Louis Kessler’s Behold Blog

Update: August 12, 2018:  This notation has been revised.

See my blog post:  BGRN Revised.

I’m leaving this older post here for historical context.

Yeah, I know. BGRN is a horrible acronym. So if you can get to the end of this post and can think of a better thing to call it, I’ll definitely consider your suggestions.

Almost six months ago, I started thinking about A New Notation for DNA Relationships. I’ve now worked out the bugs, and I’m ready to define the notation more formally, as I prepare to implement it in Behold.

Behold’s Genetic Relationship Notation defines a string of characters that represent how person A connects to person B. With this string, you should be able to:

a) Determine the expected amount of DNA shared by the two people, and
b) Describe the relationship in words.

The notation uses the following characters to make up the string:

• X is a female and is the mother of the previous person in the string.
• Y is a male and is the father of the previous person in the string
• U is a person of unknown sex and is the a parent of the previous person in the string
• A is for a non-genetic parent (e.g. adoptive or step-parent)
• (YX) is for the Most Recent Common Ancestors (MRCA) of person A and person B when the MRCAs are both the father and mother (usual case).
• (Y) is for the MRCA of person A and person B when the MRCA is only the father (half-family on the father’s side).
• (X) is for the MRCA of person A and person B when the MRCA is only the mother (half-family on the mother’s side).
• 2 follows an X, Y, U and represents an identical twin. In this case, the MRCA above is not needed nor shown.   
• x is the daughter of the previous person in the string
• y is the son of the previous person in the string
• u is the child of the previous person in the string
• a is for a non-genetic child (e.g. adopted or step-child) of the previous person in the string
• - (a hyphen) is for the spouse or partner of the previous person in the string.

The core rules, for purely genetic relationships, are:

1. The string starts with 1 or more:  X, Y or U’s, the first of which represent person A.
2. These are followed by at most 1 of:  (YX), (Y), (X), or 2.
3. These are followed by 0 or more:  x, y, or u’s.
4. The last character in the string represents person B.
5. A, a and – (hyphen) specify non-genetic relationships (zero DNA match) and are not part of the core rules.

Examples of the notation for genetic relationships and the relationship in words that can be generated from it:

Y = male person
YXY = male person’s mother’s father
YXY(YX)xy = male person’s mother’s fathers’ sister’s son.
U(Y)xxx = person’s paternal half-sister’s daughter’s daughter’s daughter.
Xyy = female person’s son’s son.
XY2x = female person’s father’s identical twin’s daughter.

Here’s example of some common relationships:

X = female person
XX = mother
XXX = grandmother
XXXX = great-grandmother
XX(YX)x = aunt
XX(YX)xx = 1st cousin
XX(YX)xxx = 1st cousin, once removed
XXX(YX)xx = 1st cousin, once removed (the other way)
XX(YX)xxxx = 1st cousin, twice removed
XXXX(YX)xx = 1st cousin, twice removed (the other way)
XXX(YX)x = great-aunt
XXX(YX)xxx = 2nd cousin
XXX(YX)xxxx = 2nd cousin, once removed
XXXX(YX)xxx = 2nd cousin, once removed (the other way)

(Any of the X’s and x’s (except the last x for the first 4 strings above) can be replaced by Y’s or U’s and y’s or u’s)

Hopefully, you’re getting the idea.

Note that a line up to a direct ancestor will not have a MRCA, and neither will a line down to a direct descendent. When the step is across to an identical twin, the MRCA parents are not shown.

I won’t go into the calculation of how much DNA is shared since it’s worthy of another post, but let me say that the expected values can be easily obtained from strings written in Behold Genetic Relationship Notation, BGRN.

Now I would like to extend this notation to handle more than just Genetic relationships and include all possible genealogical relationships. So let’s define the extended rules as:

1. The string starts with an X, Y or U, which represents person A.
2. These are followed by any number of X, Y, U, A, (YX), (Y), (X), 2, x, y, u, a, and - (hyphens).
3. (YX), (Y) and (X) must be followed by x, y, u or a.  Therefore the string cannot end with (YX), (Y) or (X),
4. 2 must follow X, Y, U, A, x, y, u, a or – (hyphen).
5. The same person is never represented more than once in the string.
6. The last character in the string represents person B.

