Monday, March 30, 2015

Matching Sandee

I manage sixty-eight family autosomal kits on GEDmatch, with six more either in the lab or expected there soon. It is no wonder, then, that I get more than a dozen inquiries a week from people who find matches. Many have very little information and others have some meat on them. These range from:
Dear Mr. Pikholz,

WE ARE COUSINS! I have a match with you on GEDmatch. Can you tell me how we are related?
Israel, I have seen your work on Facebook and I know you are very busy. I have matches with twenty-eight of your kits and my sister has thirty-four. Our GEDmatch kit numbers are [redacted] and [redacted]. Here is a list of the best of our matches: [redacted]

My families ancestral surnames are: [redacted]

Can you suggest how we might be related and how I should proceed?
I have a boiler-plate response, to which I add comments as relevant. Here is that response in its current incarnation:
Let me explain who I am and why I am here. That will help you understand why there are things I know about, while other things are way out of my orbit. And why I am not diving into this with you as enthusiastically as you might like..

Most people do genetic genealogy in order to find new relatives, either for their own sake or in order to help move back in time. That is not what I am doing. My main genealogy interest is the single-surname Pikholz Project, which has as its mission to identify and reconnect all Pikholz descendants.

The Pikholz families divide into two groups, both from east Galicia, one (mine) in Skalat and one in Rozdol. If the two are connected it involves people who lived at least two hundred years ago. We have about eighteen Pikholz families from Skalat, some going back to the edge of the pre-surname period and a few where we dead-end more recently. I am trying to use DNA to see how they fit together. I have a number of successes under my belt and several theories that should be testable when the newest batch of results comes in.

Among the Skalaters alone, I have over forty people who have done Family Finder and more on order.

So you see that I have limited interest and resources to get back much further right now, even though there are a few dozen non-Pikholz who match thirty or more of my people, almost surely from the pre-surname period. (I can get a dozen inquiries like yours in a typical week.)
I don't mind answering specific questions, such as "Is UB related to Sam", but if you want me to do the work (the only guarantee is my best effort), I will have to turn the meter on. (I am a professional genealogist.
In general, chromosome browser on GEDmatch is the best way to go to find patterns.
Some never reply. Some thank me and go on their way. Some expect me to do the work for them, which accounts for the bit about turning on the meter.

The more serious inquirers actually do some work, but invariably their results are all over the genealogical map. If they match any of the dozen descendants of my Pikholz great-grandparents, chances are they match at least eight or nine. They usually match a few of the descendants of Nachman Pikholz and Peretz Pikholz and Moshe Hersch Pikholz and Mordecai Pikholz and sometimes Uncle Selig Pikholz. They often match a few of the dozen Rozdol Pikholz descendants or my Kwoczka cousins or the two first and four second cousins on my mother's side or the two on my paternal grandmother's side.

Or my mother-in-law.

But because they are all over the map, it is very hard to say anything useful. Another untraceable relative from the period before records and surnames, with a common ancestor in the early or mid-1700s at best.

Then early last week there was Sandee.
Hi Israel,

I am on ancestry, Family Tree DNA, and GED match.  You and I are 4th cousins, according to ftDNA.  More interestingly,  I have matches of varying closeness to 29 of the people you manage on GED.  Now I am curious.  How can this be?  I have looked at your family tree and nothing jumps out at me.  I also have tested my mtDNA, and apparently that is not where we relate.  My family is primarily from Ukraine, Poland and Romania.  

I did compare in the chromosome browser.  I wish I had a better understanding of what it means.  

My name is ... Sandee ....  I would be interested in your take on this.  I hope you will be kind enough to respond.  My email is [redacted].  

My best,

My boilerplate response did not scare her off and in fact she went to work on the chromosome browser and although it took a few days, I was finally able to see her results - which she sent me as a pdf file. My first impression was that she matched well in Skalat, but not everyone. Her results seemed more focused. She also sent me her ancestral surnames, which were not familiar to me and did not line up geographically with my people.

