Looking at Madeline’s AncestryDNA Frazer Shared Clusters

I hope the title makes sense. I just finished working on Joanna’s Shared Clusters here. It turns out I generated more questions than answers in that Blog. One question had to do with the lack of DNA matches that Joanna had with Madeline’s line.

Madeline’s Genealogy

Here are Madeline and Joanna. They show as third cousins. Third cousins should match each other about 90% of the time according to FTDNA:

However, in my previous Blog, I didn’t see consistent matching between Joanna and Madeline’s branch which should have showed up in Shared Clustering.

Madeline’s Shared Clustering

I’ll just jump into Madeline’s Shared Clustering. The good (and bad) thing about that is that I’ll need to know a little more about Madeline’s genealogy. It looks like I have access to Madeline’s sister Charlotte’s DNA also. Well I’ve already downloaded all of Madeline’s matches, so I’ll go with her.

Here are Madeline’s matches according to the Shared Clustering Program:

Here is Madeline’s sister Charlotte’s tree:

The basic things I need to know is that Madeline’s and Charlotte’s four grandparents are:

  • Crowley
  • Cronin
  • Emmet
  • Frazer

I’ll start with the basic matches for Madeline at 50 cM or greater. This was a bit basic as it gave me just one embryonic cluster.

I say embryonic, because I can see things happening within Cluster 1 and outside of it. Next I’ll ratchet Madeline’s matches down to 35 cM.

Madeline’s 7 Shared Clusters at 35 cM

This is everything you need to know about Madeline’s AncestryDNA matches at 35 cM and above:

Madeline has a large Cluster 4. The next step is to do some basic identification of the Clusters and narrow down to Madeline’s Frazer side. The easiest way to identify common ancestors is by the Common Ancestors column in the image above.

Cluster 4 appears to be on Madeline’s paternal side which is Crowley and Cronin. Kathy who is Charlotte’s daughter has done some work and she has the small Cluster 1 as Frazer:

Kathy’s code for pink in the second colored column is for Catherine Matilda Frazer and the reddish color in Cluster 1 is labeled Frazer.

The next Cluster that has mostly Frazer is Madeline’s Cluster 7:

However, Kathy has Lyall coded as yellow which is Crowley.

Madeline’s Clusters at 27 cM

Ancestry’s shared matches go down to 20 cM, so 27 cM is about halfway down from 35 cM. Kathy shows some possible Frazer connection in Clusters 5, 8 and 9.

 

Basically Clusters 5-9 may have a connection with each other. However, still no common ancestors have been identified by AncestryDNA.

Madeline’s 20 cM Clusters

This is the lowest level Ancestry has Shared Matches. The Shared Clustering Program calls this ‘all visible shared matches’. At this level, Madeline has 45 Shared Clusters. Now Madeline has some clusters with identified Frazer common ancestors:

One thing I notice is that the clusters with Frazer common ancestors don’t have affiliated clusters. That is the last column before the colored columns. The other observation is that the Cluster 8 Frazer is separated from the Frazers at the bottom of the list in Clusters 38 and 40.

Madeline’s Cluster 8 match is Clyde:

I see that I also have access to Toni’s AncestryDNA matches. In order to see this tree from the sides of the three brothers (William Fitzgerald, Edward Wynn and Thomas Henry Frazer, it would be a good idea to look at Toni’s shared clusters also. However, due to changes at AncestryDNA, it has been difficult to download to Shared Clustering lately, so this third analysis may have to wait.

More on Cluster 8 and David

David from Cluster 8 has a Fraser in his ancestry, so let’s take a look:

Hopefully this is more than a coincidence. I’ll see if I can build out David’s tree. According to the 1930 Census,

Keith’s father was from Canada and mother from Ohio.

Further, I see from the 1910 Census, that Augusta’s parents were from Ireland. That means I may have been on the wrong track with the Fraser lead:

It turns out that the Wilson side was from Ireland also:

The clusters are easy, but the genealogy is a pain. I did find this information at Ancestry, but I don’t know if it is accurate:

This family would have been in an area of County Roscommon that would not have been too far from where the Frazers lived.

Gerald From Cluster 8

Gerald is the only other match from Cluster 8. I notice that Gerald has Michigan ancestry which makes me curious.. That means that I get to build out Gerald’s tree. Fortunately, I don’t enjoy watching TV much. This is what Gerald has:

That means in Madeline’s Cluster 8, we see 100% of the matches have trees. Here is where I got with Gerald’s tree:

The paternal side seemed to be going to French names, so I looked at the maternal side and found Gertrude Dwignan. This sounds like the Duigan that I found above.

Here is Getrude “Dwignan” in the 1900 Census:

Things get a little complicated with the extended family. Gertrude is living in her husband’s mother’s house. There is a boarder named William Duignan also living in the house who is likely Gertrude’s brother. 

Here is Gertrude in 1880:

We see Gertrude’s brother William and her father John also. Unfortunately, her father was a widower at age 30. You may recognize Augusta from the previous genealogy I did above.

That leads me to this tree:

I have seen Henry Fry from a marriage certificate. I have seen Abigail Knott in many trees, but I have not verified that. However, my assumption is that Madeline’s Cluster 8 goes back to the Knott name. Here is one Archibald Frazer who married a Knott:

All this to say that it appears that Cluster 8 could be a Knott Cluster. I notice that Clyde who matches Madeline from Cluster 8 has a William Knott who married Elizabeth Knott in her tree also – though not as a direct ancestor.

Taking Shared Clustering Down to 6 cM

This doesn’t add new clusters, but adds matches around the clusters that are associated with the clusters. For Cluster 8, there was not a lot of change:

Madeline’s unaffiliated matches above Cluster 8 are associated with Cluster 7.  Madeline’s unaffiliated matches below her Cluster 8 are affiliated with Cluster 9 except for two matches. These matches were above 20 cM, so could have been in a cluster but they didn’t have enough matches. Again, Cluster 8 is not associated with any other cluster, so that makes me think that it could be an old cluster, perhaps going back to a Knott or other Irish ancestor.

Madeline’s Clusters 38 and 40

I had identified these two Clusters as having Frazer common ancestors:

However, as I look at Cluster 38, I see some connection with Cluster 37 and no connection with Cluster 36. That means that it would make sense to look at Cluster 37 also.

Cluster 38

Let’s see what we are dealing with in this Cluster first. Madeline has two matches: SC and LC. From checking AncestryDNA, I see that these two are siblings. Here are Madeline’s ThruLines for her common ancestor James Frazer born in the early 1700’s:

LC and SC are shown above. My claim to fame in the James Line is that I descend from Margaret Frazer. I am also a match to CK. I am pretty sure that this is the line that Margaret Frazer fits into.

Some SC Genealogy

This is what SC’s administrator has for a maternal side tree for SC:

This answers a question I had as Kathy had a category for Catherine Matilda Fraser and mentions her a lot. Here she is above in SC’s tree.

Here is Catherine in Leeds, Ontario in 1871:

This shows that George and Catherine were born in the United States, which may be wrong. Here is Catherine and family in 1851 in Leeds, Ontario:

Now she shows that she was born in Ireland. There appears to be some confusion as to the parentage of Catherine Matilda Frazer. Ancestry suggests that her mother could have been Mary Wooleghan and that Mary Wooleghan’s father was Michael Frazer. However, I am not seeing an obvious connection there. The tree that has Mary Wooleghan has this:

From this tree, it looks like Mary is the daughter of Michael Frazer and Margaret Stuart. I assume that the intention was to have William the son of Michael Frazer and Margaret Stewart.

Claire from Cluster 38

I’m hoping that Claire will shed some light on Cluster 38:

I took a guess on building out Claire’s maternal side. I got a little lazy and accepted some of Ancestry’s suggested hints. On that side I saw three of Claire’s ancestors from Ontario:

I have that Fanny Bellows was from Beachburg, Ontario. Beachburg is WNW of Ottawa. I have that Crysler Cook and Susan Wallace were from Stamford Township, Welland County, Ontario. Here is Susan Wallace in 1851:

Susan’s father appears to be Alex ‘Wallis’ a Presbyterian from Ireland. Crysler’s father was a Baptist named Abner who was born in Canada. I see that Madeline has Wallace in her tree but she was on the Emmet side.

Madeline’s Cluster 40

The next and last Cluster that Madeline has with Frazer common ancestors is Cluster 40.

This is a fairly small cluster with five people in it. However, there are many people outside the cluster. Those outside the Cluster have a low DNA match below 20 cM which could mean a more distant match that 4th cousin. Ancestry uses 20 cM as its cutoff for 4th cousin and for shared matches.

Gail

Here is Gail:

By the chart, Madeline and Gail are third cousins.

NB and BZ

NB has the largest match with Madeline in the cluster (33.9 cM). She has a large tree, but unfortunately, it is private. BZ is Betty on the chart above.

DO and PO

D.O.’s tree is confusing because the tree opens up to one person who I assume is DO’s spouse. Here is the tree I get for the spouse:

I assume that the O in the DO is for O’Hair, but I can’t be sure. Here is the Henry name mentioned as a possible common ancestor between Madeline and Gail:

Archibald Frazer, Mary Ann Henry (?)

Here is what DO’s tree has for Anne Henry:

If the connection is with Henry, then that could mean that Cluster 40 is a Henry Cluster and not a Frazer cluster. Also I see Palmer in the tree which is also in Gail in Betty’s tree.

PO is one generation younger than DO. Here is PO’s tree:

This means that Madeline matches PO on POs paternal side.

Matches Affiliated with Cluster 40

From the small sampling of matches in Cluster 40, the common denominator could be Henry. Let’s look at some of the other matches affiliated with Cluster 40 that are genetically more distant.

Toni

Toni does not have a tree, but I have her in my tree here:

Toni and Madeline have a small match of 7 cM.

SH

Madeline also has a match with SH of 16.2 cM. SH shows as the child of Walter Frazer along with Gail here:

Kay

Kay appears to be Betty’s niece and matches Madeline by 17 cM:

WG

Going down the list of those affiliated with Cluster 40, I see WG. WG only matches Madeline by 7 cM. As WG should be Madeline’s second cousin once removed, this seems like a very small match.

According to AncestryDNA, this is the chance of WG being Madeline’s 2nd cousin once removed:

The closest shared match between WG and Madeline is Sandy. Sandy and Madeline share 188 cM. Sandy is in Madeline’s Cluster 16 where Madeline has a White and Burrows common ancestor:

Here is Sandy’s tree:

I see that I already made my own tree for Sandy, so I must have gone down this road before:

These ancestors have New York State births, so fit in geographically with Madeline’s ancestors. However, I don’t see the specific matches as I probably didn’t go back far enough in time. The name of Susan Price is interesting as there was a Susannah Price who married a George Frazer who was born in Martinsburg, NY in 1858. However, he was from a different Frazer line.

Remember Sandy is in Cluster 16. She is not in Cluster 40 like WG, but is related in some secondary way.

Timothy

Timothy is next in the list of small matches above Madeline’s Cluster 40 that are affiliated with that Cluster. Here is his tree:

Here I notice a Mary Fraser married to Hugh McKay. I have come across Mary in the past and have built my own tree here:

Above is the paternal side for Timothy. There is even a photo for Mary M Fraser. I have that she was born in New York, but I think that Canada is a possibility also. The last time I looked at Mary Fraser, I was looking at the Michael Line:

However, I don’t think I made the genealogical connection.

Patricia

Patricia is the 6th person associated with Cluster 40, but outside the cluster due to her DNA match of less than 20 cM with Madeline. Here is her tree:

The maternal side of Patricia’s tree goes back to a Mary Armstrong who was born in Manorhamilton. Patricia has her mother as Margaret Palmer.

Joanna’s Brother and Sister

Madeline also matches Frazer researcher Joanna’s brother and sister in the group of DNA matches that is affiliated with Cluster 40. They should be third cousins, but by the DNA they would be less than that.

One of Joanna’s siblings matches Madeline at 9.4 cM and matches four out of five in Cluster 40. The other sibling matches Madeline by DNA at 10.0 cM and matches 3 out of 5 in Cluster 40. However, it is not totally clear to me that the match is on the Frazer side. Note above that Patricia also has a Palmer ancestor as does Joanna’s family. Kimberly also shows between Joanna’s two siblings but does not have a tree.

SW

SW shows this connection to Madeline:

SW is not on my chart.

I’m not sure if SW is the daughter of Sharon or of another daughter of Walter James Frazer. However, SW would be a second cousin once removed to Joanna. SW matches Joanna at 93 cM across 5 segments, which is about right for a 2nd cousin once removed. That tells me that something strange may be going on in the Edward Wynn Frazer line as Madeline has a low match with WG and a low match with Joanna’s family.

After SW there is a small match with WW, but I can’t figure out WW’s tree.

