Recreating a Grandmother’s Genome – Part 2

In my last post (see “Recreating a Grandmother’s Genome – Part 1”), I introduced my grandmother Jane, who died when I was just 8 years old. Although I have only a few memories of my grandmother, I have 25% of her DNA. To explore this rich genetic legacy, I’m trying to recreate as much of my grandmother’s genome as possible using the GEDmatch Tier 1 tool called “Lazarus.”

In the last post we also learned about the new Lazarus tool. In today’s post, we’ll choose what kits to use for my grandmother’s Lazarus kit.

Finding DNA Kits for Lazarus

GROUP 1 – DESCENDANTS ONLY

So GROUP 1 must be descendants of the target Lazarus kit. My grandmother has six children, twelve grandchildren, and eleven great-granchildren. Any of these 29 people are candidates for GROUP 1. Of those 29 people, I’ve tested four of the six children and one of the grandchildren (myself). Yeah, I know, I have more testing to do!

So in my GROUP 1, there will be the following people: four children (two daughters, two sons) of the target Lazarus person. Since my mother is one of the four children being utilized in GROUP 1, I won’t add myself to the group; any DNA that I could add would already be included in my mother’s kit.

GROUP2 – NON-DESCENDANT RELATIVES ONLY

Like most people, GROUP 2 presents the biggest challenge to my project. I need the closest relatives possible. Siblings and parents of the target Lazarus kit would be preferred, but my grandmother’s siblings and parents are, unfortunately, all long deceased.

My grandmother’s closest match is someone I tested myself, my grandmother’s first cousin. However, the first cousin is related to my grandmother on both sides of my grandmother’s family, and thus shares more DNA than a first cousin otherwise would. For example, the first cousin shares between 525 and 700 cM with my grandmother’s four children, suggesting that she could have shared 1050-1400 cM with my grandmother, well above the approximately 850 cM expected for a first cousin.

LAZARUS RUN #1 – Using the four children in GROUP 1 and the first cousin in GROUP 2, I get a Lazarus kit with 1381.5 cM (248,288 SNPs), which is probably very close to the entirety of the DNA that my grandmother and her first cousin shared in common. Interestingly the kit has no DNA from chromosomes 8, 17, or 22, although this is not unexpected; first cousins will often not share DNA on every chromosome.

GEDmatch Lazarus1

GEDmatch Lazarus2

TIP: Don’t forget to try the “DNA File Diagnostic Utility” to review your generated Lazarus kit; it provides a slightly different way of looking at the file although the information is similar to the Lazarus interface.

Image 5

If you forgot to save the Lazarus page when you generated a kit, you can use the “DNA File Diagnostic Utility” to analyze your kit.

Finding Other GROUP 2 Relatives

While I can run a One-to-Many comparison using any of the four children, I won’t be able to tell if the matches are related to my grandmother or to my grandfather. I might have more luck with the double first cousin, as she is related on both sides of the family.

To identify potential cousins, I used the “People who match one or both of 2 kits” tool. As the name suggests, the tool lets you enter two kit numbers and identify: (1) everyone at GEDmatch that matches BOTH kits; and (2) everyone at GEDmatch who matches only ONE kit. — USE THIS WITH EXTREME CAUTION! The danger of this tool is that you’ll identify cousins that might be related via the line you aren’t investigating (e.g., my grandfather). However, my grandmother’s family is from another country and the likelihood of recent overlap is so remote that I wasn’t concerned about this analysis.

I used the “People who match one or both of 2 kits” tool with the following people:

  • Kit 1: My grandmother’s first cousin
  • Kit 2: Each of the children

Using the tool I found kits that match both [First Cousin] and [Child]. The red boxes show kits that are my own, including myself and the other siblings.

3

These other 10 individuals are all possible candidates for my grandmother’s Lazarus kit.

I performed this with each of the four siblings, creating a final list of 26 individuals that are shared between [First Cousin] and one or more of my grandmother’s children. These individuals were all predicted to be between 3-5 generations removed from my grandmother’s children. The amount of DNA shared between my grandmother’s children and these 26 individuals ranging from 10 cM to 350 cMs.

LAZARUS RUN #2 – For the second Lazarus run, I used the four children in GROUP 1, and the following in GROUP 2:

  • The known first cousin; and
  • The 26 individuals above (based on my knowing that they had relatives/ancestors from the area where my grandmother came from).

Using these individuals, I get a Lazarus kit with 1685.5 cM (315,032 SNPs), an increase of 304 cM from Lazarus Run #1. I also gained a few segments on chromosomes 17 and 22.

Image 6

Image 7

Importantly, this technique allowed me to surpass the 1,500 cM threshold for processing and inclusion in One-to-Many comparisons.

A One-to-One compare of the resulting Lazarus file provides a visualization of the segments of DNA in the Lazarus file:

Image 9

Recreating Genomes

Using this method, I was able to recreate genomes for 3 of my 4 grandparents, all above the One-to-Many comparison threshold (1,500 cM minimum). Here’s what I used:

Maternal Grandmother

This is the process described in this and the previous post.

  • GROUP 1 – two sons, two daughters
  • GROUP 2 – double first cousin, 25 top matches using the “People who match one or both of 2 kits” tool
  • OUTCOME – 1685.5 cM

Maternal Grandfather

I have not yet tested any of my maternal grandfather’s close cousins.

  • GROUP 1 – two sons, two daughters
  • GROUP 2 – n/a
  • OUTCOME – 0.0 cM

Paternal Grandfather

This is of special interest to me as the mother of this Lazarus kit was adopted, and I’m trying to use DNA to solve the mystery.

  • GROUP 1 – one son, one daughter
  • GROUP 2 – paternal first cousin, 50 top matches using the “People who match one or both of 2 kits” tool
  • OUTCOME – 1710.6 cM

Paternal Grandmother

You can see that having a sibling makes a huge difference in the reconstruction.

  • GROUP 1 – one son, one daughter
  • GROUP 2 – brother, 50 top matches using the “People who match one or both of 2 kits” tool
  • OUTCOME – 2581.2 cM

So Now What?

The big question now is probably, so what? I’ve created these Lazarus kits, but what can I do with them? We will take a look at this in the next post.

 

 

 

 

 

 

 

 

.

6 Responses

  1. Leif Sommerdal 30 June 2017 / 1:37 am

    Thank you for the inspiration usefull for recreating my fathers DNA
    Did you ever write Part 3?

  2. Geneva Bailey 3 September 2017 / 12:03 pm

    What method or kit would I use to determine if I believe I need to prove a young man is my grandson? I do not have my son’s DNA, but I have whom I believe to be his son? He only shows up as 8 segment match however in phasing I see many matches. My son won’t take a DNA test. I have my ged match number
    And I have the boys, who is now 18 years old. and I have his mother’s GEDmatch number and the man that raised him is evidently not a match.

  3. spmcc 8 August 2018 / 2:25 am

    Can I use the Lazarus kit produced for my gr-grandfather (using 6 of his grandchildren and a bunch of his cousins) in the “People who match one or both of 2 kits” comparing the LL# with each of his grandchildren to find out who is related to my gr-grandfather’s wife? Thank you!

Comments are closed.