't make much sense to me to contact the dozens of potential cousin matches.  Unless there is a shared surname in their online trees, which so far there hasn't been. (I am contacting the Scandinavian matches. There are less of them.)

I also understand that it may be complicated and science-y (is that a word?) - but I'd like to see the complicated, scientific explanation. It might confuse me, but I consider myself intelligent. I'd like Ancestry to include the scientific explanation on their website.

So...in summary:

• Those of mostly European Jewish descent may not find Ancestry's DNA test very useful due to the over-abundance of false-positive cousin matches, and the lack of specificity in ethnic origins. I am unsure if the results are similar on the tests provided by other companies. 

• I did find my results interesting, and I am in the process of contacting some potential cousins.

SNGF: Matrilineal Line

For his weekly Saturday Night Genealogy Fun, Randy Seaver at Geneamusings challenged:

Your mission, should you decide to accept it, is to:


1) List your matrilineal line - your mother, her mother, etc. back to the first identifiable mother. Note: this line is how your mitochondrial DNA was passed to you!


2) Tell us if you have had your mitochondrial DNA tested, and if so, which Haplogroup you are in.

Perhaps he came up with this idea because tomorrow is Mother's Day?

a) Me - John C. Newmark (1969 - )
b) (classified)
c) Myrtle Van Every (1900-1951) married Martin Joel Deutsch (1907-1991)
d) Margaret Jane McAlpin Monteroy Denyer (1868-1923) married Melvin Elijah Van Every (1863-1929)
e) Sarah Ann Hartley (1836-1898) married Ebenezer Ophan Denyer (1828-1872)
f) Eliza Beasley (?-?) married George W. Hartley (?-?)

The names of Eliza Beasley and George W. Hartley come from the Dawes Commission testimony of Sarah's brother Samuel Tillman Hartley.  And that is where my knowledge ends.

My mother has had her mtDNA tested and she was in Haplogroup U5 (Clan Ursula) - This suggests a possible origin in Northern Europe - such as Sweden, Norway, or Finland.  This discovery didn't disprove that Sarah Hartley was part Choctaw (she allegedly claimed to be 1/8), but it does suggest that her brother's claim in front of the Dawes Commission that their mother was 100% Native American (1/2 Choctaw, 1/2 Cherokee) was inaccurate.

Tombstone Tuesday: Samuel Deutsch (1861-1938)

When I first saw the below photo of my great grandfather's tombstone, I knew instantly, whoever was responsible for the inscription believed he was either a descendant of Aaron, brother of Moses, or from the planet, Vulcan. Either way, it was new information, and the precise descent equally difficult to trace. Since he died in 1938, when Star Trek creator Gene Roddenberry was only 17 years old, my bet was on the former.

Hebrew Transcription:
Shlomo Zalman Bar Avraham "The Cohen"
Died on the 20th of Sh'vat 5698

Last Week I discussed how I finally obtained the photographs of my great grandparents' tombstones, and both of them provided previously unknown details. I have long been a Star Trek fan, so I instantly recognized the hand gesture inscribed at the top of the tombstone as the Vulcan salute (rotated 45 degrees). I also knew that actor Leonard Nimoy had borrowed the gesture from his observances of the Priestly blessing as a child.

But just because someone thought he was a Cohen doesn't make it necessarily so. The most likely person responsible for the inscription would have been his widow, and there's a chance she may have been basing this on actually having met his father. So is there any way to prove this?

No. The only way to prove it would be to document the descent individual by individual back to Aaron. I don't see that happening. Yes, there is a Y-Chromosome marker that half of Cohanim who have been tested have been shown to have. Of course, the other half don't have it. Maybe the half that do have it are all descended from the same individual who falsely assumed Cohenship a Millennia ago. Is that impossible? Here's more information on the: Cohen Modal Haplotype.

Let's say we accept the marker as a test for descent. Can I be tested for this?

Sure, but my Y-chromosome is meaningless, since this is my mother's line. (And I have a paternal uncle who has already been tested, and my Y-Chromosome haplogroup should be identical to his.)

My mother can't be tested, because she doesn't have a Y-chromosome, and neither does her sister. There are some male Deutsch cousins whose arms we might be able to twist into taking the test.

I'll also point out that since the Cohen designation is traditionally something that gets passed down only through the male, which is why the Y-Chromosome test makes sense, I am not a Cohen, even if my maternal grandfather was.

He should get a re-vote

Snoop Dogg is shocked by some DNA results taken for the Lopez Tonight show, which indicated he was 6% European, and 23% Native American.

"We got to have a re-vote, this ain't right," said the 38-year-old.

According to Megan Smolenyak Smolenyak, in a Huffington Post article from back in December, the test the Lopez Tonight show is using is outdated, and often inaccurate. There is a newer test that is more accurate. So, Snoop Dogg ought to get his re-vote. (As should Larry David, from back in December.)

***

I'm not too interested in this test. Instead of testing your mt-DNA or Y chromosome origins, it tries to compute the percentages of nationalities in your personal genetic makeup. As I understand it, even if 100% accurate (which even the newer test isn't), it's not testing your ancestral heritage. Some may recall from genetics discussions in school that due to recessive and dominant genes, you might get more than half of your genes from your mother or father. It's not an even split. Over the generations, the difference is multiplied. Whole parts of your family tree may be missing in your genes, but present in your siblings or cousins, while they're missing other parts.