Below Cluster 40

Below Cluster 40, Madeline has four matches affiliated with Cluster 40:

Dean and Dot

Dean has a large tree, but it is private. Dot has a smaller tree, but I can see it:

Dot’s maternal side goes back to Henry. That was a possible spouse for Archibald Frazer born about 1795. I see there is some circumstantial evidence for the Palmer name also:

I’m too lazy too do my own tree for Dot, so I’ll skip that part.

Pamela and Christy

Pamela has no tree. Christy has this one:

A quick look at this tree shows Buchanan in common with the previous tree. Francis was from Belfast and his two parents are listed as being from Ireland.

Where Are Madeline’s Frazer Matches?

Madeline is one quarter Frazer, so theoretically one quarter of her matches should be on her Frazer side. As I saw above, it seems like some of the matches that are attributed to Frazer could be explained by matches to other ancestors. The problems seem to start here:

Madeline has normal matches to Mike, Charlotte, Kathy and Mary, but the match to WG is small.

At one step out, the matches on the to Thomas Henry Frazer Line are small also.

Let’s check WG, to see how this person matches others. Madeline has shared matches with WG:

David has no tree. NB has a large private tree. NB and DO are managed by the same person. BZ is the great grandaughter of Thomas Henry Frazer. It seems significant that none of Madeline’s very close relatives match WG – at least not at 20 cM or more.

An Analogy from My Own Genealogy

Actually the analogy is from my mother’s genealogy. My mother’s mother was a Lentz. My mother is one quarter Lentz and one quarter Nicholson. However, a lot of the closer matches are both Nicholson and Lentz, so it is difficult to separate the two. Here is my Lentz DNA/genealogy tree:

The Lentz/Nicholson relatives are on the left. The common ancestors go back to Jacob George Lentz and Annie Nicholson. The were born in the 1860’s. The DNA matches that are only Lentz have an common ancestor that goes way back to 1792 or three generations further back. It would be possible that Madeline has this same type of issue with her Frazer ancestors. However, Madeline’s situation seems a little different in that she has WG that she should match and doesn’t also other Frazer branches that should match more closely do not. But they do match faintly, so it gets confusing. Madeline could be matching on a more distant Frazer line or a more distant line that is collateral to the Frazers.

Summary and Conclusions

  • The small DNA matches suggest that something could be amiss with Madeline’s Frazer side genealogy.
  • However, she does have some small matches which suggest a match further back in time on the Frazer side or on lines that married the Frazers.
  • Madeline has a very small match with her second cousin WG. If WG matched other Frazer lines, that that would tell me that his genealogy is right and that Madeline’s is wrong. However, he doesn’t, so that means that there could be a double problem in the genealogy.
  •  I had looked at Joanna’s line [Thomas Henry Frazer] in a previous Blog. I had also wanted to look at the Willliam Fitzgeral Frazer Line (born 1821) to see how they fit in with the other two lines, but there was a trouble with Ancestry working with Shared Clustering, so that will have to wait.

 

Looking At Joanna’s Frazer Shared DNA Matches and Shared Clustering

Frazer genealogist Joanna recently expressed some interest in Shared Clustering, so I thought I’d take a look at her shared DNA matches and shared clusters. I won’t go into how shared clusters work, but the output is supposed to organize your DNA matches into ancestral lines.

Joanna’s Ancestry

I am mostly interested in Joanna’s Frazer ancestry, as I am distantly related to her on at least one of her Frazer lines. Here is what I see for Joanna’s tree:

Joanna is divided into four parts: Frazer, Seymour, Dickins and Williams.

Shared Clustering

Shared Clustering has a basic 50 cM setting. Here are Joanna’s matches sorted at 50 cM and above:

Cluster 1 looks like Joanna’s Williams side. Cluster 2 is Joanna’s close relatives and Dickins side. Cluster 3 is Joanna’s Frazer side. I didn’t see anyone from Joanna’s Seymour side. That means that she doesn’t have close matches on that line or not enough close matches to make a cluster. Because Joanna has many close relatives tested, the demarcations between the clusters don’t show up well. Close relatives will be in all clusters. I put a yellow box around the Frazer Cluster 3.

At 30 cM Joanna Has 7 Clusters

Here is a shrunken down image of all of Joanna’s matches and how they are clustered into 7 groups:

The places where the red markers go a long way in a line are Joanna’s closest relatives.

The Cluster that I recognize at this level is Cluster 2 – Gail and Sharon are in this Cluster. They are Joanna’s second cousins with common ancestors Thomas Henry Frazer and Eliza Jane Palmer. Gail and Sharon also show that they have a correlated cluster in Cluster 1. I take that to mean that Cluster 1 could be a Palmer Cluster.

More on Joanna’s Cluster 2 at 35 cM

Joanna’s Cluster Two is a bit of a compound Cluster:

The first part of Cluster Two is somewhat separated from the second part but not totally.  Let’s look at my Frazer DNA/Genealogy Chart:

Here I’m at a bit of a loss as I am not an expert on Joanna’s side of the Frazer tree. However, one interpretation is that the two sides of Cluster Two could be for William Fitzgerald Frazer and Thomas Henry Frazer. The question could then be: what happened to the descendants of Edward Wynn Frazer born 1830? There has been a problem with Joanna’s family matching this branch in the past, so let’s keep an eye on this branch. I see that I have permission to look at Madeline’s DNA, so it may make sense to look at her shared matches to see how they fit in. This will probably require a separate Blog.

Kelly in Cluster 2

Joanna has a match in Cluster 2a with Kelly who has Frazer genealogy. Let’s take a look. Cluster 2a is the William Fitzgerald Frazer Branch.

Here we see that Kelly shows her ancestry goes back to a William Frazer born 1824. Also notice that Kelly’s ancestor has a middle initial of W. Here is how Joanna shows Kelly’s ancestor:

So while I was thinking that Kelly would fit into the William Fitzgerald Line, the genealogy shows that she is in the William Wynn Line. Here is Kelly’s ancestor grave marker:

This seems further supported by the 1851 Census of Ontario:

That leaves me with this possible interpretation of Joanna’s Cluster:

 

 

This shows that Joanna matches Thomas Henry closely (through second cousins). She matches William Fitzgerald descendants who are third cousins once removed. Then she matches a group that appears to be descended from William (not Edward) Wynn Frazer. This is probably William Frazer:

That leads to this crazy situation:

This shows that Joanna is a 5th cousin twice removed to Kelly. I say it’s crazy because the relationship is so far out. Joanna matches Kelly on two segments which means that they could be related on more than one line. For example, Joanna doesn’t have a wife for her ancestor Archibald Frazer born 1792. Perhaps his wife was a Knott. Or Kelly and Joanna just have a randomly large DNA match considering the relational distance.

A Transitional Clustering at 30 cM

I already ran this Shared Clustering for Joanna, so I’ll show it.

Here, Joanna’s second cousins popped out into their own Cluster 3. Cluster 2 now appears to just be William Wynn Frazer and Kelly and William Fitzgerald Frazer in the lower right part of the Cluster.

Thomas From Cluster Two

Because I only have one known person (Kelly) in the first part of Cluster Two, I’ll look at Thomas. He has an Alexander Frazer in his tree from Pennsylvania. I’ll take what Thomas has and see if I can make a connection:

This appears to be Alexander and family in 1920:

Alexander’s brother William was said to be born in Connecticut. Here is Archey’s Declaration of Intention:

Archibald was transcribed as being born in “Roocommon”.  That narrows it down. It appears that Joanna was ahead of me and already has this line:

However, this is in a different Frazer Line than Joanna is in, so more mystery. I’ll be glad to add him to my Frazer/Stinson Tree:

So the mystery deepens. Joanna, what is going on? Again, I suspect that there could be a non-Frazer connection somewhere. The question now is how Kelly and Thomas are related?

Joanna also matches Emily on this Line. My guess is that Emily is Thomas’ daughter. In fact, that is how Joanna has Emily.

Taking Joanna’s Shared Clustering Down to 20 cM

At this level of DNA matches, Joanna has 31 Clusters. Let’s see what they show.

Emily, Thomas and Kelly are still in the same Cluster, but now it is Cluster 22. The fact that these three are still in the same cluster tells me that this could be a fairly old set of common ancestors. This is one possibility in addition to this being a non-Frazer cluster (or a cluster that has Frazer and another common ancestor).

Going Down to 6 cM for Joanna’s Clusters

Doing this doesn’t create any new clusters, but it shows people that probably would be in clusters if clusters did go down that low. I am setting my expectations low here. I am just hoping for not too many more questions to come out of this.

The extra matches that Joanna has are above and below Cluster 22. Here are the ones above:

What I find interesting is that the top part of Cluster 22 has a pink designation. Joanna has this as McMaster/Frazer/Haire/Bowles if I am reading it right. The green is Frazer.

Here are the extra matches at the bottom of Cluster 22:

They form at transition between Clusters 22 and 23. Cluster 23 has Toni from the William Fitzgerald Frazer Branch. Notice that Joanna’s pink designation has mostly gone away for Cluster 23.

Joanna’s Other Clusters

Joanna’s Cluster 26 is the McPartland Cluster:

There are only three in that Cluster. That McPartlands can trace back to an Ann Frazer. Here is where Joanna has Ann:

However, this is genealogy is just a guess at this point. I have written many Blogs about the McPartland including one on Joanna’s sister.  However, after looking at how the clusters are associated, I can see where Joanna would come to the conclusion that this is the place to put the McPartland family in her genealogy. The McPartland Cluster 26 has an affinity for Joanna’s Cluster 27:

Cluster 27 and Different Common Ancestors

Here is another thing to sort out. Cluster 27 and matches associated with that Cluster have differrent common ancestors:

 

As per above, the McPartland Cluster had common ancestor Archibald of Tullynure. That Cluster had an affinity for Cluster 27. It looks like all the Common Ancestors that are in Cluster 27 proper have Archibald of Tullynure (or Thomas Henry Frazer). It doesn’t seem normal that there should be a group that is associated with Cluster 27 includes Palmer in Notley. If these were really Palmer and Notley, I would assume that these matches should be associated with Palmer and Notley Clusters. That leads me to believe that could be some relationship between Archibald of Tullynure (or his wife) and Palmer or Notley (or their spouses).

Here I have filtered Joanna’s clusters and matches by Palmer and Notley common ancestors:

 

I also included the row number to the left. Cluster 1 has a Notley common ancestor in the cluster. Cluster 10 has a Thomas Palmer common ancestor. Then rob has a Notley common ancestor associated with Clusters 1 and 10 which makes sense. I’m saying that it doesn’t make as much sense that Claire, Alexander, etc have Notley and Palmer common ancestors that are associated with Cluster 27 which is a Thomas Henry or Archibald of Tullynure Cluster.

Summary and Conclusions

  • Many Blogs I set out trying to solve all the genealogical problems, but in reality I end up inching toward the truth
  • The Blog raised more questions than providing answers
  • One question: How does the Edward Wynn Frazer/Ismena Jane White branch fit in as the DNA connection is not clear? I hope to look at the DNA from this branch in my next Blog
  • Next questions: Joanna is in a cluster with Kelly who appears to be a 5th cousin twice removed.
  • This Kelly is also in a Cluster with Thomas who is in an even more distant Line of Archibald Frazer/Stinson. How are these two connected?
  • My last question had to do with Joanna’s Cluster 27 at the lowest resolution of clustering. She had many matches with Palmer or Notley ancestors in a Cluster which had her ancestor Archibald of Tullynure. This suggests that the two families could be related.
  • I looked at a small McPartland Cluster. This cluster seems to support the way Joanna has this family in her genealogy. However, I also match this family through X Chromosome matches. It would be interesting to try to integrate the connections between Joanna, McPartlands and my family.

Continue reading “Looking At Joanna’s Frazer Shared DNA Matches and Shared Clustering”

A New Nicholson DNA Match with Robert at MyHeritage

MyHeritage occasionally sends me updates on my new DNA matches. Robert was one that stood out recently. Here is how my DNA match looks like with Robert:

Robert and I match on the above 5 chromosomes.

Robert’s Genealogy

Because of past experience, I know that I will match Robert on his maternal side. I believe that Martha Pote is a daughter of a Nicholson. Here is the relative spot where Robert fits in on my DNA matching/Nicholson Genealogy Tree:

I say relative place,  because I don’t know for sure whether Robert is Sadie’s father or uncle. If they had both tested at MyHeritage, I would be able to tell. Actually, I may be able to tell, because I have Sadie’s DNA painted. Either way, Robert is my third cousin and we share Nicholson and Ellis DNA – specifically from William Nicholson and Martha Ellis.

DNA Painter

Here is Sadie at DNA Painter. This is how she matches me:

If Sadie only had DNA matches that contain Roberts, she is most likely his daughter. If she has DNA matches that are different than Robert, then Robert is likely not her father. This is based on the fact that Sadie got all the DNA that matches me from her father. When I check DNA Painter, Sadie’s DNA is contained in Robert’s DNA matches, so it is likely her father. Plus I see Sadie’s email listed for Robert at Gedmatch, where his results also show up.

