Login to participate
Register   Lost ID/password?
Louis Kessler's Behold Blog » Blog Entry           prev Prev   Next next

Help Needed for DMT - Thank You! - EAST Part 3 - Sat, 2 Jul 2016

I have the basics of my Double Match Triangulation program working, but before I can release it to the world (as freeware!), I must put it through its paces and test it with some real data and ensure that it will correctly analyze and display the data and relationships.

Since I’ve only DNA tested my 93 year old uncle Harry, and since two people’s Chromosome match files are needed for the program to work on, I cannot do this by myself. So I contacted several of the people listed as matches on my uncle’s Family Finder page at FamilyTreeDNA to see if they would help out with my research.

I was overwhelmed by the enthusiastic response. Everyone, myself included, is looking to find some way to make some sense out of their autosomal matches, and then there’s the potential promise that true triangulation made easy by my DMT program could save loads of time and help us figure out how some of our matches are related.

My uncle’s match list (which is growing daily as FamilyTreeDNA finds new matches) is currently up to 7,865 matches and still only has one confirmed relative.

The one confirmed relative is Joel, who is my 3rd cousin, and my uncle’s 2nd cousin once removed on my uncle (father’s brother’s) mother’s father’s side. Joel is 3rd out of 7,865 on my uncle’s match list with 134.8 cM shared. Joel and I have been communicating for years working with several other cousins on that common side of our families. Joel sent me his chromosome match file.

Then I found Seth, whose ancestral surname was Braunstein (the same as my uncle), whose family originated in a town in Romania less than 100 km from where my uncle’s Braunstein ancestors came from. He didn’t show up prominently in my uncle’s matches until FamilyTreeDNA’s recent algorithm update. Seth moved up from a 5th to remote cousin sharing 127.1 cM to a 2nd to 3rd cousin sharing 130.7 cM. I’m very hopeful we’ll find the connection between Seth and my uncle because we know it will be on both our paternal lines. Seth sent me his chromosome match file.

Another person high up on my uncle’s match list was Erika, listed as a 2nd to 3rd cousin at 160.0 cM. She caught my attention when I was putting all my Pikholz connections together in preparation for my day at the Ontario Genealogical Society Conference where Israel Pickholtz was going to speak. About half the people in Israel’s book: Endogamy, One Family, One People were on my Uncle’s match list. But Erika was the closest of anyone with a Pickholtz connection. I contacted George, Erika’s cousin who was administering her account and is himself listed as a 3rd to 5th cousin at 92.2 cM shared. George sent me both his and Erika’s chromosome match files.

Then there’s a FamilyTreeDNA project for an area of Ukraine that i joined on behalf of my Uncle. Four people from there, Sandy, Barbara (2nd-4th cousin, 102.5 cM), Bruce (2nd – 4th cousin, 97.0 cM) and Mark who have connections to my uncle, sent me their chromosome match files. Barbara and Bruce also each sent me two more of the files they administer. Sandy sent me 23 files in total covering quite a few relatives in her family, many of which are among my uncle’s matches. Sandy has considerable experience in triangulation and has given talks on her analysis using it. I look forward to working with Sandy to help figure out her/our families.

Last but not least is Arnold Chamove who has been a Behold user for almost a year. He and I have had many good talks since then about Behold and what it does and should do. So it was a bit surprising when I found 6 people whose DNA Arnold administers listed in my Uncle’s match lists, the closest of whom is his cousin Roger (2nd – 3rd cousin, 144.2 cM). Arnold has given me access to 23 of the chromosome match files that he administers. It will be fun helping Arnold put his families together and finding out what our connection is.

It is very interesting that I can’t offhand connect yet to any of these 2nd to 4th cousins except for Joel. Most Ashkenazi lines only go back about 5 generations, and due to endogamy, 2nd to 4th cousins can mean 3rd to 6th cousins, even though FamilyTreeDNA says they try to correct for this.

I’ll be taking these 58 Chromosome Match files and use them for testing and to determine how best to analyze, interpret and present the triangulation data.

Most of the chromosome match files are from full Ashkenazi heritage with all its endogamy. These files range from 8 MB to 14 MB in size and the largest have more than 200,000 chromosome segment matches to 8,000 people. Non-Jewish chromosome files I’ve been sent seem to be about one tenth that size.

9 x 8 = 72 combinations both waysAnd the DMT program does Double Match Triangulation, meaning it needs two match files for a comparison. I will do every pair of comparisons. That will be 58 times 57 or 3,306 comparisons both ways. The program takes about 5 seconds per comparison (comparing two files of 200,000 lines each), so once I get the automated selector working, I’ll let it run for several hours to do them all.

There was one person I asked who would not give me his chromosome match file. It wasn’t that he wanted to keep his information private. Au contraire, Meir is a world expert at Y-DNA research, specializing in the Levite line, and he receives hundreds of DNA files from people willing to help. I know he’d be more than willing to help me.

But Meir’s reason was very interesting. He said to me:

The “autosomal soup” is not science. far from it.
It is pseudo science on verge of charlatanism.
Leave me out of this fiasco.

I pressed him further on this, and he told me he’d do an exception for me if I could meet a challenge. Putting aside the known relations, If I could show him how a mere 7 unknowns out of the 7,600 are related to Harry, he’d be willing to participate. So sort of like Sodom and Gomorrah which needed 10, I’ve got to find 7 good people who I can match to. 

I told Meir this is a fair offer. I said I don’t know if I will succeed in identifying 7 relationship paths just using the triangulation information, but I shall try. The rewards of succeeding are just too great to ignore.

It’s going to be fun!

Obviously, 3,306 pairs of test files is enough for me for now. But if you check your FamilyTreeDNA matches and notice that Harry Braunstein is listed as one of your matches, contact me, and I’ll try to include you in my tests. 

Extreme Autosomal Segment Triangulation (EAST) - Part 1
EAST Part 2 - Double Match Triangulation

No Comments Yet

Leave a Comment

You must login to comment.

Login to participate
Register   Lost ID/password?