Some may be interested in knowing which parts they got, but if I am correct in my understanding, it doesn't really interest me.

Clan Ursula

In 1902 Samuel Tillman Hartley testified in front of The Dawes Commission that his mother was half-Choctaw and half-Cherokee. His sister was my great-great grandmother, on my maternal mtDNA line. Assuming Samuel was correct about his mother, and assuming he had the same mother as his sister, an mtDNA test of myself or my mother should have placed us in a Native American haplogroup.

Alas, it didn't. My mother's test came back this week, and we are in Haplogroup U5. (Clan Ursula)

Haplogroup U5 and its subclades U5a and U5b form the highest population concentrations in the far north, in Sami, Finns, and Estonians, but it is spread widely at lower levels throughout Europe. This distribution, and the age of the haplogroup, indicate individuals from this haplogroup were part the initial expansion tracking the retreat of ice sheets from Europe.

Haplogroup U5 is found also in small frequencies and at much lower diversity in the Near East and parts of Africa, suggesting back-migration of people from northern Europe to the south.

Haplogroup U5, with its own multiple lineages nested within, is the oldest European-specific haplogroup, and its origin dates to approximately 50,000 years ago. Most likely arising in the Near East, and spreading into Europe in a very early expansion, the presence of haplogroup U5 in Europe pre-dates the expansion of agriculture in Europe...Interestingly, individuals with haplogroup U5 and U5a may have been come in contact with Neandertals living in Europe at the time. 11% of modern day Europeans share this origin.

Nothing is wrong with being in this haplogroup, but we had hoped to confirm our Native American ancestry. It appears that Samuel and Sarah Hartley's mother wasn't 100% Native American. How much was she is uncertain.

With some research, I might be able to find a Y-DNA descendant of Sarah's brother to test the Hartley line. However, Samuel thought his father, George Hartley, was 1/2 Choctaw. If that meant George's parents included one full-blood Native American, that almost always meant the mother. Still, I think I would be interested in finding out his ancestry. Of course, even if I am able to find a direct male descendant, I don't know if I will be able to convince him to take the DNA test.

According to FamilyTreeDNA - they know of three matches for both HyperVariable Regions 1 & 2. They make it clear this doesn't mean they are necessarily closely related, and the common ancestor could be as much as 50 generations back, since mtDNA changes so slowly. However, they do provide contact information. The three recorded their origin as England, Ireland and Northern Ireland. Of course, there is no assurance that they were correct with the origin.

This leads me back to thinking about the names given to my great grandmother, Margaret Jane McAlpin Monteroy Denyer. The daughter of Sarah Hartley and Ebenezer Denyer, family names weren't uncommon as middle names in the Denyer line. However, the names could equally have come from fellow soldiers Ebenezer fought with during the Civil War - or elsewhere. I wouldn't mind having McAlpin lineage, though.

Paternal Grandmother's Patrilineal Line

Randy at Geneamusings provides his weekly Saturday Night Fun challenge. This week it was to map the patrilineal line of your paternal grandmother, with Y-DNA tests in mind.

Here are the questions he asks, and my answers:

* What was your father's mother's maiden name?

My paternal grandmother’s maiden name was Belle Feinstein

* What was your father's mother's father's name?

Her father was Herman Feinstein

* What is your father's mother's father's patrilineal line? That is, his father's father's father's ... back to the most distant male ancestor in that line?

Herman’s father was Selig Dudelsack (who changed his surname upon coming to America), and his father was Samuel Zvi Dudelsack.

* Can you identify male sibling(s) of your father's mother, and any living male descendants from those male sibling(s)? If so, you have a candidate to do a Y-DNA test on that patrilineal line. If not, you may have to find male siblings, and their descendants, of the next generation back, or even further.

My grandmother had two brothers. One brother had a daughter, the other brother had a daughter and a son. That son had a daughter and two sons. So there are some living male Feinstein-Dudelsacks who could have a Y-DNA test.

There are also some male Feinstein descendants of Herman’s brother Morris. (His other brothers Harry and Ben seem to have ‘daughtered out’, and his brother Aaron had no children.)

Selig had two male siblings. One, Gershon, remained in Poland. I have no idea what happened to that branch of Dudelsacks. The other, Yidel, came to America and changed his name to Julius Odelson. Julius’s son Samuel had 13 children, including ten sons. (Eight of whom served in WWII and were featured in a newsreel.) There are also some male Odelson-Dudelsacks available for a Y-DNA test. One that could prove scientifically that our branches are related.

If there were interest from both the Feinstein and Odelson branches, I might encourage it. Otherwise, it's not very important to me to find out which haplogroup each of my lines came from. Especially on my father's side where I expect them to be geographically similar. There can always be surprises, but I'm more interested in my maternal lineage DNA since I know there is more diversity there.

My mother sent an mtDNA test in a week or so ago, and perhaps we will be told that she belongs to haplogroups A,B,C,D or X which are the known Native American ones.