I have quite a few Nicholson matches already, but Robert adds some new DNA (see circled below):

The new areas of Nicholson/Ellis matches are on Chromosomes 1 and 5.

Here are some other matches on my maternal side Chromosome 1:

I have left out the names for privacy, but Robert is the purple match. He is right over Judith who is a closer match under Lentz/Nicholson. This tells me that my match with Judith is on her Nicholson side and not on her Lentz side.

Here are all my maternal matches:

Note that Robert’s purple match on Chromosome 1 is obscured by the green matches. Expanding Chromosome 1 shows the details.

Summary and Conclusions

  • Robert descends from Sarah Nicholson who was the older sister of my great grandmother Annie Nicholson, making us third cousins
  • I was able to paint Robert’s DNA onto my Chromosome map. This DNA represents the DNA Robert and I share from either William Nicholson or Martha Ellis. This couple was born in Sheffield, England and moved to Pennsylvania.
  • Robert’s DNA added to my Nicholson/Ellis DNA as well as adding to my mother’s DNA profile and my siblings’ chromosome maps.
  • In one case, due to a matching overlap, I was able to tell that a closer Lentz/Nicholson match was actually on the Nicholson side.
  • When we match an ancestral couple, the DNA is actually from one or the other of the couple, but we usually don’t know which. For the example of the match with cousin Judy, Robert identified that the DNA was from Annie Nicholson and not her Lentz husband.

 

 

A New AncestryDNA Match with McMaster and Frazer Ancestry and My Sister Lori’s Shared Clusters

I was checking my new matches at AncestryDNA and came upon a match named Rebecca who has a Frazer ancestor from Sligo. This got me interested.

Rebecca’s Frazer Genealogy

Rebecca has a small tree. Here is Rebecca’s maternal side:

I sent a message off to Rebecca, but I’ll also try to build out her tree. It seems like this James may be the James Robert Frazer I have in my tree:

The death date is close.

My guess is that here is how I am related to Rebecca:

That makes us third cousins. She is also the first DNA match that I know of from the William Frazer Line. I did hear back from Rebecca who confirmed that James Robert Frazer was her grandfather.

My DNA Match with Rebecca

Rebecca and I match at 64 cM over 6 segments. Rebecca shows on my list as a 4th cousin, however, the AncestryDNA table for a match of 64 cM shows this:

We have a 37% chance of being third cousins and a 4% chance of being fourth cousins. Rebecca and I don’t share Whitney as a DNA match. Whitney also tested at AncestryDNA:

It gets complicated because Rebecca’s grandmother was also a McMaster. Here is Rebecca’s grandmother’s McMaster side:

I also descend from James and Fanny McMaster. Here is another way to show this:

 

This shows that Rebecca is Keith’s 1st cousin once removed. Also Rebecca is both my third cousin as well as my fourth cousin. This shows how Rebecca would be related to me on my McMaster side. Rebecca descends from Margaret and Hubert McMaster who were the children of Fanny and James McMaster. The people in green have uploaded their DNA results to Gedmatch.com. If Rebecca did this, I should be able to tease out the Frazer DNA from the McMaster DNA. A further complication is that Fanny McMaster who is at the top of the tree is the daughter of Margaret Frazer. However, this Margaret Frazer is from a different Frazer line than George Frazer who is in the tree.

Shared Matches With Rebecca

Here are my shared DNA matches with Rebecca in addition to the shared matches of my siblings and Rebecca at AncestryDNA:

I have information on how I match BV:

BV is a third cousin, twice removed to me, my siblings and Rebecca. This seems to indicate that the DNA connection is on the McMaster side.

Lori and Marshall

In doing the shared matches, with Rebecca, Lori and Rebecca have a shared match with Marshall:

Here is what Marshall has about George McMaster – his 2nd great-grandfather:

It would be nice if I could fit George into the William McMaster/Margaret Frazer Tree above. Margaret Frazer moved to Ontario with most of her children. The best way I know to review Marshall’s tree is to create my own Marshall Tree.

Here is Frederick McMaster with his family in 1911:

Father George was a farmer and Frederick Herbert appeared to be a young twin.

1871 Census

Here is Abraham McMaster and probably his father George in the 1871 Census:

1861 Census

I am finding the 1861 Census difficult to interpret:

There is a note by Abraham which appears to say married. These appear to be the same people as in the 1871 Census. However, the last name is spelled McMastrin. Also Jane shows an “m” for married but also appears to be only 10 years old! I assume that this is the same Jane who shows as 27 in the 1871 Census. A little squiggle in the 10 could be interpreted above as an 18. Here, at least George and Catherine are the same age.

1851 Census

The 1851 Census is stranger still:

Here are 5 McMasters and the oldest is 12. Did the family leave them temporarily? I assume that this is the same family, but I am not sure. If this is correct, and two year old James McMaster was born in Ireland, then that means that they must have moved to Canada around 1850.

The assumption is that Abraham was born about 1837. This seem born out by this grave marker from St. Mary’s Anglican Cemetery in Osprey, Ontario:

Getting from Abraham Back to George McMaster

If Abraham was born in Kilmactranny Parish, that would have been during the time when records were missing. Here is one George shown in the Kilmactranny records:

William son of George and Jane McMaster

Born                                      Bapt. Aug 18, 1843

However, the timeframe is off for the William above unless he was born a while before he was baptized.

Here is where Osprey or Ospry is:

Another George McMaster?

I mentioned another George McMaster in this Blog. This George was married to a Jane and had a son named George Arthur McMaster:

This George was in Vaughan as early as 1847 when his son George Arthur was born:

Vaughan is in York County a little above Toronto.

The question is, was this George, the same as the father of Abraham above? Or put another way, did this George and Jane McMaster have a son named Abraham? The George from my previous Blog married Jane McMaster. The Jane above is believed to be Jane McMaster who is a sister of my ancestor.

Just to confuse things further, Ancestry gives this hint for the George in my Tree:

That hint was based on this tree:

I put a box around my ancestors.

Any Conclusions?

From the above, I would say that the George McMaster who married Jane McMaster is most likely not the same as the George McMaster listed above in the 1861 Census along-side of Abraham McMaster. Two trees at Ancestry have the George who married Jane McMaster dying in 1847. This is likely right as Jane marries William Thompson in 1851.

The fact that Abraham was living in the same or next house as George McMaster in 1861, implies but does not prove that George was the father of Abraham:

My further guess based on the genealogy and DNA is that the shared match between Lori, Rebecca and Marshall goes back further into McMaster history. I have been unable to connect all the McMasters, but the DNA seems to be pointing in the direction of that connection.

More DNA

It seems I have made this simple match with Rebecca into a complicated Blog. That is because Rebecca’s DNA match with my family seems to point to the McMaster side of my genealogy which is a bit convoluted – not unlike the Frazer side.

Here is some more detail on Rebecca’s shared DNA matches with my family:

 

Here I have added my match numbers with Rebecca as well as my siblings’ match numbers. I have also add match numbers where they appear with shared matches. In scanning from left to right, it appears that Lori has the most shared matches. After that, I added whether the match had an Ancestry Tree. CA means that they have a tree and Ancestry indicates that we share a common ancestor. I share common ancestors with Rebecca, but either Ancestry hasn’t picked up on that yet or Rebecca’s tree is not detailed enough. There are few directions I could take at this point. I could look at more trees or at more DNA. I’ll try the DNA route.

My Sister Lori’s Shared Clusters

Jonathan Brecher has a utility called Shared Clustering. I’ll run Lori through that program as she has the most shared DNA matches with Rebecca. Here is how the Lori’s shared clusters look at a limit of 40 cM:

At that level, Lori has 4 clusters. Cluster 1 represents her paternal grandfather side. Cluster 2 represents Lori’s maternal side and Clusters 3 and 4 on Lori’s paternal grandmother’s side. Rebecca’s match is right in the middle of Cluster 3. Right above Cluster 3 is a match with Michael. He and Lori have a Frazer common ancestor, but because Cluster 3 appears to be a McMaster Cluster, Michael is not in Cluster 3.

I have circled the different Clusters in green.

Whitney also shows up in Lori’s Cluster 3:

Melanie, Emily and Paul don’t show up because they had their DNA tested at different companies. Melanie doesn’t show up on Lori’s shared match list with Rebecca. However, because both Rebecca and Melanie match other who are in the Cluster, they are included in Cluster 3.

Bringing Lori’s Shared Clusters Down to 30 cM

I’ll try Lori at 30 cM to see who we pick up along the way. At this point, I’m just interested in the Cluster Rebecca ends up in:

Now Rebecca is in Cluster 1. A new Cluster for Frazers has appeared in Cluster 2. A few matches who were previously in Rebecca’s cluster jumped ship to the new Cluster 2 at this level. The last two members of Cluster 1 appear to be fairly closely related to Cluster 2.

Here is a summary of Lori’s clusters so far:

Bringing Lori’s Shared Clusters Down to 25 cM

At this point, Lori has gone from 9 clusters to 19:

Rebecca is now in Cluster 9 with a Frazer Cluster 10 right below.

Bringing Lori’s Shared Clusters Down to 20 cM

At this level, I should see Marshall. I had looked at his Irish/Canadian McMaster ancestry earlier.

Believe it or not, I find this cluster detail very interesting.

Clusters 27 – 33

Here is my markup:

There is a connection between the two pink squares. These represent Lori’s Frazer ancestry.

It appears that Violet Frazer and James Frazer were first cousins. The second pink chart above Represents Richard Frazer and his daughter Violet. That leaves the first pink box in the Shared Cluster Chart which should represent the James Frazer side.

However, notice that the second pink square is split into two. A possible explanation is that one of the two clusters could represent the unknown spouse of Richard Frazer born about 1777. I would assume that Cluster 30 could represent Richard’s unknown spouse. Whitney is in this cluster, so I am proposing that Whitney may have gotten more of the DNA from Richard Frazer’s wife:

It’s a theory.

A Clue to Marshall’s Ancestry?

Marshall is in Cluster 29 which subtly separated from Cluster 28. However, notice that Cluster 28 has more matches with other clusters. Cluster 29 has fewer matches. That leads me to believe that Marshall’s genealogy goes back on the James McMaster Line.

This leads me to another observation. That is that Cluster 29 appears to match up better with Cluster 33:

I had that Clusters 32 and 33 probably represent William McMaster and Margaret Frazer. So that tells me that William McMaster could be represented by Cluster 33.

Lori’s Final Shared Clustering

Lori’s final shared clustering should look the same as the 20 cM clusters. The difference is that small matches are added in down to 6 cM. My understanding is that these matches will generally be outside the clusters but associated with the clusters. Let’s take a look:

Here is the detail for Rebecca and Marshall. They are now in Cluster 40 and 41. In the past two runs Lori had 46 clusters. However, in this final run, the clusters got shuffled around – I suppose due to the smaller matches being added in.

Well, I could keep on going, but I’ve gone on too long already.

Summary and Conclusions

  • I was happy to find a DNA match with Rebecca. We are third cousins through our common ancestors of George William Frazer and Margaret McMaster.
  • Based on the DNA, Rebecca and I are more related on the McMaster side. That is not a surprise as our common ancestor Margaret McMaster had two McMaster parents and Rebecca’s grandmother is also a McMaster.
  • I also looked at some more distant DNA matches going back William McMaster born about 1790 and his wife Margaret Frazer. They left for Ontario in the 1800’s with their children but left one daughter Fanny stayed behind in Kilmactranny Parish, County Sligo, Ireland.
  • I then did a somewhat complicated cluster analysis focusing in on my sister Lori, her match with Rebecca and their common matches.

 

 

 

Shared Clustering for My Father’s First Cousin Joyce

Shared Clustering is a program that was brought to my attention by Jim Bartlett. As Joyce is my father’s first cousin, I share common ancestors with two out of four of her grandparents:

However, once I go down this route, I need to know something about Joyce’s Gurney and Rounseville sides to make sure I’m not going down the wrong path.

Jumping Into Joyce’s Shared Clustering

I had thought that I had downloaded Joyce’s matches from Ancestry already, but I guess it didn’t save the results. I’ll need to try again:

Joyce has quite a few matches – over 84,000. This could take a while to download. The advantage of using Joyce’s matches is that she is one generation closer than me to get matches from my great-grandparents. Stated another way, her Hartley and Snell matches should be about twice as strong as mine.

Now that Joyce’s matches are downloaded, I can begin to cluster. I’ll choose a lower limit of 50 cM:

 

I’ll put the results in a chart:

The good news is that all four of Joyce’s grandparents are represented in the chart above. There were six clusters. There were two more on the Gurney side that I didn’t enter.

