Playing with Phased Grandparent DNA at Gedmatch

Some kind person worked on my DNA to phase it to my four grandparents and uploaded the results to Gedmatch. One of those 4 grandparents was my my maternal grandmother Frazer side. This is exciting territory as this is an area of genetic genealogy that I haven’t looked at before in over 500 Blogs that I have written.

My Top Frazer Match at Gedmatch

I can run this kit which would be equivalent to my siblings’ Frazer matches at Gedmatch. My top match at Gedmatch is Tony. Let’s see if that match pans out. The phasing was not perfect, so there could be some errors.  The largest match with Tony is 193 cM, but when I run the details, it is only a small match:

My guess is that other matches with large largest segments are not correct. My first real match appears to be my second cousin once removed Paul. Here is what the match with my composite Frazer kit looks like:

Matches in Common with Paul

I ran matches in common with Paul and the composite Frazer Kit and then put those matches in a matrix:

I recognize at least half of these matches.

Painting My Frazer Matches

It occurs to me that I can paint my Frazer matches. As this DNA painting represents my grandmother, I’ll say that the person is female:

The first person on the match list under Paul is rosco. This is actually Keith:

Keith is in the last green box on the right next to Paul. I put that Keith is on my grandmother’s paternal side:

DNA Painter has my grandmother <1% painted. The next question is how closely should I ‘paint’ my grandmother’s DNA? In the past I have only gone to grandparent level. I think I’ll change and go to 1st cousin level. A first cousin shares two grandparents.

Adding Paul

Paul would be a first cousin once removed to my Frazer grandmother:

This gets my grandmother up to about 2% painted. I next added Emily. Emily’s common ancestors with mine are the same as Paul’s: George Frazer and Margaret McMaster.

Adding Gladys

Gladys’ common ancestors go back a generation to James Frazer and Violet Frazer:

Here, I changed the dark green to a lighter green, so it wouldn’t blot out the older DNA. So far, I have been only painting the Frazer paternal side. The maternal side is Clarke and I know less about Clarke genealogy than Frazer genealogy.

Painting Michael: Common Ancestor Richard Frazer

Notice Chromosome 1. The blue segment shows as all Richard Frazer. However, this is how it breaks out:

That means that the green segment is really Frazer and not McMaster. The Orange segment under the blue is Violet Frazer as she was the daughter of Richard Frazer.

Clarke DNA

My grandmother’s mother was Margaret Clarke. However, Margaret died when my grandmother was young.  I have a few fairly good DNA matches on the Clarke side, but the best matches are not at Gedmatch. Stephen is probably the closest match. Here is Stephen:

Stephen is also related on the McMaster side which confuses things. However, the closer match is on the Clarke side.

Now my grandmother is about 6% painted. Yellow is her first maternal DNA. I haven’t looked at X Chromosome matches as those are treated separately at Gedmatch.

Next, I looked for matches in common with Stephen and my phased Frazer kit:

The problem is that some of these matches are the same that were in common with my cousin Paul. Paul has no Clarke relatives. That means that this comparison is probably picking up the McMaster connection also. However, it may be possible to tease the two apart.

Grandmother Frazer AutoCluster

I thought that I had done these before, but perhaps not. At standard Gedmatch defaults, I get 19 clusters for my Frazer Grandmother:

There are a narrow range of DNA matches. They range from about 19 to 23 cM. I recognize matches from the last three clusters. Cluster 17 has Jonathan. Our common ancestor goes back to about 1690, but I have a feeling that there is a more recent match – probably on the James Line. James Frazer was born around 1717. Here is where Jonathan matches:

There appears to be a small overlap between Paul and Jonathan.

Pat and Bill are in Cluster 18. Here is how Pat and Bill fit in:

Our common ancestors are James Frazer and Violet Frazer.

In Cluster 19, I recognize Marilee who is related on one of my Frazer lines also.

I changed the limits to between 22 and 250 cM and got this autocluster:

In this AutoCluster reiteration, Clusters 11-14 are the Frazer Clusters. The good news is that I can identify 4 clusters. The bad news is that there are 16 clusters which I cannot identify. Actually, Jonathan is in Cluster 2, so that is one more Frazer cluster that I am aware of. However, the match for this Frazer Cluster ois probably through my McMaster side:

This is the ancestry of my 2nd great-grandmother Margaret Frazer. Turns out that she had a Frazer ancestor that went to a common ancestor with Jonathan. They were James Frazer and Katherine Fitzgerald, born in the first half of the 1700’s.

Barry is in Cluster 11. This is how I think I am related to Barry:

For some reason, I don’t see Barry on my DNA Painter profile. I’ll add him in:

This shows that Barry overlaps with Michael who I match on the Richard Frazer Line. Richard was a brother of Philip. We are showing we match on the common ancestor of Archibald Frazer who was born about 1720. Some of these dates are relative. I have a chart showing Archibald born in 1743. Also many charts have Philip and Richard born earlier than the 1770’s.