Examples of the extended notation and the relationship in words that can be generated from it:

Y- = Male person’s spouse
Y-(YX)xaY = Male person’s spouse’s sister’s adopted child’s father.
XXXX–x-X- = Female person’s mother’s mother’s mother’s spouse’s daughter’s spouse’s mother’s spouse.
Y(YX)xy(Y)y-(YX)x- = Male person’s sister’s son’s paternal half-brother’s spouse’s brother’s spouse.

So BGRN can handle any relationship, no matter how complicated.

I am interested in hearing any and all comments, criticisms and suggestions.

It was Judy Russell @legalgen on our recent Unlock the Past Genealogy cruise who implored me to get my 93 year old uncle DNA tested while I still can. Judy made me realize that it was necessary.

I have always understood the importance of DNA research as a valuable science to help discover relationships that works hand-in-hand with traditional research. For the past few years, I have been learning everything there is to learn about DNA, I’ve become a member of the ISOGG and keep up with the latest advancements by following a good number of genetic genealogists through their blogs and twitter accounts. Whenever I have questions, I also go to my daughter who finished her University degree in genetics and microbiology who can set me straight.

Over the past 20 years, doing my own genealogy has taken a back seat to my day job, my family, and my development of Behold. But I realized Judy was right. Similar to talking to your oldest relatives about everything they know, doing this was different and time-urgent. My uncle is the last of my living uncles and aunts either my father or mother’s sides. This was my last chance to get a directly connected generation further back that would help me in my future research to isolate my father’s side from my mother’s site once I get around to doing a DNA test of myself and starting to ask other relatives if they would do so.

I was curious as well. My father’s (and my uncle’s) parents both originate from a small region in what is now north eastern Romania. Unless something unexpected happened somewhere along the way, their parentage should be 100% Ashkenazi and that would likely bring into play all the DNA puzzles associated with endogamy.

So I met with my uncle, I got his permission, I ordered the FamilyFinder, Y37 and mtDNA+ tests from FamilyTreeDNA. We waited a few weeks for it to come. My uncle swabbed his cheeks and we sent back the kit for analysis.

Yesterday I got an email from FamilyTreeDNA that my uncle’s autosomal results were available. How many cousins would it find?  I was hoping for at least a few. To be honest, I didn’t expect a lot. How many did it find? Well, would you believe 7,017?

Of those, 126 were suggested as 2nd cousins, 226 as 3rd cousins, 879 as 4th cousins, and the remaining 5,900 as having significant but remote relationships further than 4th cousins. For those who understand what the following means, the average match was 77.6 cM (centimorgans) with a maximum of 168.9 and a mininum of 20.0. The longest matches averaged 10.5 cM (maximum 32.7, minimum 6.9). I do understand that ancestor collapse occurs in endogenous populations and the number of cM for a match may indicate a relationship closer than the true relationship. FamilyTreeDNA does state that they adjust for this: “Beginning on April 21, 2011, we have modified our Family Finder matching algorithm to address this. The changes affect the match list for Ashkenazi Jews. The outcome is calculated Family Finder relationships that more accurately reflect relationships to other Ashkenazi Jews.” What they don’t state anywhere is whether or not they’ve applied this rule to my uncle’s results.

So let’s see what they consider my uncle to be via FamilyTreeDNA’s “my Origins” page. Will there be any surprises?

Sort of what was expected. A bit of middle eastern can originate from the deep roots of the Ashkenazi people who were in the Middle East 2000 years ago. I could see the 2% European being base people who were used for FamilyTreeDNA’s ethnic makeup who had some Ashkenazi ancestry in them but didn’t realize it.

So the next task was to see how many of those 7,017 matches I recognize as people who I know are related on my father’s side.

My first surprise was that the 3rd person listed was a fellow researcher on my father’s mother’s father’s side. I have communicated with him many times in the past number of years and we have shared much information about our common line and determined, but not proved, that we are cousins. He shows up as a suggested 2nd cousin of my uncle, and that we believe that to be true.

My second surprise was that none of the other 7,016 matches, even the others listed as 2nd cousins, were known to be related to my uncle (or me) on my father’s side.

