Not So Fast! When a DNA Match Isn't What It Appears To Be

You’ve found a match who shares a segment of DNA with you AND common ancestors. Is this enough evidence to make a conclusion that this is proof of your ancestry? The common refrain from researchers exploring their DNA is that a shared segment of DNA combined with a shared common ancestor or ancestral pair, is “proof” that DNA has been passed down through that exact path. We’re going to look at “proof” and why you should proceed with caution when it comes to your matches. Any discussion of “proof” cannot proceed without first examining the Genealogical Proof Standard (GPS) which requires that a reasonable genealogical conclusion be made only after meeting five criteria:

Keep in mind that DNA is evidence, just as documents such as birth, marriage, and death records, wills, and land records are evidence. DNA must be analyzed just as other evidence is analyzed. And, when it comes to DNA, items 3 and 4 of the GPS are paramount. Let’s explore some of the reasons why . . . When a relationship is close, the amount of shared centiMorgans can reasonably estimate the origins of a relationship through shared matches and the comparison of trees, when trees are available and accurate. However, beyond the relationship of 3rd cousin, it becomes more difficult to predict the source of the DNA, especially in a situation in which you might be related to a person in more than one way. For more on centiMorgans, please see Using CentMorgans to Estimate Relationships Let’s explore an example: My 5th great grandfather, Jacob Barncord, was married at least twice. The majority of trees found online indicate that the mother of his children was a woman by the name of Catherine Brandt. In Jacob’s household in the 1850 census is a woman named Catherine of the right age to be a spouse and others with either the Barncord surname or with surnames not clearly pointing to an obvious direction to explore.

As the 1850 census does not identify relationships between household members, we have to look to other sources to verify. One of those sources often cited by others is a DAR record from the 1960s that lists Catherine as a "first wife" and the mother of Jacob’s children. But was she? While Jacob's 1853 will lists his present wife as Catherine, it is Catherine's will that answers the question about her children. All of the children she lists have the surname of Sheller and none is listed as Barncord. These children include a daughter named Elizabeth who is almost certainly the woman by that name found in the 1850 household of Jacob Barncord and also with Catherine in the 1860 census.

Source 2: Pennsylvania Probate Records FamilySearch

Given the evidence presented, it would be a reasonable conclusion to assume that Catherine was at least a 2nd wife and that we need to look for additional information to identify the mother of Jacob's children. What other evidence do we have? While Jacob's first wife’s given name or surname has not yet been identified by me in records, the author Richard Warren Davis has stated in Emigrants, refugees and prisoners: (an aid to Mennonite family research) that a man by the name of Abraham Brubaker had a daughter who married a Jacob Barncord.

While the birth years of these children are not stated by Davis, he notes that Abraham was still single in 1788 and 1789. Various family trees place these children as being born about 1790, which would make Miss Brubaker a contemporary of my 5th great grandfather, Jacob Barncord. Not only is Barncord an uncommon name, these Brubaker and Barncord families resided in the same county, contemporaneous yet circumstantial evidence that justifies further exploration of this potential hint. Davis’s book does not cite the original source and a probe of the available works at the Family History Library in Salt Lake City has also failed to identify the source of Davis’s claims. The foreword to his book indicates that he did extensive research in local records in Pennsylvania, so any verification of Davis's claim would need to replicate his research locally. Thus, we have not, by any means, conducted reasonable exhaustive research in this matter. Additional research might explore other records associated with Abraham Brubaker or his other children, looking specifically for any records that might mention the Barncord family. But, what does the DNA say? Several descendants of Jacob Barncord have listed Catherine Brandt as the mother of his children and, because of that, Ancestry has indicated in the ThruLines tool (and previously in DNA Circles) the potential of shared DNA to Catherine. Because the system uses customer added tree data to identify beyond what is listed in the trees of DNA testers, hints may be generated that would appear to verify Catherine Brandt as a potential source of DNA. To many researchers, they’ll see this as “proof”. This is the exact situation in which exhaustive research should be done, with thorough analysis of conflicting evidence as well, including Catherine's will. While ThruLines is unlikely to offer support for the Brubaker theory, we can locate Brubaker DNA matches across multiple testing platforms, and can begin to build a potential DNA profile via DNA Painter.

In the four matches grouped together on the far right, we have the beginnings of a potential for DNA matching related to the Brubaker lineage. All matches are to me, a descendant of Jacob Barncord and his wife. Our goal is to identify potential evidence supporting Davis's Barncord - Brubaker union. Let's take a closer look:

Source 4: Extracted Image of DNA Painter

While we do not have thorough data on Match 4 to tie her Brubaker line to the others, she's an important link in this exploration. Matches 3 and 4 tested at Ancestry and uploaded to the free sharing site, GEDmatch so that we’re able to identify the segment shared. At Ancestry, our only option is to use Shared Matches because of the lack of the chromosome browser. Match 4 provides the greatest potential for data due to the size of the shared segment, 39 cMs. At Ancestry, there are 17 Shared Matches between me and Match 4. Five of those are descendants of Jacob Barncord and wife, 3 more are Brubaker descendants, and 9 do not have enough tree information to make a determination. Furthermore, one of those Ancestry Barncord descendants shares my 4th great grandparents, George Gipe and Elizabeth Barncord and provides another opportunity to evaluate the DNA data.