Joyce’s 40 cM Clusters

Going from 50 to 40 cM adds two clusters:

The split is pretty even between Gurney in the upper left and Hartley in the lower right. Joyce’s Hartley mother was one of 13 children, so she has a lot of close relatives.

When I compare the clusters at two levels in MS Access, I get this:

Hartley was Clusters 5 and 6 previously. This shows that old Cluster 5 mapped to Cluster 6 and Cluster 6 mapped to new Clusters 6 and 7. That shows that the clusters are reorganizing.

Victoria in Cluster 6 with Howorth Ancestry

One person in Cluster 6 has Howorth ancestry. It would be worth looking into Victoria’s tree:

This is Victoria’s paternal side. I’ll try to recreate Victoria’s tree.  I want to focus on Margaret Howorth. According to the 1910 Census, James’ mother was Margaret and Margaret’s father was from England:

The 1910 Census has her born in Iowa in about 1878. Here is some information about Margaret from her marriage record:

Edmund Howorth from England

In 1880, Edmund was living on a Farm in Iowa with his family:

This appears to be Edmund living with his parents in Ohio in 1850:

The next record which is a hint at Ancestry seems to be a leap of faith except that one of my ancestors attended this same Baptist Church:

The record makes it clear that Edmund’s mother’s maiden name was also Howorth:

That means that Joyce could be related on either side of the family or both.

Edmund the father was a weaver in 1823. I’m not sure where Long Bridge is or was. I assume in Bacup. This also matches Edmund’s birth on the 1900 Census:

By 1856 the elder Edmund had made his way to Crawford, Iowa:

Howorths in England

It appears that the Howorth’s should have been in England in 1841 for the Census. This could be the younger Edmund:

He was living at Waterside which I assume is in Bacup with perhaps his grandmother and Uncles?

Back to Iowa for Edmunds 1861 Will

Edmund mentions first his daughter Sarah, then his wife Mary:

Next. he mentions his four children and finally his brother Daniel and his son Daniel executor:

The son Edmund is not mentioned. However Edmund is referred to as Senr. which I take to be Senior and thus the son Edmund is implied in the will.

Liverpool to New York in 1842

This appears to be the Howorth family traveling to New York in 1842:

However, some of the family appears to be missing. I assume that the elder James was a brother of Edmund.

Connecting the Howorths

I would like to connect this family to my line:

Perhaps more DNA review will help clear this up. These two lines are from Australia on the left and the US on the right and are connected by genealogy and DNA.

On To Joyce’s 30 cM Clusters

At 30 cM, Joyce has 21 clusters. Using Access, I get the new cross-cluster comparison:

I’ll try this for now:

However, there were two Cluster 6’s last time, so I’ll have to check. Also, I’ll want to see if there are any new Hartley side clusters.

Gurney Clusters 7 and 11

I have a note that one of the people has many of my ancestor names, but the DNA match is in a Gurney Cluster. That tells me that I should not pursue the ancestry of this one match. That is, unless this match has other DNA matches on my Hartley side.

There is a similar situation at Gurney Cluster 11. However, this Cluster has common ancestors from Joyce’s maternal and paternal sides.

First, if there is more than one common ancestor, I tend to go with the closest common ancestors.

Hartley Cluster 13

This is the big Cluster this time:

It also has a funny bump at the bottom. The top left of the Cluster is a part I’m interested in.

I already looked at Victoria’s tree above. That is the tree with two Howorth Lines. Above are my Greenwood Hartley matches. Here is Greenwood and his wife Ann Emmet:

It was Ann’s mother who was a Howorth.

Of the 7 I show in the first part of Cluster 13, I have already been tracking the last 5. I have Kristen and Emily in a tree already:

I have been unable to figure out where saudet fits in. Jennifer doesn’t have a tree and I haven’t been able to get in touch with her. Perhaps the first two matches, Janet and Stephen, will help place saudet.

From comparing trees, it is clear that saudet is related to Stephen. So I will look at Janet’s tree:

It appears that both sides of Janet’s tree are from Lancashire. I see that James Stott was born in Cloughfold, to the West of Bacup:

Janet had this marriage for James:

I’ll assume that Janet knows her genealogy.

Here Janet has James mother as Malley Stott. The 1881 Census mentions a Sarah Stott. So there is some confusion. Was Malley a nickname? So I didn’t make the genealogical connection but now Janet is on my radar. I did make a geographical connection.

The Other Side of Cluster 13

In the middle of Cluster 13 are a lot of close relatives. It comes out of that on someone named Bessey. I’ll take a look at her tree:

It’s a bit skimpy, but I get some hints:

Althea was married to a Fearing. I thought that I was related tot he Fearings through the Snells and this shows how. We both descend from Bradford and Snell.

It looks like I was already tracking a Katherine. This person must be the same person or a sister:

The part of the Cluster in green is Snell:

Moving on to Joyce’s 25 cM Clusters

This gets Joyce up to 43 clusters.

Cluster 2

Cluster 2 is broken into two major parts and a small part at the end. The first part is probably Snell/Bradford. The second part is probably Hathaway/Clifton. The third small part seems to merge back into Gurney (not related to me).

Running the 25 cM Clusters Through Access

Access makes reviewing the clusters easier:

This is only part of the answer. Looking at the last two columns, this tells us that the previous Cluster 1 mapped to Cluster 2 at the 25 cM limit. However, that is not the whole story:

When I sort by new Cluster 2, I see that it comes from other places. Usually the clustering is going from few to many. In this case, it went from many to one. From the chart above, at 50 cM, Clusters 3 and 4 are Gurney Clusters. This may be a result of the way I did my Access query or have to do with relatives of relatives.

Here is my Access query:

It starts with the 25 cM Clusters. There is a right handed arrow which means to include all the 25 cM Clusters plus the clusters where the 25 cM clusters are the same as the 30 cM clusters. I do that the same way down the line. I tried to query a slightly different way and got the same results.

Here are Clusters 3, 4 and 5 from 50 cM:

Clusters 3 and 4 are Gurney matches and Cluster 5 shows Hartley matches.

I can simplify my Access Query by only including those Hartley Clusters I considered at 30 cM:

Interestingly when I restrict the 30 cM Cluster, I no longer show Gurney Clusters 3 and 4 at the 50 cM level. This shows that previous Cluster 13 may now be 2, 4, 5, 6, 7, or 9. Cluster 2 is interesting in that it also came from Cluster 1.

Here is another way to slice and dice it:

Cluster 5 at 50 cM was my Snell side. Cluster 6 was my Hartley side. That means that Clusters 6, 7, 9, and 25 should be on the Hartley side. This is generally true except for Cluster 25:

There is someone there who is related to Joyce on her Gurney side. However, he also has a Howorth ancestor which would be on Joyce’s and my Hartley side. I had discussed this some above.

Restricting Members

In my previous Blog I restricted my wife’s large French Canadian clusters to 40. I wonder what would happen if I restricted Joyce’s clusters? One of the great things about the Shared Clustering Program is that it only takes a minute to see what happens. I tried it and it didn’t help. I only lost one cluster and I lost the English Hartley Clusters that I am interested in. So where it helped in one case, it didn’t help here.

Looking At Clusters 2, 6, 7, 9, 25

Actually, I already mentioned that Cluster 25 is a Gurney Cluster, so that is confusing. I already looked at Cluster 2.

Plus, Victoria’s tree with the Howorth families is now in Cluster 6. After shuffling things around a bit, I get this:

This also has Cluster 5 which I missed above. It takes a little bit to sort these out.

Cluster 4

I haven’t mentioned Cluster 4. This also came from busy previous Cluster 13. Cluster 4 is hanging off of the top left of Cluster 5:

There are only 6 matches in Cluster 6, but two are associated with Cluster 5 and one with Cluster 9. As Cluster 5 and 9 both have Snell’s in them, that tells me something about Cluster 4. Here is one tree in Cluster 4:

I’ll fill out this tree to see if I find any common ancestors:

Now if Ancestry could have figured this out, it would have saved me some time. Here is my tree:

Our common ancestors are Snell and Head. However, my guess is that there are other connections. I see also Palmer in LPs tree. I had to extend my existing chart from Isaiah Snell:

This is how LP matches Joyce:

More Snells in Cluster 5

Joyce’s first match in Cluster 5 has a very large tree. Joyce and John’s ancestors go two steps beyond the tree above to Thomas Snell. I might as well add in John. However, it will make the names tiny. That makes John and Joyce 8th cousins:

The other Snell match who I know of is OT:

Here is OT added in:

Cluster 7 From Lancashire

Cluster 7 has an interesting shape:

New in this Cluster is Ruth. Joyce and Ruth’s common ancestry go back to Pilling.

Ruth is in yellow as are the other 2 in this Cluster. One theory is that the overlapping clusters in Cluster 7 are for Hartley and Emmet. As Ruth has no known Emmet DNA, that could mean that Kristen is related to Joyce more on the Hartley side and Emily is related to Joyce more on the Emmet side.

Cluster 8

This Cluster is situated between my Hartley and Snell Clusters. There are only three in this Cluster. One match, LD, has a tree:

Here is my buildout on LD’s maternal side:

I’m curious about this part of the tree:

I did a genealogy on the Snell’s years ago and didn’t have that Samuel had a son named Peter. This Peter was from Freetown and Samuel was from Tiverton. So there must be a Snell connection, but perhaps not through Samuel. Also looking at Cluster 8, there are some matches to Cluster 9 which is a Snell Cluster.

Taking Joyce’s Clusters Down to the Minimum at 20 cM

This involves correlated matches that are down to 6 cM. This gets me 104 clusters. At the 6th cousin level, Joyce has 128 5th great-grandparents. However, I have already mapped out a match for Joyce at the 8th cousin level. At this level, Joyce has 512 7th great-grandparents. That is far back and old.

Comparison in Access

I brought part of the cluster spreadsheet into Access and compared it to the two previous cluster runs. I set the previous run to only include non-null values. Then I sorted by new cluster number:

So that tells me that the new Cluster 1 and 4 came from the old Cluster 6. Cluster 6 is the one I had as Howorth previously.

Here the new Cluster 1 has many notes on Howorth. However, I have no notes on Clusters 2, 3, and 4. Cluster 4 extends down quite far with a lot of small matches.

Starting at the Top – Joan’s Tree

There are some names above Cluster 1:

  • Sammy doesn’t have a usable tree
  • Ann has a common ancestor with Joyce of James and Mary Howorth
  • Joan has a tree with three people

Here is Joan’s tree:

Here is Mary Jane in 1940:

I think I already tracked this family. Here is Victoria’s tree:

Joan’s tree matches up with Edmund and Sabrina:

Actually, I have a guess for Edmund at the top:

This is from the Ebebezer Baptist Church in Bacup where my Howorth were baptized. Edmund mentions his brother Daniel in his will. That must be his brother Daniel above.

Here are the dates of the births:

The Tong Connection

Edmund and Ann’s children were born in Tong. I mentioned Tong in a previous Blog. Here are the children of my ancestor James Howorth:

This timeline is confusing becuase there are two Betty’s. Also Abram is another name for Abraham, so there appear to be two of those. Esther is my ancestor. I have that her youngest two siblings were born in Tong. Tong is to the East or SouthEast of Bacup:

Cluster 1

Cluster 1 adds Sara:

Sara’s tree already made the connection with Edmund Howorth of 1823.

Gretchen has an unlinked tree:

Now that I’ve gone down this route, I feel obliged to check for a Howorth connection:

I have a feeling the connection is through Hetty Nuttall, but I had trouble figuring out where in England she was born.

From Bacup to Trawden: Cluster 1 to Cluster 54

Cluster 53 has a Snell/Bradford common ancestor. Then there are some closer cousins. Then there are two with Hartley/Emmet ancestors. So Cluster 53 is a compound Cluster. Cluster 54 has some Hartley/Emmet common ancestors. Below that, there is a Sarah Pilling ancestor that I would like to look at. The first person in Cluster 55 has a Hartley ancestor from the area where my Hartley’s lived. The first person in Cluster 56 has Hartley also but no tree.

Sarah Pilling Common Ancestor

A match associated with Cluster 54 shows Charles and Joyce as having a common ancestor with Sarah Pilling:

I would like to take a look at this connection. In my tree, I have that Sarah Pilling was a single mother:

Charles’ tree stops short of the older Sarah Pilling:

Apparently, Ancestry made the connection based on my tree:

This is the record I had for John:

I don’t know if Charles got his/her information from my tree or if it was gotten independently. apparently, there was a record that John was the illegitimate son of Sarah. I had assumed that Greenwood Pilling was also an illegitimate son of Sarah who was not baptized (or at least I found no record).

Also, I note that if I have my tree right, there would be Stansfield in this line:

I have come up with Stansfield connections in other DNA matches, so that is something to consider.