I asked her to run it again, but leaving out the Rozdolers and my mother's family. And my mother-in-law. That left mostly the Skalaters. When I saw those newer, cleaner results, I suggested we speak. It seemed more efficient than writing it all out, with explanations and discussion.

We went through them together chromosome by chromosome. We pretty much ignored the matches that were smaller than 5 cM, though it is hard to ignore a column of six or seven of them all in the same place.  We looked at matches of 5-10 cM only when we had a few on the same segments. And we took note of the segments that were larger than 10 cM. 

Sandee's matches were indeed well-focused. There were very few matches with descendants of Old Nachman or Peretz, none with even two on matching segments. There was no trace of the descendants of Moshe Hersch except Leonora. Nothing either from the two great-great-grandchildren of Uncle Selig. A few matches of over 10 cM with the second cousins Craig and Ron, but never together. Very little of Vladimir or Joyce and neither of the Kwoczka cousins. And of my known Pikholz third cousins, Elaine was there and Joe was not.

But Sandee does have many matches with my sisters and me. And alot with Aunt Betty, but not with Uncle Bob. Quite a bit with Herb and some with Dalia, Lloyd, JudyT and Sam's daughter. And as I mentioned, Leonora. And some with my second cousins Terry, Rhoda, Marty and especially Lee - but less than I would have thought. These results had "Descendants of Mordecai and Taube" written all over them, but not only.

Aunt Betty, my sisters and me, without Terry, Rhoda and Marty looks like my grandmother's side, which includes Lee as well. It looks like Sandee's connection with us has to do with my grandmother's Slovakian-Hungarian side, in addition to a Galician connection through Mordecai or Taube. And since Sandee's matches include Cousin Shabtai and not Cousin Fred, that seems to point to my grandmother's maternal (Hungarian) side.

The matches with Herb are interesting, especially where Sandee matches him and Dalia together. There seems to be a connection somewhere between Herb's father's family and someone on Dalia's other side. Sandee could fit in there too, especially since Elaine's great-grandfather and Herb's grandfather are brothers.

It was a well-spent half hour on the phone. It is normal to expect that distant connections of this sort involve multiple common ancestors, but I have not seen someone whose results are so focused, particularly so heavily weighted to one of the Pikholz lines. So far it doesn't mean anything. We have not done proper triangulation. And until Sandee can get some of her cousins to test, we cannot know where we should be looking for a paper trail on her side.

But I really like where this is going and especially how it is getting there.

Housekeeping notes - literally, for a change
NOTICE: We leave Egypt this weekend. May you all find it real and meaningful. And a joyous occasion.

Tuesday, March 24, 2015

A Day of Memory, A Day of Joy

This Thursday the sixth of Nisan is the 119th yahrzeit of my great-grandmother Etta Bryna Rosenbloom. (I have discussed her here, here, here and here.)

In brief, my great-grandmother died probably in her late thirties and we nothing about her except that her father was Yehudah and that he was a Levi. In particular, we do not know her maiden name or where she came from.

One of those links above discusses a fairly close conection I made with Deborah Sirotkin Butler via MtDNA, which looks at the maternal line. But nothing came of it.

Then this:
 and in particular his most distant maternal-line ancestor:

Yenta Bryna. Sounds an awful lot like Etta Bryna.

So how close is this man? He is zero genetic distance from me, but mitochondrial DNA mutates very slowly, so our common ancestor could be five hundred years ago.

Well, it says "FF," so he has done a Family Finder. Let's do that search.
Third cousin - Fifth cousin. 91.81 cM in common, longest block is 12.72 cM. Well fourth cousins means that my Etta Bryna and his Yenta Bryna could be first cousins. That would be fabulous.

I looked at the chromosomes. There were some matches between this new fellow B and us - us being my sisters and me, my first cousins Kay and Leonard and our second cousins Sam and Beth, all of whom I discussed about a month ago. The matches did not triangulate well - not on FTDNA and not on GEDmatch.

I also compared chromosomes with B and Deborah Sirotkin Butler. She too shows up as a perfect MtDNA match and a third-fifth cousin Family Finder match. But I do not have any matches with both of then, save a few very small segments.