This match, A.B., shares a single, 59 cM segment that overlaps both Match 4, and two other Ancestry testers who also tested at 23andMe, giving us location data on Chromosome 19. These two are shown above with the orange arrows. That A.B. shares a long single segment and matches others on both the earlier and later portions of Chromosome 19 is suggestive that A.B.'s match is also likely on Chromosome 19. Furthermore, the long lavender segment that stretches across the middle of Chromosome 19 is another 23andMe match who also descends from George Gipe and Elizabeth Barncord through the same son from which A.B. also descends. These illustrations on my Chromosome 19 are a number of matches that might be suggestive of a relationship to the Brubaker family. Can we make a reasonable conclusion yet? Absolutely not. At this point, DNA is evidence and that evidence must be examined for conflicting evidence as well. One potential conflicting evidence involves endogamy or the custom of marrying within one’s own group whether that be a local community, tribe, religious group, or clan. The Mennonites would be considered to be an endogamous population and therefore open to potential opportunities to be related in more than one way. But, let’s look at another example of a chromosome 19 match with an ancestral line to the Brubaker clan of Pennsylvania: This match is from position 36 - 48 for 16.5 cMs, within the same region as Match 1 above (position 36 - 54 for 34 cMs).

But, there are problems here. Not only does this person, G.S.M., not share segments with Match 1, this is a maternal match to me whereas the other matches are all presumed to be paternal. Because my mother has tested her DNA and has her DNA at all of the sites where I am tested, I can narrow my matches to maternal and not maternal, or presumably paternal, given that the segment is of a substantial enough of a size to be a valid match vs a chance match. Let's take a look at how G.S.M. matches me and my mother at MyHeritage. The symbol notated with the arrow is a Triangulation hint, suggesting that the segment that I share with G.S.M. is not only shared with my mother, but is shared in the same location.

By clicking on that Triangulation symbol, we're taken to a new screen, showing how each of them matches to me. As expected, my mother (shown on the bottom) matches me on each chromosome from end to end. G.S.M. matches me only on the chromosome 19 segment, and MyHeritage designates that this is a triangulated segment. G.S.M. and my mother match me on the same segment and therefore, G.S.M., unlike the others, is a maternal match to me.

A close look at G.S.M's tree shows a potential of THREE different relationships on entirely different branches of my tree! Two of those relationships are paternal to me (the potential Brubaker branch and another from the UK), whereas one is on my maternal grandfather’s branch. In this particular case it is entirely coincidental that the match to this gentleman just happens to have a Brubaker ancestor and that it just happens to land in the same chromosome region as my potential paternal Brubaker cousins' matches to me. I also know from researching the paternal U.K. branch that the surname he has in his tree settled in the same part of the USA as my maternal grandfather. While some of his closer cousins share paternal DNA with me and also share matches to my maternal side, G.S.M. shares maternal DNA with me and also matches others to my paternal side. Confusing? You bet! This is why you should always approach your DNA matches analytically before making conclusions. If one had not thoroughly examined this match, it would be quite easy to misinterpret that segment. Rather than being supportive of the Brubaker theory, it is merely a DNA distraction and clear evidence for caution when exploring DNA matches. When you begin to view your DNA results through the lens of the Genealogical Proof Standard you'll gain a far better understanding of your results. Source Citations: 1. Image: "United States Census, 1850," database with images, FamilySearch (https://familysearch.org/ark:/61903/1:1:M4HF-6XY : 12 April 2016), Jacob Barncords, Greencastle, Franklin, Pennsylvania, United States; citing family 80, NARA microfilm publication M432 (Washington, D.C.: National Archives and Records Administration, n.d.). 2. Image: "Pennsylvania Probate Records, 1683-1994," images, FamilySearch (https://familysearch.org/ark:/61903/3:1:3QSQ-G99B-GRZT?cc=1999196&wc=9PMD-6TG%3A268495201%2C271074601 : 3 July 2014), Adams > Estates 1859-1870 and 1973-1976 no 4376-4500 and 1/76/124-1/76/159 > image 2395 of 3181; county courthouses, Pennsylvania. 3. Extracted Information: Davis, Richard W. Emigrants, Refugees and Prisoners (an aid to Mennonite family research). Volume II. Provo, Utah. R.W. Davis, 1997; p 98. 4. Images: DNA Painter Chromosome Mapping Tool. [cited: 2019 Jun 4] https://dnapainter.com/ 5. Image: MyHeritage Chromosome 19 match. [cited: 2019 Jun 4] https://www.myheritage.com 6. Image: MyHeritage Match List. [cited: 2019 Jun 4] https://www.myheritage.com 7. Image: MyHeritage Triangulation [cited: 2019 Jun 4] https://www.myheritage.com #DNA #GenealogicalProofStandard #GPS #Triangulation #DNAMatch #Geneaogy #FamilyTree