Cluster 55

Here is how these Clusters compare in Access with Clusters from earlier runs:

That check that by walking it forward. By forward, I mean from the 20 cM run to the 25 cM run to the 30 cM run. 55 goes to Cluster 1 then Cluster 14. The problem is, I haven’t been tracking these Clusters.

The previous Cluster 1 was under the radar with only three matches:

This further mapped back to Cluster 14 at the 30 cM cutoff. That was also under the radar:

However, since Cluster 55 below has a Shackleton, perhaps it would be be worthwhile to look at Claire’s tree. Claire is also in Clusters 55 and 1.

Claire’s Cluster 55 Tree

This tree has dates but no places. I’ll be making my own tree. My tree shows that this family was from Wharfesdale which is between Ilkley and Otley:

I brought Claire’s tree back to 1823 int Otley. There I found an Elizabeth Hartley:

The clues for Elizabeth’s parents lead me to John Hartley of Colne and Sarah West of Otley. Here is Elizabeth’s birth and baptismal record:

The 1851 Census says that John Hartley was from Colne:

Here is Elizabeth in 1841:

Elizabether’s father, John, is listed on the previous page as an agricultural laborer. He apparently became a paper maker because of his father in laws. Also there is a George Hartley living next door – perhaps a brother of John who was born in Yorkshire.

Here is George’s birth in 1815 to Samuel Hartley:

It looks like I am stuck on John Hartley.

Lee’s Cluster 55 Tree

Lee also has a Hartley in his tree:

Confusingly, Margaret’s father is given as Lawrence Halstead:

 

Rebecca’s Tree

Rebecca in Cluster 55 has a tree with Shackleton:

Ancestry has shared surnames in green. Here some more of Rebecca’s tree:

The tree I built out for Rebecca also has a John Shackleton:

I’ll look into Rebecca’s tree. In 1871, Thomas Shackleton with his father Henry and family were living in Harden, Yorkshire, but were born in Heptonstall:

Here is Henry’s marriage record:

Beyond that, I cannot be sure.

Summary and Conclusions

  • In this survey on Joyce’s clustered, I focused mostly on the Hartley side. However, I looked at a few Snell matches
  • I used the clusters to identify new people in them with new trees that I could build out. It helps to know that you are looking for a certain branch of the family that lived in or probably near a certain location.
  • I didn’t have any major breakthroughs but attempted more contacts and got closer to finding out more about my ancestry.

Shared Clustering for My Wife’s Aunt Suzy

My wife’s brother recently took a YDNA test. I can’t do an AncestryDNA Shared Clustering for my wife’s father because he didn’t test at AncestryDNA, but I can look at Aunt Suzy’s Shared Clustering

What Is Shared Clustering?

Shared Clustering is looking at shared DNA matches. The theory if you have a bunch of shared matches that also match each other they are likely from a common ancestor. Jim Bartlett got the idea of walking back your shared clusters. That means starting with your more recent clusters and going back through the years to check on and verify your older clusters which represent more distant ancestors. Jim used Jonathan Brecher’s Shared Clustering Program.

Let’s Download Aunt Suzy

I used Jonathan Brecher’s program to download Aunt Suzy’s AncestryDNA results:

Aunt Suzy also goes by Virginia which is probably a more official name. Suzy has over 68,000 total DNA matches. That should take a couple of hours to download. Once the download is down I’ll be able to look at Suzy’s clusters.

What I Expect Based on Suzy’s Ancestry

I already know what to expect in general from this excercise. Half of Suzy’s ancesty is French Canadian. They will result in many matches. The other half is Irish and they will result in fewer matches and shared ancestors.

Suzy’s dad was from an Irish background. Suzy’s mom was more recently from French Canada.

A Test Run of Shared Clustering

Suzy has 82 third cousins. Where do I set my first cluster level? Ancestry says that Suzy should start getting third cousins at 199 cM. However, I clicked through a lot of Suzy’s actual cousins and the third cousins didn’t show up until somewhere bettween 128 and 118 cM. So I’ll set the first cutoff at 123 cM. Odd number:

This was good enough to give me a paternal and maternal view of Suzy’s ancestry:

Cluster 1 is paternal and Cluster 2 is maternal. The thing that looks like a Red Cross plus sign in the middle would be Suzy’s closer relates. They match on her father’s and mother’s sides.

I’m going to put Suzy’s cluster results in a chart sorting her clusters by grandparent:

Spoiler alert is that Suzy should have a lot of LeFevre matches. I could put in Suzy’s paternal and maternal single clusters in the chart, but that would be a waste of space. Instead, I’ll bring Suzy’s cluster level down to a maximum of 50 cM from 123 cM:

Still there are only three numbered clusters, though, as above, I can see clusters within the clusters. I’ll go lower still.

At a 40 cM Limit, Suzy Has 8 Clusters

I’ll be happy to start here:

Things get a bit blurry at this level. There are 216 matches on this chart. They are not all in clusters, but most of them are.

Cluster 1

This should be easy:

Here as a Common Ancestor, Joseph Pouliot is listed as well as Pepin. However, this relationship goes back several generations:

Maybe not so easy. The Pouliot that I was thinking of is in Cluster 5:

Three of the matches have the common ancestor of Joseph Pouliot and Josephine Fortin:

I’ve been in touch with Fred. Because there are two common ancestors, I don’t know which one the DNA goes with, so for now I choose the daughter in Virginia’s line who is Emma Pouliot:

On to Cluster 2

I put Cluster 1 on hold. Cluster 2 is quite large:

In order to tame it, I retrieved all the common ancestors from Cluster 2 and put them in alphabetical order:

This name comes up quite often:

However, this goes quite a ways back. I think I need a new set of clusters. I’ll go back up to 45 cM to see if this makes things simpler:

5 Clusters for Suzy at 45 cM

The thought is, that if these clusters are obvious, they should map to the 40 cM clusters.

Cluster 1 Is Large

By the common ancestor names, this must be a LeFevre cluster. Here is a popular common ancestor:

One step backward, two steps forward?

Cluster 2

This should be the same Cluster that Fred was in above:

Cluster 3 Goes To Old Common Ancestors

This was the problem I was having with Cluster 1 at 40 cM. In fact, I’ll go to Access to compare the 40 cM clusters with the 45 cM Clusters:

I start with the 40 cM clusters as there are more with them. I join them to the 45 cM clusters by the Test ID. Then for the join I choose option 2. That says include all the records from Virginia’s 40 cM cluster and only those records from Virginia’s 45 cM Clusters where the joined fields are equal. Then on the part below in the query screen, I reverse the order with the 45 cM results first, so it will look more like my summary table.

It is possible to do this without Access but it would take quite a bit of time. I also group my query results. This takes out the duplicates:

This shows that the Cluster 3 I was looking at with the 45 cM cutoff, maps to Cluster 2 at 40 cM. Cluster 2 at 40 was the same one I looked at above and was having trouble with because the common ancestors were so ancient. I then took the clusters up to 47 cM:

This shows Clusters 4 and 5 at 47 cM. Cluster 4 is the more recent Pouliot Cluster. That tells me that there are a lot of connections between the two Clusters and that the bigger cluster must be an ancestor of Pouliot (or Fortin).

Here are the same two corresponding clusters at th 45 cM cutoff:

Now they are Clusters 2 and 3. As the larger cluster is an ancestor of Pouliot, I’ll extend Emma Pouliot’s tree:

This brings Suzy out to her 5th cousin level. However, some of the common ancestors were out at the 6th cousin level. The confusion could be that there are two Trembely or Tremblay lines. However, both Tremblay lines go through Josephine Fortin. Here is her line with the Cluster 3 common ancestors highlighted. Actually, these are the Cluster 2 Common ancestors from the 40 cM run:

All these common ancestors funnel down through Emma Fortin. Imagine that you are John Brecher’s computer program and you had all of these ancestors. What would you do? You would probably come up with something like Cluster 3.

Bottom line:

Here I mapped Cluster3 at 45 cM to Cluster 2 at 40 cM. It’s a lot of work to map one Cluster, but I feel as though I got it right.

Cluster 4 at the 45 cM Cutoff

Cluster 4 is off to the bottom left of Cluster 3. Cluster 4 has some affinity with Cluster 3 as can be seen by some of the matches to the left of Cluster 4 and underneath Cluster 3. Cluster 4 at 45 cM maps to Cluster 1 at 40 cM. This is one that I got stuck on above.

I’ll take what I learned with Cluster 3 and apply it to Cluster 4. I’ll just say that this Cluster goes to Joseph Pouliot born 1848:

Above I highlighted the AncestryDNA suggested Common Ancestors for Cluster 4.

Cluster 5 – The First Irish Cluster

So far, I have been dealing with French Canadian clusters. This is the first Irish one.

This Cluster has both of Suzy’s Irish grandparents, so I’d rather not put it in my spreadsheet just now. In general Butler is at the top left and Kerivan is the rest of the Cluster. I note that, based on my previous Access query, this will be mapping to two Clusters at 40 cM.

Back to the 40 cM Clusters

Irish Cluster 5 at 45 cM mapped to Clusters 6 and 7 at 40 cM. I’ll start with those.

In addition, there are now Clusters 8 and 9. Due to close relatives, Cluster 6 has Butler and Kerivan still. Here is the area with just Butler:

I’ll call that 6a.

Clusters 6b and 7

Here are Clusters 6b and 7:

However, both Clusters have the same two common ancestors. It could be that one Cluster favors Kerivan and the other Rooney. I would rather assign them both to the daughter at this point.

Clusters 8 and 9

Clusters 8 and 9 are small. They seem to favor the Kerivan side. One person in Cluster 8 has a tree with 247 people:

I notice he has a Bridget Rooney in his tree. At this point, I could try to build out his tree or wait to see what happens with further clustering. I’ll create a tree as I’m curious. I’m having trouble getting back to Bridget from Alice McCusker. Here is a transcription of her baptismal record:

One of the witnesses was a Patrick Rooney.

It looks like the family was enumerated as McCarty in 1870:

In 1880, Bridget was a widow:

This record pulls things together:

From that record, I get this:

Lawrence and Suzy are third cousins.

I don’t like the look of Cluster 9, so I’ll ignore it for now:

When I say I don’t like the look of it that is because it appears that this could be two small clusters.

Clusters 3 and 4

In my mapping, the previous Cluster 1 went to Cluster 4, so that is LeFevre. Here are the listed Common Ancestors for Cluster 4:

The Butler entries are too close for clustering. Most of the Common ancestors seem to go back to Lazare Lefevre and Adelaide Boure:

However, one went back to Methot and Anger:

Ancestry would like me to evaluate this tree. I’ll bite.

I am questioning why Armand Petrelli is the son of Alphonse Moreau. The 1930 Southbridge Census has Armand as the son of Joseph Petrelli:

Actually, I see what happened.

Robert got his tree backwards:

His mother is Moreau. His maternal grandfather should be Alphonse Moreau – ot Joseph Petrelli.

Here is Jeanette in 1920:

The other problem is the Athanase I have in my tree is a man and Robert’s connection to me is shown through a female Athanase.

I think I’ll just assume that this is an earlier LeFevre Cluster rather than Methot:

That leaves the new Cluster 3:

Cluster 3 has four connections to Cluster 2 (Fortin) and two to 4 (LeFevre). I don’t have a lot of information, so I’ll just add Cluster 3 near Fortin:

This could be a long Blog, but I like that the clusters seem to be falling into place – even with the French Canadian families.

Suzy’s 18 Clusters at 35 cM

This looks like the 1,000 foot view:

In Access here is how the previous clusters map to this set of clusters:

This is interesting because the Cluster 3 I was looking at above, maps to 7 and 8 but Cluster 2 also maps to Cluster 7.

Cluster6 mapped to 9, but I had a 6a and 6b. It turns out that the Cluster 9 has not separated out yet due to Suzy’s close relatives being in that cluster. From above, it looks like a lot of the action is in the French Canadian quarters.

Looking At the Irish Side – Down to 30 cM

I’ve decided to take a new easier approach. I am more interested in the Irish side at this point. The French Canadian side is interesting, but much of the genealogy is seemingly well-defined. At least the answers seem to be there if you dig deep enough. Describing the French Canadian clusters may be interesting, but it is a lot of work. I had some luck above with Rooney, so I’ll stick with the Irish and go down another level to 30 cM.

I ran the Shared Clustering program and came up with some more large French Canadian Clusters. I notice there is a button where you can filter out the larger clusters.

I thought I wanted to use the first box, but that is for the entire file. I want the second box to get rid of the large French Canadian clusters. Without this filter I get:

  • Cluster 2 goes from row 15-266
  • 21 goes from 441 to 485
  • 22 which I want goes from 487 to 509 or about 23 rows

I’ll try setting the cutoff at 40 members. I like this because I can now see what is going on:

This reduces the number of clusters and changes the cluster numbers, but that is OK. I’ll just put another heading on my Summary Chart:

I brought these new restricted clusters into Access and compared them:

Here I have only 5 clusters. That made a huge difference. Without the 40 member cutoff, I get 31 clusters. Life just got simpler.