I spoke to Leonid, the cousin of B who manages his family kits. He refers to his ancestor alternately as Yetta Bryna and Yenta Bryna and tells me she was born in 1868 in Lodz and died in 1928 in New York. That would make her about ten years younger than our Etta Bryna. We have no way to bridge the gap between Lodz and Borisov. Leonid has five kits that match B, so perhaps I can learn something by looking at all five compared to ours.

In any case, without some documents to take us back another generation, the DNA is not going to get us very far.

So all we can do today is remember great-grandmother Etta Bryna bat Yehudah HaLevi.

Actually, that is not quite all we can do. We can rejoice with cousin Baer, Sam's son, who is to be married to Dina, Wednesday evening, 6 Nisan, on the anniversary of his great-great--grandmother Etta Bryna's death. The wedding is in far off Australia, but the couple will be living just outside Jerusalem where Baer is studying in yeshiva. Dina's birthday is also 6 Nisan.

Mazal tov to the couple and the families.

Wednesday, March 18, 2015

Imperfect Triangulation

While examining the autosomal DNA results of Ira and Steve on GEDmatch, a curious thing turned up. Ira and Steve are second cousins, great-grandsons if Isak Pikholz of Rozdol and his wife Toby Blum.

In the course of my analysis, I saw the following on chromosome 7.
Ira and Steve have a match of just over 66 cM, as represented by the orange bars.

Below that, each has a green bar representing a match of about 13 cM that each has with Esther. But Esther's matches are not identical. Her match with Steve is 13.2114 cM, while her match with Ira is 13.3094 cM. Close but no cigar. It fails the triangulation test.

Let's look at an analogy. Ira and Steve sit down to dinner together for sixty-six minutes. Esther comes to join them for part of the meal, about thirteen minutes worth. But since Steve and Ira are together, Esther cannot spend even one second more with one than with the other. Her matches with them must be identical to the last decimal.

It is possible in theory that she matches both of them on their other sides - not the Pikholz side where they match each other, but Ira's mother and Steve's father. But it would be too much of a coincidence for those matches to be so similar.

In order to confirm the GEDmatch numbers, I did a chromosome browser for Esther.
Esther's matches with Steve and Ira are the same as theirs with her, so we see no internal inconsistency.

So just for fun, I did the same chromosome browser on FTDNA.

The bar graphs look just like the GEDmatch chromosome browser. But the numbers are slightly different. Particularly the 13 cM segments in blue. Esther matches both Steve and Ira from position 67,711,326 to 80,123,786, for a total of 13.04 cM.

GEDmatch is much easier to work with when comparing more than five kits. Some say that it is more accurate. It may be, much of the time. But not this time. This time FTDNA's numbers work and the GEDmatch numbers do not. Sometimes these tiny variations do not matter. In this case, it's the difference between possible and impossible.

Thursday, March 12, 2015

Lazerus Kits For My Mother

The kits I was able to produce for my father using the Lazerus tool at GEDmatch have been easy. I have four descendants (one more is stuck in the lab in Houston) for Group 1 and various combinations of Group 2, which allow me to weed out most of the endogamous relationships. But no matter how I have done it, I get kits with at least 3400 cM.

Considering that GEDmatch defines success by "Batch processing will be performed if resulting kit achieves required threshold of 1500 cM," my father's kits are excellent.

My mother is a different matter. Mother has the same four descendants for Group 1, but candidates for her Group 2 are in short supply. My father has a living sister and brother - my mother has none. I have two first cousins on my mother's side - a son of my uncle and a daughter of my aunt - who qualify, plus two second cousins on either side, though one of those four, Beth, is not yet on GEDmatch.

So for now, I went with what I had. All the tests are FTDNA's Family Finder.

Everyone in the bottom row has done a Family Finder test.

I decided to create three Lazerus kits for my other - one with the five cousins, one with just my two first cousins and one with Kay alone.

The kit with four of Mother's children in Group 1 and five of my cousins in Group 2 gave my mother 3137.4 cM.

But that is problematic. All of my mother's grandparents are from the same general area. Israel David is from Borisov, in what is now Belarus. Etta Bryna is likely from Borisov or nearby. So is Aunt Mary's husband.