But there were other people among the matches that very much surprised me:

1. Listed 142nd as a 3rd cousin is Brooke Shreier Ganz, a fellow genealogy software developer (of Leafseek) who I met at RootsTech 2014. There were another 5 people in my matches that are submitted with Brooke’s email address. We must be related, but I don’t yet know how.

2.  Listed 537th as a 4th cousin is someone who I’ve shared a lot of information with about our common families. The trouble is that the shared information is about my wife’s family, not my father’s family. I have no idea how he might be related to me.

3. In 1,495th spot is Gary Mokotoff, the publisher of Avotaynu, the International Review of Jewish Genealogy. He is a big name in Jewish Genealogy and knows a few things.

4. In 2,003rd spot was Israel Pickholtz. He is one of the speakers I had planned to see at the Ontario Genealogical Society conference on June 5th.  I’m half way through his book: Endogamy: One Family, One People. At the back of his book, he has 3 pages (2 shown below) of the people who’s DNA was used for his projects. I went through his list and found almost half of them (marked in red below) are in my list of matches, and some of them were listed as high as 2nd and 3rd cousins, so we must somehow connect. Now I’ll have a lot more to talk to Israel about when I see him in Toronto.

5. Finally, last but not least, in 2,364th spot is Lara Diamond, who I was planning to attend two of her talks at OGS on June 5. Interestingly, Lara recently blogged about her finding a connection to Israel’s family. I went to Lara’s blog post on How Endogamy Looks in Practice. Almost all the people Lara lists on that page are in my matches, and 6 of them are shown as 3rd or 4th cousins. Lara and I will also have lots to talk about.

The FamilyTreeDNA listings include ancestral surnames and places as well as ancestral trees for the people who have entered them. Other than the one relative I know of, only about a dozen of the 7,000 matches listed any of the ancestral names and places that I have already researched on my father’s side, so at the moment, I don’t know how I’m related to any of these people.

I will contact the people I mentioned above regarding our connections, and I’m going to see if I can figure out ways to make chromosome matching between 7,000 people a bit easier. This will be a very interesting adventure as I start to sort all this out. This insight could also allow me to develop some more ideas for DNA tools that I can add to Behold. I’ll keep you posted.

As CeCe Moore said in a Roots Tech 2015 interview: “People who might not think they’re going to find anything will be very surprised.”

@OGSConference. I had noticed that the Ontario Genealogical Society was holding their annual Conference in June. The Sunday program had several speakers I really wanted to hear. Lara Diamond will be talking on “Movement Between Towns in Eastern Europe” and “Jewish Genealogical Research in Ukraine”, and Israel Pickholtz will be talking about his successes in his DNA research in his endogamous family, and I’m currently half way through reading his new book: Endogamy: One Family, One People. To top that off, Sunday has a panel discussing “The Future of Genetic Genealogy” with panel members: Elizabeth A. R. Kaegi (moderator), Maurice Gleeson, CeCe Moore, David Pike and Judy Russell. The closing keynote for the day is CeCe Moore with “Lessons from the Cutting Edge” of genetic genealogy.

I didn’t have time in June to attend the full Conference. But with a lineup like that on the Sunday, I just couldn’t pass up that one day of the Conference. And I knew it would be great to meet up with Judy again, who I had spent a few weeks with a few months ago on the 10th Unlock the Past Genealogy cruise. I booked a flight to Toronto on Saturday June 4, the Conference hotel for the one night, registered for the one day of the Conference, and then my flight home to Winnipeg on Sunday night.

The Chair of the Conference, Paul Jones, noticed that I had registered and recognized my name as the author of Behold. He asked if I’d be interested in becoming an exhibitor or giving a pop-up presentation or becoming a sponsor. Due to my tight schedule for the day, becoming an exhibitor or giving a presentation did not work well, but I did like the idea of becoming a sponsor. So I’m proud and excited to announce that:

Behold Genealogy will be sponsoring
the panel discussion: Session 39: 
“The Future of Genetic Genealogy”
on Sunday June 5, 1:15 pm
at the OGS Conference, Toronto, Ontario

So that’s now a day I’m really looking forward to. It will be great talking to Behold and GenSoftReviews users and meeting for the first time many people I know and who know me only from the web.