Above Cluster 9 goes to Clusters 1 and 2. Cluster 9 was the one that I had split previously. Here are the new Clusters 1 and 2:

Cluster 1 is the Kerivan side and Cluster 2 is the Butler side.

This is interesting because former Clusters 10 and 11 reclustered into the new Cluster 1. Now I’m curious to see how the unfiltered Clusters would have mapped:

This goes off the page above. One unexpected result is that under this configuration the previous Cluster 9 now only maps to one Cluster 4. However, now the old Clusters 10 and 11 map to their own clusters. With the restriction of the large clusters, I was expecting no effect on the Irish clusters, only on the large French Canadian clusters. However, the restriction also had an effect on the Irish Clusters.

The bottom line is that I could work with either the restricted or non-restricted clusters, however, for the purpose of looking at the Irish clusters, I would rather deal with the restricted clusters.

Irish Cluster 1

At 30 cM with a restiction 40 per cluster, I get two Irish Clusters. The first one is the Kerivan side. The first person on the list is not in the cluster but has matches to Cluster 1. He has a small tree with a Hanrahan in it. The fourth on the list shows a correclation to Cluster 1 and has a public tree. I don’t see any obvious match and would prefer not to build out this tree.

The last person before the Kerivan Cluster 1 has a curious common ancestor named Mary:

This looks worth pursuing. The relationship is shown as half first cousin, but it is likely a full cousin relationship unless Mary married two different Rooney’s. Here is Suzy’s tree starting with her Kerivan grandfather:

The Ancestry shared ancestry tree looks suspicious also because it appears to show that there were two brothers named Timothy Rooney. I’ll try to recreate Daniel’s tree to see what is going on. I was having trouble with the tree, then I found a chart I had made a while ago:

This shows John Henry descending from a first wife of Timothy Rooney. I’ll put that on my new Rooney Tree:

I’ve ended up with a half relationship after all. However, it now appears that Daniel and Suzy are 1/2 third cousins once removed. I tried to distinguish the two branches with different colors.

The next person in Cluster 1 who I haven’t looked at yet is FG:

FG has an unlinked tree:

I noticed the name McCusker:

I like the easy ones. See, I’m a Rooney specialist.

Andrew from Cluster 1

The next person in Cluster 1 with a usable tree is Andrew:

After a little snooping around Ancestry, I came up with this:

Andrew goes on my Rooney Tree:

That brings up a point. The first three matches are not Kerivan matches. Technically, that makes Cluster 1 a Rooney Cluster.

Russell and Sandra

Russell and Sandra should be easy as they show Common Ancestors at Ancestry:

Bob’s Tree Looks Hopeful

Bob, who is in Cluster 1, has this tree:

Here is a quick tree to get him on my Chart:

Donna and John

I have a note that Donna is the mother of John and that I wrote a Blog about John. Here is John at Ancestry:

It looks like a good time to start a separate Kerivan Tree:

I found Donna’s sister in the mix also.

H6 and a More Challenging Tree?

H6 is in Cluster 1 with a match of 141 cM:

These families lived in New Jersey and some came from Sweden or Italy, so I had trouble getting this back to Kerivan easily.

On To Butler Cluster 2

The first two with small unlinked trees are Kerry and Janice.

Kerry’s tree:

Janice’s tree:

 

I’m thinking it would be a lot of work to track these trees down. Here is some more on Mary Ann:

I decided to look into this tree based on the fact that I also found a Rooney in the ancestry. Mary Ann’s mother was Rose Rooney.

That could explain the link of Cluster 2 to the Rooney Cluster 1:

Butler’s I’m Already Tracking

 

I already had Donna, Janice and Barbara. I added Harry also from Cluster 1.

Suzy’s Clusters of 40 Members or Less at 25 cM

I checked Suzy’s clusters at 25 cM without the 40 member limit and she had 69 clusters. With all those clusters, I couldn’t see all the clusters on the chart even at the minimum 10%. With the 40 member limit, Suzy had 12 clusters:

 

 

I’m curious what these clusters will show.

Running the 25 cM Clusters by Access

I’ll compare these clusters to the 30 cM clusters in Access.

This says that Clusters 1 and 2 both split three ways. That means that there were also 6 new clusters. I expect the new clusters were a mix of Irish and French Canadian. It seems odd that only the two Irish Clusters that I was working on at the 30 cM limit mapped as I had 5 clusters previously. I’m guessing that the other three clusters mapped clusters of over 40 members, so were screened out.

Cluster 1

I don’t like how Cluster 1 doesn’t cluster well, but I’ll take a look anyway:

In looking over these 4 mathces, I don’t see an obvious connection. Something more may show up at the 20 cM Cluster run.

Cluster 2

This Cluster seems to have an affinity with part of Cluster 7, which looks like a Kerivan Cluster.

Cluster 3

This mapped from previous Cluster 1. I see one person who has a Rooney in their ancestry. This is a person I looked at in the previous Cluster 1:

I’ll put Cluster 3 here:

However, I’ll have to go back and check in Access to see if that looks right:

It is right as it mapped back to Cluster 11 at the 35 cM limit. Also I see that the Cluster 1 at 35 cM that I had looked at above came from the previous Cluster 2 at 30 cM which was a Butler Cluster.

Cluster 4

Cluster 4 is new. Here is the tree for Suzy’s first match in Cluster 4:

Ancestry has Rooney in green as it notices that name in Suzy’s tree. Thomas Rooney above was said to be from County Roscommon. Here is the largest tree from Jean of Cluster 4:

 

Timothy Rooney from Suzy’s tree is supposed to be from Leitrim. Daddy Mick Roone is supposed to be from Galoway (Galway?). Thomas above shows from Roscommon. They could all be right and all be related.

Cluster 5

The person above Cluster 5 seems associated with Cluster 5 and has a tree. However, that person also has connection with Rooney Cluster 3. Here is some Rooney information from the tree:

Here is more of the tree:

Suzy has both Rooney and Sullivan ancestry:

The question is whether Cluster 5 is a Sullivan or Rooney Cluster.

I’ll build out the other unlinked tree from Kerry in Cluster 5:

I’m guessing the connection is on the Gilmartin side as that name sounds familiar. However, I couldn’t get the tree to match with a familar surnam, so I gave up for now.

Anyway, based on the Clusters, I should be looking on the Bulter side for these Cluster 5 matches:

It is easy to get thrown off my conmon surnames. That is why we have to pay attention to the shared clusters. However, having said that Cluster 5 did seem to match both Clusters 4 and 6. So for now, I’ll put Cluster 5 on the Butler side:

Cluster 6 – A Butler Cluster

This is an interesting Cluster because it includes a Cincinnati branch of Butlers who are related to Suzy’s Butlers, but we don’t know how yet. Here is my guess how they could match:

The George Butler Branch from Cincinnati is on the left. The Edward Butler branch who lived for a while in Cincinnati ended up in the Boston area.

Within Cluster 6 is the Boston group and the Cincinnati group:

The last two matches that I don’t have a green box around are closer Butler relatives.

Cluster 7 – Back to the Kerivan Side

I don’t see any new matches in this group, so no new news.

Clusters 8-12

These are all new clusters:

I don’t know if these are Irish or French Canadian Clusters.

On to the Final Clustering at 20 cM

In this run, I’ll choose 6 cM and have a maximum cluster size of 40 keeping in mind that Ancestry’s shared matches are generally at the 20 cM size or larger. This gives me 20 Clusters and some greater detail on what I had already.

First I’ll pull this new information into Access to see how the Clusters map:

This cut off a few clusters at the top. First I see in the 7th row from the bottom, Cluster 2 that went to 1 has no cluster in the last row. That was the cluster that I was wary of.

Here is a sort by the previous run of Clusters:

This tells me the new Cluster 1 is from the previous Cluster 2. The new Cluster 2 had no precedent. One large tree in Cluster 2 had a Crowley ancestor from Cork, but that may be a coincidence.

Cluster Three maps to previous Cluster 9 which I haven’t looked at. I see another Gilmarting in Cluster 3. His ancestry goes back to Leitrim, Ireland. After clicking through a lot of connections, I would guess that the connection is through Ireland. Also a lot of the connections are in Leitrim and I found one tree with a Rooney in it from Leitrim:

Here is an interesting match between Clusters 3 and 4:

This wants to tie together the Jeremiah and John Rooney Lines.

ThruLines suggests another Rooney Line:

Skipping Down to the Butler Clusters

The new Cluster 11 mapped to the previous Cluster 6:

There is nothing new in Cluster 11 itself, but the interesting part is outside Cluster 11 proper:

However, there are four matches above Cluster 11 that show association with Cluster 11. The first person is managed by Patty who I already know about:

Patty is on the George Butler Line. George Butler came to live in Cincinnati.

The third person in the group of matches that are correlated to Cluster 11 is Regina. Here is her tree:

I recognize the Branch name above.  Here is the marriage record for Alma from 1925:

Another person in the group that is correlated with Cluster 11 is Ryan. Here is his tree:

Both of Ryan’s parents were born in Cincinnati. I need to build yet another tree to see if I can find a match:

With a little snooping around, I was able to get back to Alma Branch:

Christy is on the other side of the Cluster 11:

She fits in on the Boston George side:

Emily married an unrelated Butler.

Next under Christy is Larry:

Larry matches clusters on either side. Here is here small tree:

The tree is interesting as both parents are from Wexford. Some of the Butler ancestors were also from Wexford. Unfortunately, the tree is a bit confusing to follow. Was Bridget’s maiden name Browne or Hesse?

Larry has shared matches with Suzy, Brian, Barbara and Janice:

That seems to make Larry a sort of missing link.

Summary and Conclusions

  • I did a first shot at looking at Suzy’s Shared Clusters
  • Not too far into looking at Suzy’s clusters, her French Canadian Clusters got very large.
  • I restricted the size of Suzy’s clusters so I could look more at her smaller clusters on her Irish side.
  • This lead me to some interesting finds on her Kerivan and Butler side.
  • One Butler match was very interesting as it was shared with two Butler branches that I have been trying to connect.

 

 

 

 

 

 

 

An Analysis of My Mom’s Shared Clusters

I’ve been playing around lately with John Brecher’s Shared Cluster Program. I wrote a series of three Blogs looking at my owned Shared Clusters. The last Blog is here.

Downloading My Mom’s AncestryDNA Matches

First I signed into my account using the Shared Clustering program. I chose my mom’s kit and downloaded it.

I used the middle button which says ‘Slow and complete’. It was supposed to take several hours but only took 32 minutes. My mom has 26947 total matches and 438 fourth cousins. That is about half the matches that I have.

First Try: 50 cM Gets Two Clusters

My first try was conservative at 50 cM. I only got two clusters. I was sure to name the output file indicating that it was for 50 cM:

This gave me two named clusters. I say named, because I see that there are probably actually four clusters altogether. My guess is that these represent her four grandparents. The top cluster represents my mom’s mom who was a Lentz.

By the way, my analysis looked like this:

I have a green box around my mom’s side. I will be expanding that like this:

For now, I’ll keep it simple with a paternal and maternal side for my Mom:

3 Clusters at 40 cM

These clusters are in no hurry to separate. That is partly due to my mother, her niece and two grand-nieces being in Cluster 3:

One person I know in Cluster 2, Otis, descends from a Schwechheimer:

Also Gangnus, but more distantly.

At 35 cM, Mom Has 6 Clusters

Now my mother’s maternal grandparents, Lentz and Nicholson are starting to separate. Nicholson is Cluster 2 and Lentz is Cluster 4:

Here is a simple comparison of Clusters between the 40 cM and 35 cM cutoffs:

This shows that the Lentz 1900 went to Lentz and Nicholson as I have them. That means I need to figure out where Clusters 1 and 6 go. These are new clusters with no precedent. All three in Cluster 1 match my mother’s Rathfelder cousin, so Cluster 1 could be Rathfelder or Gangnus:

Cluster 6 also appears to be on my mother’s father’s side which could be Rathfelder or Gangnus. I’ll put them under Gangnus as I have nothing else there so far.

My Mom’s 10 Clusters Down at the 30 cM Match Level

My MS Access query gives me this:

At the 30 cM level, this shows that Clusters 3, 6 and 10 are new. The previously new Clusters 1 and 6 at the 35 cM level map to new Clusters 9 and 1. Cluster 7 is interesting it has no precedent at the 35 cM level but one of those non-precedents was previously a Cluster 2 (Scwhechheimer) at the 40 cM level. So that may be a clue. Here is how these 35 cM clusters map at the 30 cM level:

Three maps to 8 and 7. 5 maps to 2. Technically 5 did not map to 7 but someone who was in the precursor to 5 mapped to 7. This probably gets into intermarriage in the German colony of Hirschenhof, Latvia where some of my ancestors lived. I suppose I could have omitted the second 7. Here is Cluster 7:

It looks to have two parts to it.