Chana Kugel is from Pleschenitsy, now in Belarus, and Hirsch was born in Dolginov, just across the border in Lithuania. Uncle Hymen's wife is from Schedrin, a hundred miles from Borisov. So there is a high probability of a significant amount of overlapping ancestry.

So I cut back to my first cousins, eliminating any endogamy from my second cousins' other grandparents. The size of my mother's kit decreased by twelve percent, to 2766.4 cM. I thought it would shrink more. Some of that decrease is real and some is because endogamy had created a larger kit to begin with.

However, I was able to eliminate the endogamy entirely. Kay's father has no Jewish DNA, so if I used her alone in Group 2, there would be no endogamy at all. The results here surprised me even more. This Lazerus kit had 2188.8 cM. That's nearly eighty percent of the kit that was made from Kay and Leonard together in group 2.

Finally, for sport, I did a fourth kit using Leonard alone in group 2. That kit had 1527.6 cM. Kay's match with us is 43% greater than Leonard's, even with any endogamous effect that Leonard's mother might have.

Note that all these kits, including those with one person in Group 2, are above the threshold that GEDmatch sets for batch processing.

Blaine Bettinger who blogs at The Genetic Genealogist, is doing a study on the amount of shared DNA in known relationships. If you haven't participated, you should. I imagine the findings above will interest him.

Sunday, March 8, 2015

Sam's Daughter and the (Isak) Josef - Mordecai Complex

Early in our research on the Pikholz families, Jacob Laor obtained a copy of the marriage record of Berl and Dwojre Pikholz, with the help of Alex Dunai. The document was from 1887, but was obvously just a formality, as the couple had been having children together since at least the 1860s.

According to the marriage record, Berl was born in 1816 to Josef and Rojse Pikholz and Dwojre (or as we spell it today, Devorah) was born in 1837 to Motie (=Mordecai) and Taube Pikholz. Everyone was from Skalat.

Is the course of our subsequent research, we learned that Josef was actually Isak Josef, who died in 1862 at age 78 and Mordecai and Taube were born 1805 and 1802, based on their death records. I never found anything else about Rojse, but grandchildren were named for her before Josef's death.

The family structure appeared thus:
Note that Mordecai has one grandson called (Isak) Josef and another named Josef Mordecai, neither of whom seems to be a descendant of (Isak) Josef the elder.

Everything pointed to a fairly close relationship between the two couples, aside from the marriage between their children Devorah and Berl. Twenty-one years separate Josef and Mordecai, enough that they could have been father and son, but not so much that they could not have been brothers. They could also have been uncle and nephew or first cousins or more distantly related. For that matter, the wives could have been sisters.

The other contemporaneous Pikholz men in Skalat were Nachman (1795-1865), Berl (~1789-1877) and Leib (~1780-1844). And of course my own great-great-grandfather Isak Fischel born maybe 1820. Much of Nachman's family is documented to the present. Berl is probably the father of Peretz Pikholz (1820-1873), who has living descendants from all of his five known children. All we know about Leib is from his death record.

We have well-developed lines from Mordecai's children Chaje Cirl, Aryeh Leib and Chaim Yaakov, less so from Henie. Izak Josef's children include my great-great-grandmother Rivka Feige, her brother Selig and very likely Moshe Hersch about whom I have written several times. From Berl and Devorah's three or four died in childhood, one has living descendants and descendants of the others appear to have been murdered in the Holocaust.

Sam's Daughter
There is another family that fits in here someplace. Simon Pikholz was married to Dwojre Waltuch who died in 1861 at age twenty-three leaving two daughters. We have information on descendants of one of those daughters, Lea Pfeffer. Simon then married his late wife's sister Chana Waltuch, with whom he had a number of children, four of whom went to the United States and have living descendants. Chana also went to the US, but Simon did not so we assume he died in Skalat in the 1880s.