Charles in Cluster 12

Charles is in Cluster 12 with Shelly and Martha. Charles has not been on my radar before. His tree is helpful in that he has a shamrock for his Irish ancestors:

The red symbol indicates Scotland. Shelly has a match in common with Gladys. I’m related to Gladys on my Frazer side. Martha from Cluster 12 is here:

One guess is that Charles could be related from this Philip tree or from the wife of Philip.

Cluster 13

I recognize three out of four in Cluster 13. They are Jane, Doreen, and Susan. Here is how they match my family by genealogy:

These families go back to Richard Frazer. The one I don’t recognize if Elizabeth. I can’t figure out how she fits in.

One Last Frazer Cluster Between 23 and 250 cM

By just shifting the lower number up to 23 cM, I go from 20 clusters to 13:

Now my Frazer Clusters are 1 and 2. Cluster 2 is interesting, beause it includes both my Frazer and McMaster Lines. Those are both double lines because two Frazer cousins married in that Cluster and two McMaster cousins married. The 145 cM match has a common ancestor with my family of James McMaster and Fanny McMaster. They had Margaret McMaster who was my 2nd great-grandmother. That match matches my closer matches in the cluster – those who descend from Margaret but not my more distant Frazer relatives. The only one in the green Cluster 2 who I don’t recognize is Nicolas and I have written to him. He appears to be a fairly close relative of Emily who my Frazer kit matches at 125 cM.

In Cluster 1, I know how I am related to all but Gary at 26 cM. However, I have been in touch and he knows my Frazer relatives in Ireland. In the above analysis, I went from further out clusters to more closely related clusters. It probably would have made more sense to start with the more recognizable clusters, but I had to start somewhere.

The last cluster is the false one I mentioned further up in the Blog.

Downloading Frazer Segment Data from Gedmatch

This could help in identifying other match groups or clusters. Roberta Estes has a helpful article on how to do this. Once I get the data from Gedmatch, I like to put it into my own format in an Excel spreadsheet. For example, I should be able to look at the region where I have Clarke matches to identify other potential Clarke Clusters. Unfortunately, when I got to the Download button at Gedmatch, it didn’t work.

I guess I’ll have to explore these matches later.

Grandfather Hartley DNA

This is all exciting and new territory for me. This time, I will start my grandfather Hartley Cluster with the range of 23-250 cM:

Even at this level, I get a lot of unrecognizable clusters. Cluster 11 has Lee in it. He has Hartley ancestry in the area of Colne where my Hartley ancestors were from, but I was unable to make a connection.

Hartley Clusters 25-1,000 cM

This brings in some of my 2nd cousins and 1st cousins once removed. Here Cluster 2 is Lee again with the Hartley ancestry. He has a shared match with my 1st cousin once removed. That shared match has Shackleton ancestry, so that is a possible connection.

For some reason, the AutoCluster at Gedmatch does not work as well on my Hartley side as it did on my Frazer side. This is possibly because I have worked to have a lot of my Frazer relatives out to the level of clustering upload their results to Gedmatch.

My Maternal Grandfather: Rathfelder

My maternal grandfather was German but grew up in Latvia. Here are some of the clusters for that side. The limits I set were between 20 and 1,000 cM:

The purple cluster has some relatives that are still in Latvia. The top match in red Cluster 3, Otis, has Schwechheimer ancestry in common with my mother’s side.

My Maternal Grandmother’s AutoClusters

This should finish off the Blog. My maternal grandmother was a Lentz. However, I get many matches to her Nicholson mother’s side. This run is between 19 and 500 cM:

Many of these names and clusters are identifiable. Cluster 3 is Nicholson. Cluster 4 has my mom’s relative who is both Nicholson and Lentz.

Sadie, Joan, Linda and Carolyn are in the red Cluster 3. The largest match in Cluster 4 is with Judy who is a first cousin once removed to my mother. As such, she is related on the Nicholson and Lentz side.

Father and Daughter in Different Clusters?

Yes it happens:

Robert is in brown Cluster 5 and his daughter Sadie is in red Cluster 3. As these both have the common ancestors with me of Nicholson and Ellis, my guess is that one cluster could favor Nicholson DNA and the other Ellis DNA.

Summary and Conclusions

  • I took one of my first looks at my four synthetic grandparent kits at Gedmatch
  • I had the best luck with my Frazer side. This is probably because I’ve spent a lot of time working on getting Frazer relatives to upload their results to Gedmatch.
  • I started a DNA Painter map for my paternal grandmother’s Frazer side.
  • I tried clustering my four grandparent results. I recognized the fewest clusters on my Hartley side.
  • I found an interesting match with Elaine on my Frazer side using clustering. However, I couldn’t find further information on her family tree and couldn’t find her Ancestry account.
  • I tried downloading segment information for my grandparents, but I couldn’t get that utility to work at Gedmatch
  • There is still a lot of work that could be done with these four Gedmatch kits that represent the DNA for my four grandparents.



Leave a Reply

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