That leaves Clusters 3, 6 and 10 to map. Because I compared the clusters above in Access, I didn’t have to know anything about who was in the clusters when I mapped Clusters 1, 2, 4, 5, 7 and 8. However, for Clusters 3, 6 and 10, I have to know something about these people to put them in the right places.

Finding a Home for Cluster 3

Cluster 3  at the top left is part of a larger Cluster 3-4-5 complex:

The is a large area below the second green box which can be ignored as that would be my mother’s close relatives. The second green box is Lentz. Below the close relative area is Nicholson. One person in Cluster 3 matches 2 people in the Lentz Cluster. However, 3 people in Cluster 3 match people in the lower right Nicholson Cluster. So I’ll say this is a Nicholson Cluster. The Nicolson’s came from Sheffield, England. One person in Cluster 3 has a tree. This tree shows both of his parents being from England.

Cluster 5 – Consider the Source

Here I have both Lentz and Nicholson mapping to Cluster 5. How does this make sense? The ones that map from Lentz (4) to 5 are my mom’s close relatives, so naturally, they would match both sides. I’ll take the Cluster 5 that comes from Cluster 4 out. That is one of the disadvantages of mapping the numbers without considering the source.

Cluster 6 – Rathfelder Side

Here are my notes on this small Cluster:

My Rathfelder grandfather was a German from Latvia. So by geographical phasing, that puts Cluster 6 somewhere on his side.

Cluster 10 – Mystery Cluster

Cluster 10 has strong internal matches indicated by the dark color of the Cluster. However, I don’t see any other matches with other groups or clusters:

Two of the matches have large trees of over 25,000 people. Comparing my mom’s tree and this large tree brings up the name Clayton:

However, JK also shares these areas of ancestry with my mom:

Here I have banished Cluster 10 to the Unknown Realm:

So now I am ready to dive into new depths of my mother’s clustering ancestry.

Mom Is Up To 18 Numbered Clusters at 25 cM

25 cM is the next to the last stop for our analysis. This is an important level, because at this vantage we can look toward the more recent 30 cM clusters and will be able to look next at the further in the past 20 cM clusters. The progression for the numbers of clusters has gone 2, 3, 6, 10, and now 18.

This is the overall shape of the 18 clusters

My mom’s close relatives have split into two clusters. The previous Cluster 10 of unknown origin is getting bigger.

Putting Access to Work Again on the 25 cM Clusters

I imported Columns A-L to Access and added it to my ongoing query:

When I view this query, I get this:

 

I’ll sort by new clusters to see which ones are new:

Next, I’ll map the non-highlighted new clusters. Here is what I get with my blind comparison:

Clusters 7 and 8 are on my mom’s maternal and paternal side. I suspect that is a result of her close relatives matching both sides. I’ll search Access for Clusters 7 and 8 and add names:

Here I see Cluster 5 going to Cluster 8 only with two close relatives, so I’ll remove that from my analysis.

However, we are still left with two Clusters 7. One is on my mother’s maternal side and one on her paternal side. That means that Cluster 7 is a compound cluster due to close relatives being there and matching both sides:

That means my mom actually has at least 19 Clusters at this level:

Ancestry uses 20 cM for a fourth cousin cutoff. The fourth cousin level represents 32 ancestors – in theory.

Next, I just need to go down the Clusters and fill in the blanks:

Clusters 1 and 2 – Sheffield Area

One person in Cluster 7a matches all four people in Cluster 1. Cluster 2 goes back to a known Nicholson ancestor born in 1765. That would bring Nicholson down to 1798 on my mother’s line in the next generation. Due to the fact that we know that ancestors, it is possible that Cluster 1 could be on Nicholson’s wife’s Clayton line:

 Cluster 3 Affinity to Cluster 7b

Cluster 3 also has an affinity to Clusters 4 and 14. That brings up an error in my Chart:

At 35 cM, I had both Clusters 3 and 5 mapping to Cluster 4 at 30 cM. A review of my Access query shows one to be wrong:

Cluster 3 only mapped to Clusters 7 and 8. Here is the corrected version:

This is bound to be a little messed up because my mother’s father had Gangnus and Schwechheimer on his paternal and maternal sides:

 Cluster 5

Fortunately, I have already done some work with Robert in this Cluster:

Here is a confusing tree. The confusing part is that the common ancestor is with Gangnus. Gangnus is also the wife of my mother’s Rathfelder grandfather. That means that this Gangnus goes in on my mother’s non-Gangnus side!

Actually, it appears that Robert is equally related on both sides:

In fact, as Otis is not in Robert’s shared group, I need to put Robert back on the Gangnus side. Good thing I’m flexible. Patrick doesn’t figure in as his results are on MyHeritage.

I needed someone on the Gangnus side that I could identify.

Summary of the 25 cM Clusters

Here I dumped more new clusters into the Unknown bucket. I don’t have much under Lentz, so some may belong there.

The Final Run of 20 cM for 36 Clusters

I run this at 6 cM which is the minimum, but the minimum shared matching at AncestryDNA is set at 20 cM. Some of the extra matches between 6 and 20 cM could give hints from the Correlated Cluster Column.

Probably the easiest part of the comparison is using Access. Here is what I can fit out of the 83 rows of results:

Here is what the mapping looks like except for Clusters 7a and 7b:

Both Clusters 10 and 5 mapped to Cluster 24. That probably has to do with the interweaving of Schwechheimer and Gangnus in my mom’s Colony of HIrschenhof ancestry. In order to map Clusters 7a and 7b, I bring in the 30 cM clusters to see which 7 I am looking at:

Here, Cluster 2 maps to 7b and Cluster 5 to 7a. This still gives some confusing results:

However, when I filter out my Mom’s close relatives of over 900 cM, I get this:

Here is how that looks in my summary chart:

I couldn’t decide whether to put Cluster 12 in under previous Cluster 7b, so I put it in parentheses. It looks like I mapped 19 clusters not counting those mapped more than once. That should leave about 17 more to map – or drop into the unknown bin.

Mapping the New 20 cM Clusters

Here are some new clusters:

All I have to do is to look at each of these clusters. My best hope is if they match another cluster or have a common ancestor or if I have some notes on them in Ancestry. Occasionally, looking at a tree or building out a tree will help, but I rarely have luck that way. One thing I find difficult is that my mom had German ancestry on both sides. One side came to the US before the American Revolution. Another side went to a German Colony in Latvia and then came to the US in the 1900’s. However, both sides were originally in Germany.

Clusters 28 and 29

Someone in Cluster 29 had a Nicholson ancestor, but it may be coincidence:

Sarah was from Bolton in the above tree. My mom’s Nicholson’s were from Sheffield, but perhaps not always from there:

The Large Cluster 30

It looks like one of mom’s unknown ancestors had a lot of descendants.

And the Answer Is…

It seems like I’ve increased my knowledge of the unknown (poor pun intended). I still have a shortage on the Lentz side. I’m not sure if there were just not many offspring or what happened. I have at least one early common ancestor at Ancestry, but that match is not in a cluster. Another question would be to identify Cluster 10 at 30 cM. I feel like I have a better idea of the weaving of DNA and families between Gangnus and Schwechheimer in my mother’s ancestry.

Summary and Conclusions

  • The combination of the Shared Clustering program and MS Access help to make a fairly quick analysis of AncestryDNA shared matches
  • My mom’s genealogy is a bit tough. I know she has some more Latvian matches out there but tracing their trees back is difficult in the period when Latvia’s records were in Russian.
  • I don’t have an explanation why my mother’s common ancestors of Lentz and Baker in Philadelphia were not part of clusters.

 

 

 

 

Walking My Clusters Backward and Forward Using MS Access

In my previous Blog, I came up with a way to show my Clusters and how they changed (or didn’t change) between Shared Cluster runs at different DNA match levels. The Shared Cluster Program is from Jonathan Brecher and the encouragement to walk my cluster back was from Jim Bartlett. As a result of my previous Blog, I thought that it might be easier to walk my clusters forward instead of back. Usually in genealogy work proceeds from the recent to the more distant path and from the known to the unknown. However, after looking at the clusters, it seems that in some ways the older clusters are more specific.

Using MS Access to Walk My Clusters Forward

I used to have some knowledge of MS Access. I wonder if I can resurrect that. I want to take my 6 cM run from my previous Blog and compare it to my 25 cM run. I hope to map 50 clusters down to 27. So I am reducing the clusters by about two to one.

I’ll start with  a new Access database:

This database had a blank table in it for me to add data to. I didn’t want that, so I got rid of it. I’ll go to External Data and import my 6 cM Shared Cluster Run. I feel like I should close out my open copy first.

I chose New Data Source and found my file. When I sorted by Date modified, my file rose to the top.

This leads me to an import wizard:

I probably won’t import all the information. I clicked the box saying I wanted my first row to be the Column Headings.

Actually, I see a better way to do this. I went back to my original file and copied columns A-L:

I put these into a new file, saved it and imported that into Access. I don’t need any of the 2500 or so columns after L. This brought me to the import wizard again. I pressed next:

Access want to add a primary key or to choose a primary key. The test ID is an ID, so I’ll use that. Otherwise, Access will assign the numbers of 1 to about 2500 to identify my new database.

Next, I give my table a name:

This now shows up in my Access Database:

Next, I do the same for my 25 cM run as I want to compare the 6 cM run to the 25 cM run.

Querying Access

The next step is to use these two tables to see how the 6 cM Clusters look like at the 25 cM level.

I’ll choose Create > Query Design:

I’ll choose these two tables. That will put them onto the blue area, then I’ll connect them by the key ID:

I wasn’t sure what to choose from the tables above to put into the query below. I put in the name. This has to be the same in each table because the Test ID in each table corresponds to the same name. Most important is to map the 6 cM Cluster Number to the 25 cM Cluster Number. I have those columns plus the 6 cM Correlated Cluster Number, the 6 cM Common Ancestors and 6 cM notes. The notes should be the same for each table. I didn’t add the 25 cM Correlated Cluster Numbers. I can add those later if I find I need them.

Next, I choose view to get the results of the query:

I get 378 results. This makes sense as there were 378 results in the 25 cM Shared Cluster Table. For some reason, Access ordered the results by the number of the 25 cM Cluster Number. This isn’t so bad.

Easy Results

Next, I’ll move the 25 cM Column in closer so I can compare the two sets of cluster numbers:

Let’s look at the 25 cM Clusters 1-3. Walking back, Cluster 1 went to mostly to 23, but one went to 24. Cluster 2 went to 24 twice and 32 once. Cluster 3

only went to Cluster 16. My guess is that the mixup between Clusters 1 and 2 have to do with the fact that these Protestant Irish families intermarried.

A Flaw in My Logic?

I’m just thinking that I need to do this exercise at least one step further back as I haven’t yet finished filling in my 25 cM Clusters. I’ll see what I can do with just the information I have sorted out from Access so far.

Without any sorting of my query, it appears that Cluster 40 comes out of nowhere. However, when I sort by the 6 cM clusters:

Here is a situation where Cluster 40 must be a compound cluster. Here is what I see:

 

The smaller square at the top left may be Lentz. The single dark red square in the bottom right is Nigel. I mentioned him in my last Blog. He represents an old  Nicholson DNA match.

Nigel would not match the Lentz side which appears to be the top left side of the Cluster. There appears to be other sub-clusters within the Nicholson Cluster which may represent Ellis or Clayton highlighted above.

Because, I’m in the area of my 25 cM Clusters that I didn’t finish last time, I want to do a new Access Query

Comparing My 30 cM Clusters to My 25 cM Clusters in Access

Here is what my new query looks like:

I’ve connected the two tables by the Test ID. Then I compared the 25 cM Cluster Number to the 30 cM Cluster Number. Plus I added some more information that I though to be helpful. Actually I had the Name column twice which was not needed.

This is my Nicholson section. This shows that the 30 cM Cluster 14 mapped to two clusters at 25 cM. I’m glad I did this, because Cluster 25 splits between Lentz and Rathfelder due to my mother being in that Cluster:

This brings up another issue. At 20 cM, my Rathfelder 2nd cousin match aligned with my maternal 1st cousin and her two daughters. This caused them to form a single Cluster that mapped from a previous two Clusters (25b and 11). It’s not a big deal, but I had to just enter 38 twice to account for it. To be consistent, I’ll call them 38a and 38b:

 

Here the green square represents 38a with my first and second Rathfelder cousins:

The matches to the lower right of the green box would be my more distant relatives with ancestry in Hirschenhof, Latvia.

