Most of you are probably familiar with AncestryDNA’s relatively new tool, Custom Groups, which allows you to add a colored dot next to your DNA matches. These colored dots can be used for whatever you decide including indicating clusters, your eight great grandparent lines, or a specific group you are working on.

A new tool at Genetic Affairs allows us to use the custom groups to create clusters. We can also use the “extend” option to grow these clusters even larger. The extend option gathers the shared matches of the DNA matches in whatever group you’ve selected.

A few days ago, I used ThruLines to concentrate on descendants of a specific immigrant ancestor: Thaddeus Koechle. I ran an AutoCluster analysis using all of my uncle’s DNA matches predicted by ThruLines to be descendants of Thaddeus. I also included all of the shared matches of those DNA matches. But, an easier way to accomplish this task would be through a new option on Genetic Affairs which allows you to use Ancestry.com’s custom groups to create clusters!

ThruLines to Custom Groups

For this example, I once again used my uncle’s DNA and one of his great, great grandfathers who was also one of our immigrant ancestors: Anton Karbach. First, I opened the ThruLines for Anton and changed it to the “list” view to see a list of those DNA matches who are predicted to be descendants of Anton.

Not including our close family members, ThruLines showed 8 likely descendants of Anton through 3 of his children: Margaret, Christina, and Maria. I created a custom group on Ancestry and marked these 8 matches with a teal dot. The first 4 of those matches are shown below with their “teal dots” in the 4th column.

Running AutoCluster with Custom Groups

From my uncle’s DNA Matches’ page on Ancestry, I can click on “sort” and see a list of custom groups I’ve created. The group labeled “02 Karbach/Reuter group” is made up of those 8 descendants identified by ThruLines to be likely descendants of Anton Karbach.

After selecting this group, I need to find the Ancestry group “identifier” which can be found in the address. With the teal group selected, this is what my address looks like:

You can see the address ends with “groupby=1001” – so the group number is “1001.” [Note: The numbers appear to be 1000, 1001, 1002, etc. And you can combine OR exclude groups with AutoCluster!]

Now that I know the group number, I can use it to run an AutoCluster analysis on all of my teal group members.

I selected a cM range of 600 to 12 cM making sure I’d covered the range of my matches in this teal group. In the “Use Ancestry groups” box, I put in the group number I’d found in the address: 1001. I selected “run on more powerful server” since it’s required at this level of cM. And, I chose “extend clusters” which will also find the Shared Matches of those matches in the teal group thus likely enlarging the number of matches used to create the clusters.

The results of this analysis – with initially 8 matches from 3 predicted different siblings – is as follows:

I now have 5 clusters with a total of 8 DNA matches.

Analysis

My next step is to try to identify as many of the actual matches as possible AND to identify the MRCA (Most Recent Common Ancestor) of as many clusters as possible. I work within each group to try to build their trees until they “meet” at a specific person – hopefully at Anton Karbach and his wife, Mary Ann Reuter. But, I’ll leave the analysis for next time!

Your Turn

Have you tried using Ancestry.com’s custom groups with AutoCluster? If so, what did you find out? If not, I hope you’ll give it a try!