Tag: DNA

I last wrote about an AncestryDNA match who was a descendant of Samuel and Judith (Wolf) Snyder, a possible cousin through Jacob and Catharine (___) Snyder and/or Jacob and Magdalena (Brey) Wolf. This post is about what I learned by mining our Shared Matches.

Besides my mother, I share four matches with this cousin, who for simplicity’s sake, I’m going to call “E.” Let’s call these four match #1 – #4. According to Ancestry they all match within the 4th through 6th cousin range. That means we share 3rd through 5th great grandparents as common ancestors.

Since I’m looking to prove a match to either Jacob and Magdalena (Brey) Wolf and/or Jacob and Catharine (___) Schneider, I would need to match these individuals through one of the following lines:

• Jacob Schneider & Catharine (___), 5x great grandparents
• Henry Schneider & Sarah Wißler, 4x great grandparents
• Joseph Schneider & Judith Deischer, 3x great grandparents
• Conrad Wolf & Catharine Yeakel, 5x great grandparents
• Jacob Wolf & Magdalena Brey, 4x great grandparents
• Joel Wolf & Elizabeth Krauss, 3x great grandparents
• Conrad Brey & Maria Magdalena Klein, 5x great grandparents

If the generational estimate is off, a match might be through the ancestors of either Jacob Schneider or his wife Catharine or Jacob Wolf or his wife Magdalena Brey. I have only determined ancestors for Jacob Wolf and Magdalena Brey, so a Snyder connection would be more difficult to figure out.

Shared Matches

Match #1 has a large online family tree, but it does not share any family members with mine. Match #1 and I share 22.6 cMs on two DNA segments. Our shared surnames are Snyder and Walker. Snyder might connect with E, but Walker would connect to other maternal ancestors.

Match #2 has a small online family tree, but it is private. We share 22.4 cMs on one segment.

Match #3 also has small online family tree, but again, no family members in common. We share 20.5 cMs on two segments.

Match #4 does not have a tree associated with their DNA results, but does have a small online tree. It only goes back a couple of generations and does not include any of my known relatives. We share 20.3 cMs on two DNA segments.

Building Out a Family Tree

Match #4’s tree was the smallest and included both dates and places. Since I could easily see the possibility of a quick connection, I started with her tree. I added the people in her tree to my own, using the “Save to Tree” function, and started building out their family trees. Since my maternal ancestry is all eastern Pennsylvania, I paid more attention to those family members who were born in and around that location.

Before too long I reached Emma Caroline (Stout) Hallowell. Emma was born 7 November 1864 and died 24 December 1948 in Montgomery County, Pennsylvania.¹ She was the daughter of Jesse Stout and Amanda Yeakle. Upon seeing the Yeakle surname, I knew I had found the right line.

Yeakle (aka Yeakel, Jäckel, Yeagle) appears in my family tree numerous times. It’s a well-known, Schwenkfelder surname from the Perkiomen region. Luckily, my ties to this community come primarily through one line, making it relatively easy to trace. I also have a resource I can turn to for this family—the Genealogical Record of the Schwenkfelder Families.²

Amanda (Yeakel) Stout, Emma’s mother, was born in November 1833 in Springfield Township, Montgomery County and died 31 October 1904 in Whitemarsh Township, Montgomery County, Pennsylvania.³ She was the daughter of Jacob Schultz and Lydia (Brey) Yeakle.

Since E may match me on the Snyder and/or Wolf family lines, seeing Lydia Brey was a “happy-dance” moment. Jacob Wolf’s wife was Magdalena Brey, daughter of Johann Conrad and Maria Magdalena (Klein) Brey. Lydia (Brey) Yeakle’s father was Philip Brey, a son of Johann Conrad and Maria Magdalena (Klein) Brey. This makes Match #4 and I 6th cousins through the Brey line.

Match #4 also matches me through Jacob Schultz Yeakle. He was born 16 October 1802 and died 30 May 1863 in Springfield Township, Montgomery County, Pennsylvania.⁴ He was the son of Isaac and Regina (Schultz) Yeakle. So, we are also 6th cousins through Andrew and Charlotte (Yeakel) Schultz, Regina’s parents.

