Finally! GEDmatch Announces a Monetization Strategy (and a Way to Raise the Dead?)

GEDmatch   Tools for DNA and genealogy researchIf you’re serious about genetic genealogy, you’ve heard of GEDmatch. The third-party site is one of the few ways to compare testing results among the three testing companies. The site

However, since GEDmatch is run by two incredible volunteers (Curtis Rogers and John Olson) with full-time jobs, the site has experienced server issues and downtime. Many have lamented that there was no monetization plan in place, but gave donations in hope that it would help the site grow.

This week, GEDmatch announced a monetization strategy, namely advanced tools that are only available to Tier 1 members at a nominal cost of $10/month:

Basic GEDmatch programs remain free and we plan to keep them this way. Donations help cover the costs associated with running this site, and will provide you with the benefit of using the additional Tier 1 tools for a period of time equal to one month for every $10 donated. You may use the ‘Donate’ button below, for a one-time donation of any amount, or the ‘Join GEDmatch’ button to establish a recurring $10 per month amount.

Once the month’s fees are paid, the following interface will appear:

Tier 1 GEDmatch Tools

A brief description of the tools:

  • GEDmatch DNA Segment Search – “This utility allows you to find other kits with matching chromosome segments”
  • GEDmatch Relationship Calculator – “This utility calculates probable relationship paths based on Autosomal and X-DNA Genetic Distances”
  • Lazarus – “Generate ‘pseudo-DNA kits’ based on segments in common with your matches”
  • Triangulation – “This utility finds people who match you with your top 300 matches as shown in the one-to-many results and below the upper threshold limit that you specify. It then compares those matches against each other.”

I’d like to focus briefly on the Lazarus tool.

Lazarus – Raising Genomes from the Dead

According to the Lazarus page, “[s]egments are included for every combination where a match occurs between a kit in group1 and group2. The resulting are combined to create the final kit.”

I decided to Lazarus my deceased grandmother’s genome using the following samples:

Group 1:

  1. Her son (my father)
  2. Her daughter (my aunt)
  3. Her grandson (me)

Group 2:

  1. Her brother (my great uncle)

The algorithm processed the kits and came up with a list of resulting segments (below is an excerpt only):

GEDmatch Lazarus 1

  • 368110 single allele SNPs were derived for the resulting kit.
  • 93351 bi-allelic SNPs were derived for the resulting kit.
  • 461461 total SNPs were derived for the resulting kit.

The kit was then available for one-to-one comparisons. To give you an idea of how the new kit was created, and how it affects matching and other uses, here are a few comparisons of Chromosome 1 ONLY of my grandmother’s Lazarus’d kit to other kits:

1. Lazarus’d Grandmother to her brother (used to create the kit):

GEDmatch 2

 

Largest segment = 209.6 cM
Total of segments > 7 cM = 2672.2 cM
Estimated number of generations to MRCA = 1.2

2. Lazarus’d Grandmother to her son (used to create the kit):

GEDmatch4

Largest segment = 180.7 cM
Total of segments > 7 cM = 2069.7 cM
Estimated number of generations to MRCA = 1.4

3. Lazarus’d Grandmother to me (used to create the kit):

GEDmatch5Largest segment = 126.0 cM
Total of segments > 7 cM = 1115.5 cM
Estimated number of generations to MRCA = 1.8

4. Lazarus’d Grandmother to an unrelated male (but, importantly, this male is paternally related to the son and daughter used to create this kit):

GEDmatch6

Largest segment = 0.0 cM
Total of segments > 7 cM = 0.0 cM
(1813) No shared DNA segments found

When you compare my grandmother’s kids to her brother, it seems apparent where the large left gap in Chromosome 1 comes from. In other words, that large gap is missing from my grandmother’s kit because it isn’t shared between her one tested brother and her children. To make that clearer, here is the female child (my aunt, used to create my grandmother’s kit) compared to my great-uncle (used to create my grandmother’s kit):

GEDmatch7

and here is the male child (my father, used to create my grandmother’s kit) compared to my great-uncle (used to create my grandmother’s kit):

GEDmatch8

 

But I’m not sure about why there’s a gap to the right in my grandmother’s Chromosome 1. I’m still studying this and trying to figure out how the algorithm is working.

There’s still a lot to learn about Lazarus, please feel free to share your experiences!

 

 

 

 

 

 

 

 

 

.

22 Responses

  1. Jim Barry 21 October 2014 / 6:39 am

    I created a Lazarus file for my deceased grandfather and the results were very illuminating. They showed how recombination resulted in my matching some relatives while my first cousin did not.