Finding Cluster 22 at 25 cM

When I compare my 25 cM Cluster to my 30 cM clusters:

Cluster 22 at 25 cM has no precedent at the 30 cM level. That has to do with the match level of Cluster 22. Here is a comparison with it’s corresponding 6 cM Cluster:

All these matches to me were under 30 cM. So basically, Access makes it so you don’t have to keep switching back and forth between the different results.

Solving the Puzzle: Filling in the 20 cM Clusters

The 20 cM and the 6 cM Clusters are the same. The Shared Cluster can look at how the matches between 6 and 20 cM are associated with Clusters but Ancestry doesn’t make shared clusters at less than 20 cM.

I have some filled in already, but need to get up to 50 clusters:

The first Cluster 19 mapping is an issue:

25 cM Cluster 19 maps to Clusters 6, 7, and 15 at the lower cutoff. Further, I can see at least three clusters within the new Cluster 7:

The two green squares represent points of interest for me. They appear to represent my Hartley English genealogy.

Backing Up a Step: 30 to 35 cM Clusters Compared with Access

Another way to use access is by using an unequal join:

The unequal join is represented by the left to right arrow above. That says, show me all the cases where there is a value in the 30 cM cluster table that is equal to a value in the 35 cM cluster table. And it adds in all the remaining 30 cM clusters that aren’t included in the 35 Cluster Table. I think that that is what I want. I could have had the arrow go the other way, but would have gotten different results.

Just for fun, I’ll do it both ways. This is the way from above:

This shows that Cluster 2 at the 30 cM level had no corresponding cluster(s) at that 35 cM level. Cluster 4 only had one corresponding Cluster. And in that corresponding cluster, Cluster 4 had a correlated Cluster 7.

Here are the results with the arrow going the other way. Think of the previous results as walking the clusters back and this new one as walking the clusters forward:

Here Cluster 2 doesn’t even show up at the 30 cM level. They both give about the same information, but the walking forward comparison should provide more detailed information.

Cluster 7 at 35 cM

I’d like to take another look at this as I had ignored this Cluster previously. Based on the above query, it turns out Cluster 7 is quite important. This is Cluster 7 at 35 cM:

The first two people I have as coming from Nantucket (my paternal grandfather’s side). The second two I had guessed as being from Ireland (my paternal grandmother’s side). I don’t think that they belong in the same Cluster, so I’ll tree this as two clusters.

Putting It All Together with Access

Here is a more complicated Access Query:

 

Above, I’m comparing the 6 cM table to the 25, 30 and 35 tables. I have the link between the 6 cM Table and the 25 cM Table a right handed link, so I’ll see all the 6 cM (or 20 cM) Clusters. I then put the 35 cM Clusters first so they will be in the order of my Summary Chart:

At 35 cM I have 10 Clusters. Cluster 1 maps to Cluster 14 at 30 cM. 14 Maps to Cluster 26 at 25 cM. Cluster 26 maps to Clusters 40 and 42 at 6 cM. However, note that other clusters are mapping to Cluster 40. That is because we have Lentz and Nicholson in Cluster 40 as well as those who only descend from  Nicholson, if I understand it correctly.

Cluster 2 at 35 cM includes my mother. Her clusters go from 2 to 14 to 25 to 21. The correlation is one to one until we get down to the 20 cM level. At that point the cluster splits into three where Cluster 40 is a child of Cluster 1. That makes me think that Cluster 40 will be a compound Cluster.

I see at least three clusters for Cluster 40 at the 20 cM cutoff:

Another Query

This one is more like the manual comparison that I did previously:

This query says take everything in the 6 cM Table plus those things that match inthe 25 cM. Then do that for each Table. That query appears to give me everything as it includes 2,452 rows:

Here is a stripped down version of this query:

Here I only include the Cluster Number columns:

This gives me 135 rows. This also points out that Cluster 6 and 10 at the 35 cM level both map to Cluster 7 at the 30 cM level. This makes sense when you look at Cluster 7:

This is a distinction that I had missed in my original mapping chart:

This corrected chart better reflects what Access is showing me:

Access then shows this:

The part of Cluster 7 that was from 10 goes to 18 and then to Cluster 5. This was not reflected in my previous summary chart:

Here is the correction:

I’ll also add in my Cluster 7 at the 35 cM level. I didn’t add it in previously as it had 2 people with Nantucket shared ancestors and 2 people with suspected Irish shared ancestors.

Mapping Cluster 4 at 30 cM

Here I have another situation like my Cluster 7 above:

This is another 4 person Cluster. And, like Cluster 7, the first two people seem to match on my Hartley side and the second two seem to match on my Frazer or Irish side. This is reflected in my Access query:

The split goes to Cluster 8 (Irish) and 14 (Hartley ancestors).

Mapping Cluster 14 at 30 cM

Here is the work I did previously in my Cluster Summary Chart:

I had split Cluster 14 in three. These are my major maternal lines. Because my mother was in this Cluster, she was related to both these sides. Lentz and Nicholson are well-tested, so I have some good separation there. This confusion is reflected in my Access query:

The Common Ancestor Column above shows three different sets of Common Ancestors. They represent the needed splits for Cluster 14. This is where the Access results come in handy. In Access, I have a 14 Cluster going to Cluster 26 which goes to 27. I had missed that in my previous analysis and only had 27 going to Cluster 40.

My Irish Cluster 15 at 30 cM

Here is what I had:

The Access query shows some additional subtleties:

This shows Cluster 2 at 25 cM going to Cluster 24 at 20 cM. However, that doesn’t mean my other Cluster 2 to 32 is wrong:

Two other people in Cluster 2 probably would have mapped back to Cluster 15 but the DNA match was not high enough to be in Cluster 15. Also note that Cluster 24 and 32 both have Common Ancestors:

The question is, if these three have the same common ancestors, then why are they in different clusters? The answer appears to be that one Cluster (24 or 32) would represent McMaster and the other Frazer due to the two common ancestors.

As a result, I split out these two clusters like this:

McMaster 1829 is Fanny in Cluster 2 above. I now have Clusters 24 and 32 as her parents. Consider it a theory.

The Good Enough Product

I have slimmed down my chart to show between 40 cM and 20 cM. I started with 5 clusters in the 40 cM Column which is enough to describe my four grandparents. In the 20 cM column, I didn’t feel a need to describe each of the 50 clusters. However, I was interested in describing each cluster in the 25 cM column and it’s corresponding cluster at 30 cM and 20 cM. The 25 cM clusters were at a good vantage point where I could check the clusters on either side using Access. The place where I may be more interested in detailed 20 cM clusters would be for my English Hartley side.

Summary and Conclusions

  • Jonathan Brecher’s Shared Cluster Program is good for sorting out your clusters
  • The use of MS Access makes it easier to see the nuances of how the clusters merge or separate between the different lower match thresholds
  • There is a question of my mind concerning the level of accuracy needed in this analysis. It’s good not to be so detailed-oriented that you miss the big picture. The big picture for me is whether the cluster is in the right grandparent group for me.
  • This was my first shot at using Shared Clusters to sort out my Ancestry Clusters. I’d like to try using the Shared Cluster program on other DNA kits that I administer. Perhaps my mother’s results would be the next logical step.

 

 

Walking Back My Clusters: Part 2

Part 1 of Walking Back My Clusters was long and rambling. I learned a few things, looked at a few family trees and reached out to a few DNA matches at AncestryDNA.  While writing my previous Blog, I came up with a better way of presenting the results of walking back my clusters. I realize that this may sound obscure if you are not already into genetic genealogy and clusters, but hopefully the readers understand the basics of DNA and clustering.

The New Cluster Results Format

Here it is:

I have my four grandparents in four colors. The thought is that even if I am lost as to what a cluster represents, I should know under which grandparent the cluster belongs. At the top, I show the cM cutoff for the clusters. I have a small column for the cluster number that the program produces. This is a relative number and changes for each analysis. To the right of the cluster number, I have the name of the closest surname that cluster represents and the date that ancestor was born. If I don’t know this, I may give a geographical hint. As far as which ancestor to use, it is somewhat subjective. On the top row I have Hartley going to Pilling 1802 and Snell 1866. It may have made more sense to use an earlier Hartley instead of Pilling, but I suspected that one of the people in that particular cluster went back to Pilling. Under Lentz 1900, that went to the two parents who were Lentz 1866 and Nicholson 1865. This new representation, so far, keeps everything close together where I can keep track of where the clusters are going.

Cluster 13 on the 30 cM Limit

This is where I left off on the chart above. There are only three in this Cluster. I have a note on one of the match’s that they have a Northern Ireland background. I’m going to peek forward to 25 cM to see if I get any more hints. This adds one more match and tree. This tree in addition to the match with the largest tree has Canadian ancestors. I’ll take a look at the largest tree in Cluster 13:

This particular match had Ontario ancestors. The parents had connections to Owen Sound, Ontario. That sounds familiar from one of my distant Frazer relatives. My hunch is that the connection is not on the McRae side as they are listed as being originally from Scotland and Presbyterian. I’m not aware of Presbyterian ancestry on my Frazer side.

Here is my best guess:

Jane or Jennie in the bottom right of the tree is from Inniskillen. I assume this to be a variant of Enniskillen, where a lot of my DNA match leads take me:

All that to make a guess at Cluster 13.

Cluster 16 on the 30 cM Limit

I’m a bit stuck on this one. I think it is on my maternal grandfather’s side. Here is what I have so far:

I know that at 20 cM, I have 50 Clusters, so I have a way to go.

25 cM Clusters

Here I have 27 clusters, but some may be compound clusters.

Clusters 1 and 2

These split out the previous Cluster 15 which I had assigned to Fanny McMaster born 1829. Let’s take a second look:

Whitney in Cluster 1 matches everyone in Cluster 2. That is because he is a closer relative to me than I am to others in that Cluster:

What is also confusing is that Margaret McMaster had two McMaster parents.  I had the previous Cluster 15 correct on Fanny McMaster. However, it would be easier for me to think of this now as having the old Cluster 15 on Margaret McMaster 1846. Then I could assign Fanny to BV and mt and James McMaster to Whitney. Here is how I’m related to Whitney:

I am Whitney’s third cousin once removed. Here is one case where I changed an earlier analysis based on a later one:

After going through some more clusters, I came up with this:

I mentioned that I Had 27 clusters at 25 cM and 50 clusters at 20 cM, so I gave up doing this for now. I think I have an easier way to go about this which I will explore in my next Blog.

One interesting thing above is that the orange Rathfelder line jumps from 1921 to 1819 in the above cluster summary. My explanation is that there were not many DNA matches for that line at Ancestry. That line represents my mother’s father who was from Latvia. He jumped ship and came to the US in 1916. I have has one Rathfelder 2nd cousin once removed who tested at Ancestry, but one person is not enough to form a cluster at that level.

Going All the Way to 6 cM with Shared Clusters

The creator of the Shared Cluster Program commented on my previous Blog and recommended I take the clusters down to 6 cM. John Brecher tells me I won’t get any more clusters but more matches associated with those clusters. At first I thought that I had to leave the “Lowest centimorgans in shared matches” as 20, but that gave me the same results as my last run using 20 for both values in that row. So now I have both values set to 6 cM:

This kicked up my spreadsheet from 912 rows to 2453 rows. I suspect that this is where the Shared Cluster Program really shines.

Filtering the 6 cM Results

Excel has a filter button. I would like to filter my results on Common Ancestors:

When I choose Filter, an arrow appears in each column’s heading. I click on the arrow under “Common Ancestors” and unclick the ‘Blanks’ option which is at the bottom of the list:

That will give me each row that has a common ancestor:

I couldn’t get all my results in one screen shot, so the top is cut off.  Cluster 7 appears to have many of my 2nd cousins, so it shows other more distant clusters that they are related to. The 13 is highlighted in the Correlated Clusters column because it gives a clue to Cluster 13 with common ancestors Snell and Luther that I didn’t have before for Cluster 13. The same is true for associated Cluster 19 with common ancestors of James McMaster and Fanny McMaster. If I add up all the clusters plus associated clusters that have Common Ancestors, that adds up to about 20. Those will be a good clues to identifying my 50 clusters.

I highlighted Nigel because he is an interesting case. He has a fairly high DNA match with me. He’s my 5th cousin, once removed:

I don’t recall Nigel being in a cluster before due to the distant of his relationship to me. So it is good to see him in Cluster 40 now.

On to the Next Blog

Part of the difficult part of comparing these Clusters is cross checking between say, a 25 cM analysis and a 20 cM analysis. For example, Charlie was in Cluster 35 at 20 cM. What Cluster was he in at 25 cM? I hope to figure out a way to make that a little easier in my next Blog using MS Access. There may be other ways. It makes sense to me also to walk the Clusters Forward instead of back. That is because the older clusters have more people in them. As noted above they also have about 20 identified Common Ancestors.