Regina (Schultz) Yeakle’s sister Christina married George Krauss. Their daughter Elizabeth Krauss married Joel Wolf, son of Jacob and Magdalena (Brey) Wolf. This couple is my gateway to the Schwenkfelders. Any connection I have to this community can be traced through them.

Since we match on two segments, it’s possible that match #4 and I share both a Brey and a Schultz/Yeakel DNA segment. E and I also share two segments, making it possible that we share both a Wolf/Brey and a Snyder segment. Or it could be Yeakel and Wolf segments as both surnames appear multiple times in Jacob Wolf’s ancestry. Only more analysis will tell.

Conclusions

What does this mean regarding my family connection to E? Based on my identifications of match #4’s ancestry, I think that my case for Judith (Wolf) Snyder being the daughter of Jacob and Magdalena (Brey) Wolf is stronger.

If the relationship is true, E and I would be 4th cousins once removed. According to the Shared CM project the average shared DNA for that relationship is 20 cMs with a range between 0 and 57 cMs. E and I share 24.5—just about average.

I’m still looking for a paper trail to document Jacob’s children. But I think the DNA analysis may be pointing to one of them.

According to Ancestry, I have 363 DNA matches who are 4th cousins or closer. That’s a lot of cousins. Parsing through them all to identify where we match is not a small undertaking. As more people test at Ancestry, that task is unlikely to get any easier.

These cousins and I share between 20 cMs (4th cousin) and 615 cMs (2nd cousin). As you might expect, I’m spending time on the matches with either higher amounts of matching DNA or Shared Ancestor Hints (aka shaking leaves)—not on those with either a severely limited online tree or no tree.

However, sometimes it pays to spend the time building out a match’s family tree.

The Snyder Family

I recently had a new match pop up in my list. According to the link to her profile, she didn’t have a family tree, but I viewed the match anyway. Turns out she did have a tree; she just hadn’t associated it with her DNA test.¹

The tree was small, only going back a few generations, none of whom matched my identifiable family members. But it included one Elizabeth Snyder. Since this person also matched my mother and Snyder is one of her family surnames, I did a little research on Elizabeth.

I was able to locate Elizabeth’s death certificate.² She was born to Charles Snyder and Coletta Zellner in Allentown and died in Lehigh County. I recognized the surnames Snyder and Zellner from my research and, of course, Allentown is right up the road from where my Schneider/Snyders lived in Upper Hanover Township, Montgomery County.

So, I dug around a little and traced Elizabeth’s tree.

Using census records, marriage records, death records, and Find A Grave, I was able to trace Elizabeth’s ancestry back to Samuel Snyder and Judith Wolf of Allentown. In 1850, the family was living in the East End of Allentown.³

Samuel was 48 years-old, a laborer, and his son Charles was 7 years-old. Jonas Snyder, identified in later records as Samuel’s brother, was also living in the household.⁴ Just one family away was the household of Absalom Wolf. Judith Snyder’s maiden name was listed as Wolf on her son’s death certificate.⁵ Could they be related?

Absalom Wolf was most likely the son of Jacob and Magdalena (Brey) Wolf of Upper Hanover Township and Allentown, my four times great grandparents.⁶ Indeed, if you look four households before the Snyders (at bottom of the previous page), you’ll find Jacob and Magdalena and their children Charles and Catharine.⁷

Isn’t it possible that Samuel and Judith were living in the same neighborhood in Allentown as her parents and brother?

I found Jacob Wolf living in multiple locations in the Allentown general area. He was living in Allentown in 1860,⁸ 1850, and most likely in 1840.⁹ In 1830, his household can be found in Upper Hanover Township near his wife Magdalena’s father Conrad Brey.¹⁰ In 1820, he was living in Milford Township¹¹ next door to Daniel Brey, son of John Conrad and Elisabeth (Schneider) Brey,¹² and in 1810 in Upper Hanover.¹³

Both the 1820 and 1830 census include an entry for a female in the appropriate age range to be Judith. Furthermore, in looking at the households on the same page as Jacob in 1830 census, I noticed two households in particular. One was his father-in-law Conrad Bry [Brey], as already mentioned. The other was Catharine Snyder.¹⁴

Yes, that Catharine Snyder! Mother of my 4x great grandfather, Henry Snyder.