Simon had a son and a grandson named Mordecai and at least one granddaughter Taube, so he may well have been the son of Mordecai and Taube. Simon also has a grandson named Shalom (Sam) which is also the name of the great-grandfather of Charles and Leonora from the Moshe Hersch line. That Sam/Shalom assisted Charles' parents in immigrating to the US in 1939 and is listed on their papers as a distant cousin.

Without jumping to conclusions, the more detailed family structure looks like this, where some of the conclusions are based on DNA.

The matches within this group are extensive, to put it mildly. Everyone pretty much matches everyone else. They do not have the same level of matches with the  descendants of Nachman or Peretz. The marriage of Berl and Devorah comes from the two families as does the marriage of Leonora's great-grandparents Isak Josef and Gittel. My own great-grandparents represent a mixture of the DNA from the same two families. And if we make the easy assumption that Simon and Szulim are brothers, sons of Mordecai, then the marriage of Szulim and Sara is another between the families.

Leonora, Judy, Charles and my own family are all clearly products of both Isak Josef and Mordecai or his related DNA.

But the reasons for matches between Nan, Jane and Selig's descendants, on one hand, and Dalia and Lloyd on the other, are less easy to explain. True, we are dealing with the general phenomenon of endogamy, and true as well Mordecai and Isak Josef are almost certainly from a recent common ancestor. But those apply to the descendants of Peretz and Nachman as well, and the matches with them appear weaker than within the Isak Josef-Mordecai complex.

So now we have the matches for Sam's Daughter, the great-granddaughter of Simon. Here are the first twenty-eight, up to the classification "fourth cousin - remote cousin."
I added a column to note the family of descent for each match
The vast majority of Sam's Daughter's matches are from the Mordecai-Isak Josef group, even though Simon seems to have a path to only one of them. As I said above, there are a few possibilities. But from looking at this data, including the individual chromosomes, over the last few weeks and discussing some of it with other people, I am thinking that there may be some other possibilities.

For instance, perhaps Lea Waltuch, the mother of Simon's two wives, was a Pikholz, maybe even the daughter of Isak Josef. That would give Simon's children paths to both Mordecai and Isak Josef. We cannot prove that, of course, even if it is true. All we know about Lea Waltuch is that there is a granddaughter named for her before 1860, so she died not live to old age.

But there is another idea I like better. As I write above, twenty-one years separate Isak Josef and Mordecai, enough that they could have been father and son, but not so much that they could not have been brothers. They could also have been uncle and nephew or first cousins or more distantly related. But what about Mordecai's wife Taube? Maybe she is the daughter of Isak Josef. That would leave any possible relationship between Isak Josef and Mordecai in play, except father and son.

Taube was born when Isak Josef was only about eighteen, but that does not have to be a problem. After all, Mordecai's daughter Chaje Cirl was born when Mordecai was only seventeen.

Taube was considerably older than Isak Josef's known children Berl, Rivka Feige and Selig, but maybe there were others we don't know or others who died in childhood. Or Taube was from an earlier wife of Isak Josef, not Rojse.

If Taube is Josef's daughter, it would certainly explain why Taube's children Chaje Cirl and Aryeh Leib had sons named Josef. That's a question that's been in the back of my mind for many years, and has never been properly addressed.

There is probably no way to prove any of this, but it does execrcise the mind on all matters of genealogy, and that too is useful. It also gives me another opportunity to say that I am happy so many names in that complicated chart are in red, indicating that they have tested - but I'd still like to get more of Simon's descendants to participate in the testing project.

Sunday, March 1, 2015

Getting the Y-Chromosome Right

The Rozdol Male-Lines
It is axiomatic that all the Pikholz descendants from Rozdol are from a single couple, Pinchas and Sara Rivka, whose youngest known child was born in 1818.

I never expected the Y (male-line) DNA to match the three identical lines from Skalat because I have been assuming that if the families in the two towns are related, it is through Sara Rivka, not Pinchas. But at least the Y-DNA of the male-lines from Rozdol should all be identical, unless there was a mutation in the last two hundred years.

Of the nine Rozdol lines, one is a maternal line back to the 1830s, two have no male line descendants, two (Robert and Gadi) have taken thirty-seven market Y tests, one (Micha) took a twelve marker Y test and three have not agreed to test, despite being asked multiple times.

