I looked at my sister Heidi’s AncestryDNA clusters here and my brother Jon’s here. I used the bottom level of 25 cM and top-level of 600 cM for their matches. This resulted in 20 clusters for Jon and 23 clusters for Heidi. This is what my clusters look like at those same levels:
I have 37 clusters. These clusters are proportional to the number of 4th cousins or closer that we have a AncestryDNA.
In my previous tries at looking at my clusters, I chose match levels that resulted in first 5 clusters and then 76 clusters. 37 clusters seems like a good number, plus it will give me a good comparison to my two siblings.
Starting to Identify My Clusters
The thought behind clusters is that a group of clusters probably indicates a group of ancestors that are all along a particular line. I have sorted my clusters by DNA match. I have my top match in the cluster followed by the match amount in cM. Then I have the cluster number, my grandparent line and then common ancestors or other notes:
I continued the cluster numbers down in the order of their match levels. Cluster 17 is out-of-order in a sense. There are only 4 people in the Cluster 17 and one is my maternal 1st cousin’s daughter. So those matches could be on either side of my maternal grandparents.
Cluster 10 and Gladys
Here is how I am related to Gladys:
That’s assuming that I have this tree right. Gladys is my third cousin, once removed. We have a double Frazer ancestry. That makes me wonder how I am related to the other people in Cluster 10:
Ancestry shows me at the same relationship to Gladys as with the rest of the group. However, I’m not so sure about that.
I tried building out Debra’s tree and found some Irish ancestors:
Little and Burns were both from Ireland if I drew Debra’s tree correctly.
Here is John’s tree:
John appears to have two Burns Lines in his tree. So that is something to keep in mind in case this is more than a coincidence.
It looks like my top 5-10 clusters are all on the Frazer side.
Some Relatives from Russia?
My Rathfelder ancestors lived in a German Colony in Latvia. In Cluster 11, I have some relatives that were from a German Colony in Russia about 1,000 miles away from Latvia. I have read of at least one connection between the two colonies. These are my top 20 clusters. I seem to favor the Frazer side as more than half of the clusters are Frazer clusters.
Diving Further Into the Unknown Clusters
The next Cluster is 30. One of the two people in that Cluster, Howard, has a tree with this person:
This looks like the same Hannah that I have in my tree:
Hannah’s grandparents would be Howard and my common ancestors: Samuel Snell and Mary Head.
Cluster 16 and a New Ancestor Discovery
I can see this from this table from AutoCluster:
A note that I had put under my AncestryDNA match Bobby turned out to be helpful.
I have no idea who Seymore is and have no known ancestors in this area of the country. I suspect that we may have common ancestors in England or Ireland. From what I can tell, this match is on my Frazer grandparent side.
My Last Seven Clusters
Here is a summary of all my clusters:
My Clusters Compared to My Brother’s and Sisters’ Clusters
I wold like to see how my clusters compare to my brother Jon’s and my sister Heidi’s. It looks like my matches tend to the Frazer side. The process was a bit annoying, so I took the data files into MS Access and compared them there. I came out with this comparison:
This shows where my clusters are equivalent to Jon’s or Heidi’s.
Here is the same chart by match size:
I match Cheryl at 69.1 cM but her Cluster has no match with Jon or Heidi.
This shows that I split Heidi and Jon’s Cluster 1 into two. They are now my Clusters 2 and 1. Likewise, I split Jon’s Cluster 20 and Heidi’s Cluster 15 in two. They are my Clusters 21 and 15. One theory is that I am related on both common ancestral lines and Heiid and Jon are related on only one. My assumption is that my Cluster 15 is related to my Cluster 21. I split Jon’s Cluster 6 into my Clusters 10 and 28. However, I don’t match Heidi on my Cluster 28. I had already determined that My Clusters 10 ane 28 involved the same Frazer couple.
My maternal Cluster 17 is described by Jon’s cluster 8 and Heidi’s Cluster 14. These are not helpful because they are on my mother’s maternal and paternal sides. I need to lower the thresholds so Taylor does not show in the matches. She is a 1st cousin once removed, so she is related on both of my maternal grandparent sides.
Summary and Conclusions
- I have 37 clusters. I matched Jon on 14 of my clusters and my sister Heidi on 19 of my clusters. This sounds about right as we should match half our our sibling’s DNA and I have more clusters than they do.
- I liked running AutoCluster with a top cuttoff of 600 cM to get an idea of how to sort the clusters. However, once those clusters are sorted, it is good to lower the top cutoff. I lowered my top cutoff for my next run with my sister Sharon to 300 cM and got good results.
- I like sorting the clusters by match size. This should put the more recent matches that are easier to identify at the top of the list.
- I like to compare my clusters to my siblings’ clusters to see where I match amd where I don’t. I was also able to see where my clusters split my siblings’ clusters in two in some cases.