Seeing Catharine got me thinking about the children she had with Jacob Schneider. I knew that they had a son Jonas—it’s not a common name in my family lines—and that he was young when Jacob died about 1829, but couldn’t remember if they had a son Samuel. Looking them up in Reunion, I found that Jonas was born after 16 November 1815 and before 19 October 1829. An 1820 birth fits into this range. It also fits with the 1820 census¹⁵ and 1830 census records for the family.

Jacob and Catharine also had a son named Samuel. According to my guesstimates, he was likely born between 1795 and 1800. If I or the census ages are off a little, Samuel, too, could fit right into this family. Since I don’t have firm birth dates or even ages from the Orphan’s Court documents, it’s currently impossible to know for sure based on the information at hand. However, if Samuel was the son of Jacob and Catharine, and Judith was the daughter of Jacob and Magdalena, the fact that both families are shown as living in Upper Hanover in 1830 and appear to be living near each other—only seven households apart—puts Samuel and Judith in the right place to meet prior to their marriage about 1833.

Conclusions

So, I have a bunch of circumstantial evidence that points to Samuel and Judith being the children of Jacob and Catharine (___) Schneider and Jacob and Magdalena (Brey) Wolf.

1. A DNA match of 24.5 cMs on 2 segments between myself and a descendant of Elizabeth Snyder, who also matches my mother
2. A family tree that includes Elizabeth Snyder
3. A death certificate naming Elizabeth’s parents as Charles Snyder and Colette Zellner
4. A death certificate for Charles Snyder naming his parents as Samuel and Julian (Wolf) Snyder—Judith/Judy in other records
5. Census records that place Samuel’s household in the same neighborhood in Allentown as Jacob Wolf and his son Absalom Wolf
6. Census records that name Jonas Snyder, born about 1820, as Samuel’s brother
7. An Orphan’s Court document naming the children of Jacob Schneider of Upper Hanover as: Henry, Elizabeth, Jacob, Catharina, Samuel, Daniel, Michael, John, Sarah, and Jonas, and stating Sarah and Jonas were minors (under 21 years-old) as of 19 October 1829
8. Census records that show both families—Snyder and Wolf—living near each other in 1830
9. Burials for members of both Jacob Wolf’s family and Samuel Snyder’s family in Union-West End Cemetery in Allentown

The facts seems to create a compelling story. I’d feel more confident with evidence that was more direct, but I may have to settle for an argument based on circumstantial evidence. I’ll keep digging into both families to see what else I can come up with.

As I reported  last year in “A Beautiful Circle,” I am a member of both the Philip Hoover and Hannah Thomas circles on AncestryDNA. This means that my DNA matches that of at least two people who have public trees in which both Philip Hoover and Hannah Thomas appear within six generations as a common ancestor. Therefore, my DNA results support my hypothesis—though does not prove—that my three times great grandfather, Christian Hoover, was the their eldest son.

However, my matches from the circle include two cousins who descend from Samuel Hoover and Victoria Walker, Christian’s son, and only one cousin who descends from another child (Sarah) of Philip and Hannah. This cousin also matches two descendants from yet a third child of Philip and Hannah, Margaret, but neither I nor the other Samuel descendants match them.

I decided to see if I could find additional evidence of the connection through my other matches.

Clustering Shared Matches

Blaine Bettinger at The Genetic Genealogist put forward a technique using the Shared Matches tool at Ancestry. By mining the shared matches of your matches, you can look for shared ancestry. Since only my Mom and I have tested, and the Hoovers are on my paternal side, I couldn’t use this technique exactly as explained.¹

However, what I did is not all that different, just not as thorough.

First I identified the Shared Matches between myself and the one Philip and Hannah descendant I match—let’s call them VR. There were ten. I listed them and included the amount of DNA and the number of segments we shared. The amounts ranged from 8.4 cMs to 615 cMs, across between one to twenty-eight segments. These amounts are included in figure A and B (amount/segments). The AncestryDNA Circle members are bold.

Then for each of these people, I listed all of the matches they shared with me. If these shared matches were not also on the list of at least one of the others, I removed them from consideration.