The problem has been that Robert (who is traceable to Pinchas and Sara Rivka as a fourth-great-grandson) and Gadi (traceable to his second-great grandfather David ben Isak and Feige, who was born about 1823) do not have matching DNA. It is possible that the connection between Gadi's David and the original couple Pinchas and Sara Rivka is through his mother Feige rather than his father Isak, but it did not sit well with me.

At the time of the recent sale on DNA tests, when more than twenty Pikholz descendants ordered tests (all but one from Skalat), I decided that we needed to upgrade Micha's test from twelve to thirty-seven markers so we could see how he compares to the other two.

In the meantime I had a closer look at Robert and Gadi.

They were identical except in two markers - DYS437 and DYS438. In the first Gadi is 14 and Robert is 11. In the other, they are reversed. In each case, there is a difference of three so together we are looking a difference of six. Six mutations in two hundred years on those markers is highly unlikely and the numbers are too symmetric for my taste. After a Facebook discussion with some of my colleagues, I became convinced that in one of the two cases, the results were reversed.

A few days ago, Micha's results came in for markers 13-37 and he is a perfect match for Gadi.

I wrote to the always reliable Janine at Family Tree DNA and asked her if she could check Robert's results to see if perhaps they had reversed DYS437 and DYS438. She did and I was right. The results that we have for all three are now identical and will be corrected to haplogroup R-M269. This is the same as our Skalat haplogroup, even though the numbers themselves are very different.

There is a moral here, of course. Sometimes the testing companies make mistakes. When you make good case, they are willing to check their work. But no one will find this kind of thing for you. You have to have enough information to make the case and you must pay enough attention to the details to realize that there is a case to be made.

I'd still like to get those other three Rozdol families to do Y tests. And all six who have not tested to do Family Finders.

And speaking of mutations
Y-37 results came in a couple of weeks ago for a member of another of my families.

There are no perfect matches here, but the 155 matches that FTDNA counts include eleven at a genetic distance of one and thirty-one at a genetic distance of two. Some of those are marked "FF" to indicate that they have also done Family Finder tests. Genetic distance refers to the number of mutations since the most recent common ancestor.

My strategy for following up with these kinds of matches is to see who has both a close Y match and a Family Finder match of "third cousin-fifth cousin" or closer. I have seen experts who say that the Y chromosome has mutations every four or five generations on average, but this can vary, depending on where the mutation occurs and on the vagaries of DNA inheritance. After all, in both the Skalat and the Rozdol Pikholz families, all three DNA tests show no mutations at all in the last two hundred years.

Following up Y matches is more than a bit of a crap-shoot, so I generally think in terms of a genetic distance of one being worth looking at and anything further being too low-probability to bother with unless there is a very good Family Finder match.

But in looking at these particular results I realized that I was missing a point of logic. I am willing to look at a match with a genetic difference of one because the most recent common ancestor might well be four-or five (or less if I am lucky) in the past. But a genetic distance of two does not have to mean double that. The second mutation could be in  the other person's line.

Let's say that Reuven and Shimon have a common ancestor and we figure an average of one mutation every five generations. If they have a genetic difference of one, that is one mutation, that would mean they are likely about fourth cousins because there was one mutation during five generations. But if there is a genetic disrtance of two, that does not mean that an average of ten generations had passed since the common ancestor. It is just as likely that there was one mutation in Reuven's line and one in Shimon's. In that case, the common ancestor might be five generations back, yet the descendants themselves would have a genetic difference of two.

I tested the reaction to that theory in the ISOGG Facebook group, and a brief but spirited discussion ensued. Some agreed that I was making sense. Some not so much. There are always some in that second group. Some of the people I most respect thought I was correct. That's good enough for me. In any case, we are talking about probability, not conclusive fact.

Housekeeping notes
Five Pikholz Family Finder results came in over the weekend. It will take time to analyze, but it appears that two have excellent results, two are about what I expected and one is very odd, having no Pikholz matches in her first fifty-five. (She has lots of known relatives.)