    The most interesting comparison was with my closest YDNA match, whose family lived about 10 miles from mine in rural County Cork. Our YDNA results indicate a 50% chance of a common ancestor within 7.5 generations. I do not have an autosomal match with him but my reconstructed grandfather does, with an estimate of 5.1 generations in distance. That’s very compatible with the YDNA result and provides some additional clues about when our common ancestor might have lived.

    And speaking of raising the dead, there is a new video on our project to do testing of ancestral remains in Ireland on YouTube:

    https://www.youtube.com/watch?v=t5znwaINIh4

    We have raised about 70% of the needed funds and welcome contributions. Please contact me at jbarry6899@me.com for further information.

    • Blaine Bettinger 21 October 2014 / 9:33 am

      Jim – did any of the kits you used to create the Lazarus’d grandfather’s kit match your Y-DNA cousin?

      • Jim Barry 21 October 2014 / 11:36 am

        Thanks, Blaine–no, none of the others I used to create the artificial kit were matches to my closest YDNA match. The matching segment was consolidated from other cousins and totaled 12.3 cM.

  2. caith 21 October 2014 / 8:30 am

    Excuse me, but Gedmatch is not charging a nominal cost of $10 per month. It is DONATION of $10 per month. And, it is well worth it. The first day I made a segment match with 4 others.

    • Blaine Bettinger 21 October 2014 / 9:29 am

      Caith – thanks for stopping by! Not sure I understand the intent behind your comment, but I think that’s because I’m not a tax attorney! 😉

      All their existing tools are still free (and you can surely still donate), but if you want access to the Tier 1 tools, you MUST donate $10/month (which in my book is a “charge” but the semantics don’t really matter. In any event, the Tier 1 tools are NOT free and will not appear unless you “donate.”

  3. Barbara Taylor 21 October 2014 / 11:28 am

    The charge is not $10 per month. It is a one month subscription to the new tools for every $10 you donate – with credits applied from January 2013 forward.

    • Blaine Bettinger 21 October 2014 / 11:42 am

      OK everyone, no more comments about “charge” versus “donation versus “subscription,” this is just getting bizarre and they will be deleted.

      No, you do not need to “subscribe”, as that means a recurring charge which is an option but not automatic (there are two options). $10 must exchange hands from you to GEDmatch in order for you to access 1 month of Tier 1 tools. In most of the world that is considered a “charge” to use the tools.

  4. Roberta 21 October 2014 / 1:37 pm

    I tried Lazarus, however, I am uncertain what or who I have created. I tried running it a couple of different ways. My goal was to create data for a maternal grandparent. However, after creating the results and then doing a comparison I wonder if I haven’t created a parent rather than a grandparent. I would appreciate it greatly if someone could give me an opinion.

    I have a test from my mother’s brother, my test, my brother’s daughter, and my sister’s daughter also test results from a second cousin who is on my uncle’s paternal side.

    The first time I ran it – first tier uncle (son) first, then me (granddaughter) – second tier my two nieces.

    Second time – first tier uncle first, then me – second tier the two nieces and the second cousin.

    The results are almost identical when I put them into my results in Genome Mate. If indeed I am creating results for one of my grandparents would this indicate that it is my grandfather? OR because of the way it matches me and my nieces at a slightly greater amount than my uncle would it indicate that these results could be for my deceased mother? My maternal side is Bohemian/Czech on both of her parents side of the tree. Many of the people I match who possibly are from that side of my tree don’t have a clue about their family history back any distance which makes finding tree information to match very difficult. I also have two huge grandfather gaps for my father which is why I am so keen to be able to do some sorting. I have been trying to build trees for some of my matches (enrolled in the current class at dnagedcom, working on the methodology) and feel as though I am chasing my tail trying to figure it all out.

    The result running one to one – with the second cousin in the mix:

    My result: Largest segment = 211.8 cM
    Total of segments > 7 cM = 2996.2 cM
    Estimated number of generations to MRCA = 1.1

    My uncle:Largest segment = 121.3 cM
    Total of segments > 7 cM = 1704.0 cM
    Estimated number of generations to MRCA = 1.5

    Niece 1: Largest segment = 173.4 cM
    Total of segments > 7 cM = 2181.1 cM
    Estimated number of generations to MRCA = 1.4

    Niece 2: Largest segment = 240.2 cM
    Total of segments > 7 cM = 2528.6 cM
    Estimated number of generations to MRCA = 1.3

    Cousin (Grandfathers were brothers):Largest segment = 67.2 cM
    Total of segments > 7 cM = 351.7 cM
    Estimated number of generations to MRCA = 2.7

    I have also run “one to one” on some of my matches from running “one to many” and find that some that I suspected might be related on the Czech side actually do match with the Lazarus number. None are close matches though 4 to 5 generations which gets me back to finding MRCA. The good thing would be IF this is a valid way to approach it is that it will help me sort out my father’s side and I have a better idea of how to approach tree building with his possible line. Any thoughts on this will be sincerely appreciated.