Within the first couple, it was clear that there were going to be additional Hoover cousins on my list who were not members of the AncestryDNA circles. Several of them were easily identified because we had Shared Ancestor Hints. Those I highlighted in green. If I recognized one as being a descendant of an associated family—a Walker, Kinnard or Thomas—I highlighted them another color.

Of the ten I shared with VR, I quickly identified four of them among my Hoover relatives—three as descendants of my two times great grandfather Samuel Hoover (NH, CH, and DC) and one as a descendant of his brother Simon (BC). I also identified another two of the recurrent matches as Hoovers (MP and BL). With some research I was able to identify three more Hoovers—one positive identification (MR) and two possibles (m50 and g65). I highlighted all of those I was sure of in green.

Four of the matches I shared with VR—la, a49, JB and DB—didn’t match these Hoovers. They matched only a group of testers that they shared with each other, myself and V. As none of them had shared ancestor hints, I decided to put them aside for later consideration and further research.

I was also curious to see if any of my known Hoover relatives matched the two other members of Philip and Hannah’s circles: IC and SH. Two of the Hoovers I’d identified (BC and MR) and one of the possibles also matched them (g65).

Putting It Together

So, now I had eleven matches for whom I could identify their descent from Philip and Hannah, and two who I could, I think, place in the appropriate branches. Maybe. The chart below shows this information (click to enlarge).

Based on the AncestryDNA matches, VR and I share DNA with descendants from two of Christian’s children: Samuel and Simon. These descendants and I also share matches that include other descendants of Samuel and one of Reuben’s. VR also matches descendants of Philip and Hannah’s daughter Margaret. Even though I do not match Margaret’s descendants, they match two of my matches who belong to Simon’s line and one of Samuel’s (presumably).

To my mind this is not a fluke.

The next question? Do I share DNA with each of these matches within the expected amount for each relationship?

The matches include relationships as follows:

• NH (615/28) : first cousin once removed
• CH (52/2) : third cousin
• DC (15/3) : third cousin
• BL (36/4) : third cousin
• g65 (40/2) : third cousin—third cousin twice removed?
• MP (52/3) : third cousin once removed
• BC (49/2) : second cousin twice removed
• MR (42/2) : second cousin twice removed
• m50 (27/4) : third cousin once removed?
• VR (8.4/1) : fourth cousin once removed

Based on the Shared cM Project, there are ranges of expected shared DNA for each relationship as shown in the following chart:

As you can see the amount of DNA I share with each cousin is lower than average in some cases, but all fall within the expected ranges for the proposed relationship. This doesn’t prove the relationship, nor does it disprove it. But, as far as I can tell, the numbers do not suggest that any of these presumed relationships are erroneous.

The Limitations

Does this prove that we all inherited our matching DNA from Philip and Hannah, and, therefore, prove our descent? No. Unfortunately, there are several limitations to this approach.

First of all, I was only able to identify my shared matches with those cousins who appeared on my shared match list with VR. I could not identify the matches shared by VR and each of these cousins. I don’t believe the other Hoover cousins I identified would have appeared on VR’s list, but maybe there would have been other’s I could have identified with whom I don’t share DNA.

Furthermore, Ancestry’s Shared Match tool only shows us people who are on both our match list and our match’s match list. The fact that we have matches in common only means that we share DNA somewhere on our chromosomes with the same people. It does not, however, mean that we share the same segments on the same chromosomes with the same people.

For instance, VR and I share both DNA with BC, so she appears on both our match lists. However, BC and I could share DNA inherited from the Kinnard family, while VR and BC may share DNA inherited from the Hoover family or another relative they share who is completely unrelated to me. It would be different DNA and wouldn’t link us genetically, but BC would still be a shared match to both VR and I. There’s no way of knowing based strictly on the data and tools available through Ancestry.

In order to prove the relationships between all these shared matches, I would need segment data for each of them. While I could use FTDNA’s chromosome browser to ensure that I matched each of these people on the same segment of the same chromosome, I could not perform the comparison to ensure that they matched each other in the same location. The only way to prove our shared DNA is inherited from Philip and/or Hannah would be to compare each of us one-on-one to be sure that we share overlapping segments.² To do this, I’d need to use the tools and data on GEDmatch.

