Monday, September 21, 2015

Another Option for GEDmatch's Lazarus Tool

I have written about the Lazarus tool on GEDmatch as have Blaine Bettinger and others. In fact, there is a Facebook group dedicated to the subject.

Lazarus is a tool which helps you recreate - in full or in part - the genome of someone who cannot himself be tested. Usually because he is long dead.

Lazarus is one of the "Tier 1" tools  which require a contribution to GEDmatch. (They do very important work and the contribution is definitely warranted, even if all you use are the standard tools.)

The idea behind Lazarus is that you look at the target's descendants and see what segments match other known relatives who are not descendants. For instance, if I have segments that match my father's sister or brother, then those segments must have come to me from my father. For a Lazarus of my father, I would be in what they call Group 1 and my aunt and uncle would be in what they call Group 2. My sisters would also be in Group 1 and they would allow us to recreate additional bits of our father's genome.

Adding other relatives of my father to Group 2 - his first and second cousins, for instance - allows us to enlarge his Lazarus kit even more. Unfortunately we do not always have enough kits of these cousins to give us a significant number of matches with Group 1.

Furthermore, in the case of endogamous families, where people are related fairly closely in mutiple directions, this must be done very carefully. I discuss this issue in detail in Chapters Fifteen and Sixteen of my book "ENDOGAMY: One Family, One People" and I after that I show practical uses of several Lazarus kits.

The form at GEDmatch tells us:
Group two is a list of kit numbers for available remaining relatives of the target Lazarus person we wish to deduce the atDNA. That will be brother/sisters, parents and cousins.
The implication here is that Group 2 must not include descendants of the target. This is stated specifically on Facebook and in other discussion groups.

The reason for this exclusion is that the members of Group 1 carry the DNA of both the target and the target's spouse (the other parent of the descendants). You want to be sure, therefore, that no one in Group 2 might share DNA with the target's spouse, DNA which would be mistakenly attributed to the target's genome.

There is, however, an exception. When the target has children from multiple spouses who are not related to one another.

Let us take as an example our forefather, the Biblical Jacob, who had thirteen children from four women. Leah gave Jacob seven children: Reuven, Shimon, Levi, Judah, Issachar, Zevulun and Dinah. Jacob had two children - Joseph and Benjamin - with Leah's sister Rachel. Jacob had four additional sons with his two concubines, Bilhah (Dan and Naftali) and Zilpah (Gad and Asher).

For the purposes of this analysis, I shall ignore the fact that Jacob was a first cousin of Leah and Rachel, in addition to other family connections. I shall also ignore the Rabbinic tradition that Bilhah and Zilpah were sisters of Rachel and Leah.
So let's say that Jacob's family has just returned to Egypt after burying him in his family plot in Hevron and Reuven says "Too bad we didn't do a Family Finder test on Dad while he was alive."

Levi, who was always particuarly conscious of his lineage, says "Well, half the DNA of each of us comes from Dad. We should be able to work with that."

Judah wasn't so sure. "We'd have to figure out a way to see what came to each of us from Dad and what came from our Moms. But we would probably need to phase our DNA using some other relatives."

Joseph, the worldly, practical one, said that he heard that GEDmatch had a new tool called Lazarus (whoever he is) which allows recreation of a dead person's genome based on matches between descendants and other known relatives. The descendants go in Group 1 and the other known relatives go in Group 2.

Dan, whose deaf son Hushim had whacked off the head of Jacob's only brother Esau when he misunderstood all the wild gesturing over the ownership of the burial plot, pointed out that they were not likely to enjoy any cooperation from Esau's kids.

"So," Reuven sighed, "We have only ourselves to work with. And we are all candidates for Group 1. It says that right here in the instructions."

"No so," said the learned, analytical Issacher. "Group 2 have to be people who are not related to the other parent of the descendants in Group 1. Generally that means not descendants of Dad. How about if we put the nine children of Leah and Rachel (who are sisters) in Group 1 and the four sons of the concubines in Group 2. Joseph and Benjamin cannot be in Group 2, because Rachel and Leah share DNA that would then be misattributed to Dad."

Dan, Naftali, Gad and Asher hated being reminded that they were the sons of concubines. When they were young, some of the others teased them by calling them "porcupines." But they couldn't argue with Issachar's logic.

"In fact," Issachar pointed out, "the DNA of the four half-brothers would be even better than the DNA of Esau's disgusting sons or even Uncle Esau himself, because the half-brothers are a full 50% from Dad." The porcupines were proud of that designation.

"Let it be written so" said Judah and Joseph in a rare moment of agreement.

Housekeeping notes
My next two speaking appearances are in Hebrew - 27 October at the IGS Jerusalem branch and 28 October in Carmiel. "ENDOGAMY: One Family, One People" will be available for purchase and signing. After those two events, the Israeli price will go up from NIS 120 to NIS 140 (plus shipping from Israel).

Outside Israel, books and genetic genealogy T shirts and tote bags are available at .


  1. Very interesting concept to recreate the DNA.

    1. It didn't start with GEDmatch. Kitty Cooper ws doing something similar. But they automated it.
      That's good but it means we endogamous folk have to be extra careful.