  5. Dan 23 October 2014 / 8:45 am

    Blaine,

    These new tools look great but being rather still an amateur at understanding exactly how I can use them let me ask this question based on my scenario.

    I have no clue who my father is but I have the results of my full brother, identified via atDNA, and a half brother. Can I use these new tools to help build a kit for mine and my full brother’s father? How would I go about doing it?

    Thanks ahead for any help you can offer.
    Dan

    • Dan 23 October 2014 / 2:28 pm

      As a followup to my own question. Here is how I build my Lazarus
      I first did two Match lists who match 1 or 2. One for my own matches against my half brother and then one for my full brother against my half brother. I then entered those into a spreadsheet and removed duplicate kit numbers since the kit in whole would be used in Lazarus. Then I went to Lazarus and entered all the matching kits in the kit list which would match Lazarus’ relatives. I entered myself and my full brother as Lazarus’ descendants. Reduced the matching criteria to Segment size 100, minumum threshold 4cM. The resultant Lazarus build came up as
      137344 single allele SNPs were derived for the resulting kit.
      54771 bi-allelic SNPs were derived for the resulting kit.
      192115 total SNPs were derived for the resulting kit.
      Total cM: 2010.3

      Ran that against myself and got
      Largest segment 35.8 cM
      Segments larger than 7cM=998.2
      MRCA 1.9 generations

      Against my full brother I got
      Largest segment 36 cM
      Segments larger than 7cM=943.7
      MRCA 2 generations

      If nothing else it’s pretty cool to play with.

  6. Marchpane 24 October 2014 / 1:17 am

    I am stumped about this Lazarus thing. I want to do a Lazarus kit for my grandfather, because I have kits for two of his children (by different mothers and three of his grandchildren, but I don’t have any kits of his siblings. Does that mean I can’t do a Lazarus kit?

  7. Kim 20 November 2014 / 4:05 pm

    I found out through DNA testing that the man who I thought was my father for almost 60 years was indeed not my biological father. I have some clues from family and from my ethnicity results about who my bio father was, but he is deceased. My mother is also deceased.

    I ran my Lazarus test using my kit # in group one, and my sister’s (now half-sister’s) kit, plus a collection of some of my top matches (whom I have no idea how I am related to) in group two.

    My Lazarus person matches me 1852 cMs, the largest being 132.36 cMs. Estimated number of generations 1.5.

  8. Kim 20 November 2014 / 4:15 pm

    One more comment regarding Lazarus. Looking through my numbers, it appears that they took all of the kit matches I included, added all the cMs from each chromosome, and came up with the resulting number (in my case 1852 cM). Could this be correct? However, I believe that some of these people who kit #s I included are NOT related to each other.

  9. Rolanda Pettersson 23 November 2014 / 9:43 pm

    Hi Roberta
    My maternal family was also Bohemian/Czech on both sides. My email is lynnepettersson@gmail.com lets talk and exchange kit numbers and surnames.
    Lynne
    Might have a match

  10. Solothurn 8 January 2015 / 7:07 am

    Would Lazarus work to ‘resurrect’ a deceased father if we have both full daughters’ DNA?

    We have the daughters’ mother’s DNA, would this be used in the calculation?

    Thanks

  11. kdspokane 15 March 2015 / 2:56 pm

    Just wondering about using nieces/nephews in group 1. I recognize you aren’t supposed to, but could this help provide a more complete genome, particularly if matches to the nieces/nephews are used in group 2?

  12. Carla 1 May 2015 / 4:52 pm

    Just tried the Lazarus kit today for a great grandfather. In Group 1, I used the only grandchild I could find that would test, Group 2, was the kits for two great grandchildren, and a 2nd great grandchild. There are two men that match his direct male descendants on YDNA but not atDNA. I’m thinking it’s just because of the inheritance pattern. Can I add the atDNA of his Y matches to Group 2?

  13. judy 2 May 2015 / 4:48 pm

    Where/how do you make a donation to GedMatch? Can’t find it anywhere.

  14. Leonard 19 January 2016 / 4:15 pm

    Judy,

    At the very bottom of your GEDMatch.com homepage is a donate section. You can use paypal or credit card online or send a check or money order.

    If you do not wish to use PayPal, you can send your check or money order to GEDmatch, c/o Curtis Rogers at 710 First Avenue South, Lake Worth, FL 33460. Please write your GEDmatch email login on your check.

Leave a Reply

Your email address will not be published. Required fields are marked *