So, I guess my next step is to contact my cousins to find out if any of them is willing to upload their results to GEDmatch (it’s free!) for analysis.

I’ve been spending a lot of time—a real lot of time—working with my Ancestry DNA, FTDNA, and GEDmatch results, working through my match lists, compiling data, and, where I can, identifying my most recent common ancestors with various matches. It’s a great deal of work. And most of the time I feel like I’m flailing about, trying to swim in water that’s really too deep for my abilities.

As I was traversing a match’s GEDCOM file searching for a common surname—unsuccessfully, I might add—it occurred to me to wonder why I was doing so. Just what did I hope to gain from all this work?

It’s common online to see folks trying to identify relatives in order to identify their or a relative’s birth family. That’s not my situation. I know who my parents, grandparents, and great grandparents are—and my DNA matches confirm this.

So, why am I spending so much time analyzing my results? What do I get out of it? I took the test for fun. I thought it would be interesting to see my ethnicity results, maybe find some cousins. I’ve already done so much research to build the “paper trail,” I didn’t really consider that it might be useful otherwise.

But recently I’ve been spending significantly more time working with the DNA results, than I have been researching. Maybe it’s time to stop and consider whether that’s a productive use of my time.

What do I want to get out of it?

Here’s that I’ve been able to come up with so far.

1. I want to use genetic genealogy to confirm my research.

Not particularly exciting and rather obvious, but a true and useful goal. My DNA matches have become another piece of source data to add to a proof argument. The fact that I share DNA with other descendants of Philip Hoover and Hannah Thomas of Armstrong County strengthens my assertion that my ancestor Christian Hoover was their eldest son. As more matches come in which share DNA with their descendants—or those of their ancestors—the stronger my argument will grow.

2. I want to “ancestor map my chromosomes.”

Since I know the majority of my ancestry back six generations, and a number of lines back even further, sounds like a simple thing to do, right?  Uh huh. Wouldn’t that be nice? There are actually several factors that complicate this process.

First, in order to find how you and a DNA match are related, you need to compare ancestors. Sometimes it’s as easy as finding a shaking leaf on Ancestry. It’s more common, however, to find no family tree to examine, a private family tree, or a tree with only couple of generations—all marked “private”—with no names or dates. Somewhat less than useful. Furthermore, when you contact matches hoping to collaborate, you get no response.

Second, my ancestors on both sides of my family tree arrived in Pennsylvania, decided it was a good place to live, and then never left. Although this really helps with the document research, it complicates the DNA research. You see, they not only stayed, but they married. Then their descendants intermarried. So, finding how I match that 4th-6th cousin prospect is not always as simple as locating the common surname in a family tree. Occasionally, there are multiple common ancestors. Which one (or more) is the reason for the match?

Which brings us to the next difficulty—Ancestry’s total lack of useful tools to analyze the actual DNA. At most, you can identify the overall amount to DNA you share and the number of segments. Because there is no means of identifying segment or chromosome information; there is no method to triangulate your matches. You can’t determine whether the DNA you share with several matches is actually the same segment on the same chromosome—or that they share that same DNA between them, too.

To actually use Ancestry DNA results for more than just guessing, you need to upload it to GEDmatch or Family Tree DNA so that you can access segment and chromosome data. You can upload to both for free—although to use FTDNA’s tools, you’ll need to pay a small fee. It was $19 when I did it and well worth the price.

Several of my Ancestry matches have already uploaded to both places. Because they’ve done so, I was able to identify the specific segments we share from our common ancestors. It’s cool to see that I likely got about 30 centimorgans (cM) of DNA on chromosome 16 from my 4x great grandfather George Walker of Boggs Township, Centre County, Pennsylvania. A shorter segment on chromosome 16 may have come from his in-laws, Andrew and Catharine Margaret (Fetter Fetzer) Walker.¹

3. Break down some genealogy brick walls.

This one is the trickiest of my three goals. Christian Hoover was a brick wall ancestor for some time. Research, however, pointed to a possible location and family. DNA, so far, has been supporting that conclusion. More on that in another post.

Another brick wall ancestor is Jefferson Force. I believe he was orphaned at a young age, possibly as early as age ten. My research to date has not yielded much. I would like very much to identify his parents. Identifiable DNA matches between other Force descendants to some of Jefferson’s descendants might provide other avenues to explore. Jefferson’s parents would be my 4x great grandparents. This is pushing the limits of usable DNA information as 25% to 50% of 4th cousins share no DNA. Someone of my generation would likely be my 5th cousin, making it even a bit more unlikely a share DNA with any given 5th cousin.

The Plan

So now that I’ve identified some goals for using DNA testing in my genealogy, it’s time to come up with a plan and action items. What exactly do I need to do to accomplish them? More testing will likely be required. Any aunts, uncles, or cousins out there who want to volunteer to take a DNA test?

When I saw the recent announcement that Family Tree DNA was now accepting Ancestry V2 DNA uploads for autosomal transfer, I was excited, to say the least. I’ve been waiting impatiently for this since I did the testing last summer. So, I immediately uploaded my raw DNA. Here are my initial thoughts after trying it out.

Looking through my matches, I was surprised how many I recognized from my match lists from Ancestry. I’ve already identified the common ancestor for some of them. However, there are plenty that I still need to sort through to try to make that determination. As on Ancestry, many do not have pedigree information associated with their account. So, if I’m to determine how we match, I’ll need to make use of common matches.

I found the chromosome browser tool to be pretty cool. It’s one of the reasons I wanted to use Family Tree DNA. I’m a visual person, so a tool like this is really helpful to aid my understanding. Now I need to study up on segment matching so I actually know what I’m doing!

When I checked today, my ethnic origins had finished processing. I know the experts recommend taking ethnicity estimates with a grain of salt (or the whole darn salt shaker), but given how close the Ancestry estimates were to the estimate I made from my pedigree, I wasn’t expecting anything different on FTDNA. Boy, was I surprised!

According to Ancestry, my ethnic makeup looks something like this:

According to Family Tree DNA, it looks more like this:

Surprise!

Here’s a comparison of the three estimates I’ve got in table form.

Region	AncestryDNA	Pedigree	FTDNA
Western Europe	63%	73.72%	9%
Ireland	16%	15.6%	—
Scandinavia	5%	3.12%	58%
Great Britain	4%	1.56%	—
Italy/Greece	2%	—	29%
Iberian Peninsula	4%	—
European Jewish	<1%	—	—
Unknown	—	6.24%	—
Central Asian	4%		3%
Middle Eastern	—	—	2%

Here’s what the maps from the two companies look like:

While the maps look similar in terms of the areas covered (except British Isles, Eastern Europe and Egypt), the ethnicity estimates do not. Why are they so different?

Part of the reason, I’m sure, has to do with the reference panel to which a DNA sample is compared. How the testing company geographically categorizes those references counts a bit, too. Ancestry separates Italy/Greece from the Iberian Peninsula, whereas FTDNA’s My Origins has one group: Southern Europe.

Genetic overlap is also a contributing factor. For instance, those from the British Isles are likely to demonstrate DNA from Scandinavia, Western Europe, and Southern Europe due to the various invasions/population influxes the islands experienced through the centuries. But I would think that at least some of this is accounted for in the process of mapping the various admixtures.

How does that help explain my results? I’m not entirely sure. I only have one ancestor who could be considered to be from Scandinavia (or Germany) going back at least six generations on either my paternal or maternal lines. My one Danish ancestor is  five generations back and would account for only 3.12% of my DNA. So, how do I get 58% of my DNA from Scandinavia? That’s more than half my DNA! Roughly the equivalent amount of DNA contributed to me by one of my parents. I should be able to account for it more recently than five generations back.

The same can be said for Southern Europe. While I can see the possibility of DNA from Southern Europe being present in some of the Swiss/German or Irish ancestors, it’s many generations back. It would likely be diluted with other ethnicities. I can’t really see how it could account for 29% of my DNA—that’s a grandparent!

Nine percent of my DNA from Western/Central Europe seems exceedingly small. Especially when I can count 46 out of 64 ancestors six generations back as most likely being of German or Swiss descent. That’s approximately 72% of my DNA. The difference is almost laughable.

So, I’m a bit at a loss to explain these results.