Category: DNA General

Posted on

A New Frazer DNA Match for Jane at Ancestry

I had an email from Jane recently about a new Frazer match she had. Jane’s match is with JD. His tree is private but searchable. That means that Ancestry can figure out likely common ancestors with his tree even though it is private. Here is how Jane and JD probably match up:

The Archibald above is from what I call the Stinson Line:

Archibald Frazer and Ann Stinson had other children, but I only show one who was another Archibald who married Catherine Parker.

Evaluating JD’s Tree

Ancestry suggests I evaluate JD’s tree, so I will. I created a tree for JD. I assume that he knows who his mother is. I have to further assume that the connection is on his mother’s side. In 1940, JD’s mother Mary was living in Bridgeport, Connecticut:

From here, we need to get the family back to Alexander Frazer who lived in County Roscommon, Ireland. So far, JD’s maternal side is checking out:

Carrie or Katherine?

Here is the 1920 Census for Bridgeport:

The one I have as Katherine in the tree appears to be Carrie. Her mother (also Carrie) is from Irish and speaks Irish. The 1900 Census has Carrie the mother immigrating in 1890, born in Ireland to parents who were born in Scotland. I seem to recall from Jane that her ancestors did move to Scotland and then back to Ireland, so this Census makes sense.

ThruLines for Jane

I see that Jane has many matches to the children of Alexand Frazer:

JD is right in the middle along with descendants of five of the children of Alexander Frazer. This seems to fill in where some of the records are missing for Carrie Frazer Ember. I like these ThruLines as they tie together many families in one image. Here is what I have for children of Alexander Frazer:

Apparently Alexander Frazer had 12 children.

Actually, I now see the marriage record for Charles Ember and Caroline Frazer after searching some at Ancestry:

This is the passenger record for Caroline from the Ellis Island records:

I found this once I typed in Fraser instead of Frazer in my search. What makes me sure about this record is that Thomas John Frazer, her older brother, is listed on the passenger record above her. Ancestry has some more information:

Adding JD to the Frazer DNA/Genealogy Tree

Here I have Anne Caroline born 1868 which is probably more accurate than Jane’s ThruLines show.

Adding in Jennifer

I still have one person missing from Jane’s ThruLines. This is Jennifer:

She descends from Elizabeth or Eliza Jane Frazer. I won’t go through the genealogy, as I think the genealogy is OK.

That now accounts for DNA matching of the descendants of six of the children of Alexander Frazer. Turns out that Jane also matches Jennifer’s father Larry by DNA. Apparently many of the children of Alexander Frazer settled in Connecticut Including Eliza Jane Frazer.

Jane doesn’t match Brenda by DNA. So the match must have been to someone else. Brenda descends from the older brother of Anne Caroline who was Thomas John Frazer. These two travelled to the US together as mentioned above.

Summary and Conclusions

  • DNA matches and ThruLines help in checking on genealogical lines. Because there are matches, it means that the records must be there to support the genealogy.
  • I was able to find the interesting shipping record for Carrie (Anne Caroline Frazer) and her older brother Thomas Joyn Frazer.
  • Because Alexander Frazer had many children, that resulted in many descendants, many of whom took DNA tests.
  • ThruLines at Ancestry continue to be a helpful tool in validating (in this case) Frazer genealogical trees. DNA matches coupled with genealogy is a powerful tool and ThruLines is good representation of the intersection of DNA and genealogy.

 

 

 

 

Posted on

Re-Running My FTDNA Autoclusters

It looks like I have not run my autoclusters at FTDNA since 2018. As I recall, I didn’t get as much out of FTDNA as I did with Ancestry autoclustering. However, Ancestry autoclustering is not a valid option. I figured that autoclustering would be a good summary to show any important new matches that I may have missed.

First Try

I used the suggested parameters the first time, but the lower cutoff for a match at 50 cM was way too high. I just got three known Frazer relatives. This was less than helpful:

My assumption is that P means paternal side match as FTDNA makes that distinction based on testing or your own input. This showed that everyone matched everyone else except for Paul and Kenneth.

Second Autocluster with a Lower Cutoff of 15 cM

15 cM is a pretty good number because that is usually considered to be a high chance that the match is not by chance. Here is the 2nd run minus the names:

The Frazer Cluster – Red

The three Frazers who were in my first botched attempt are now in the red Cluster 3. There are an additional three people in that Cluster. Two I know and one I don’t know. One that I do know is sister to Susan who showed up in the first three person autocluster. That leaves Larry who shows a tree. Good news. His tree is a little sparse:

It’s sort of fun trying to build out these trees, but often frustrating finding the link. Here are some of Larry’s surnames:

Oldham and Hamilton were helpful names:

So I didn’t find an obvious connection. Too bad. I tried. Something may show up in the future.

Frazer/McMaster Connections Between Cluster 2 and 3

I think that the gray match between Robert and Paul is in the McMaster side. However, Benjamin and Robert don’t have trees at ancestry. My guess is that Robert descends from Edward Mcmaster and Celia Clarke:

Cluster 1 – Rathfelder

Cluster 1 is on my Rathfelder side based on a test by my cousin Catherine. However, I don’t know anything about Pamela and Romy. Romy appears to be from Germany. Pamela has a tree, but I don’t feel like going back a few hundred years of genealogy to figure out the connection:

 

Cluster 4: Maternal/Paternal Connection?

Cluster 4 is a bit mysteriious as it shows as a maternal cluster.

However, Craig has matches with Susan and Margaret. My assumption is that the connection is not through either of the two lines that I am related to these three people. I thinnk that Kenneth and Margaret are siblings.

Cluster 5: Big, Brown and Paternal

My assumption is that this could be my colonial Massachusetts side due to the number of matches. I also assume that the boxes with a P in it are the larger matches. Again, we see a maternal/paternal crossover. I would assume that this person (Brandy in the pink cluster) matches my mom but also has colonial Massachusetts ancestry.

Letterless Clusters 7 and 8

I’m on my own to figure out if these two clusters are maternal or paternal.

Cluster 7 in gray has a P connection with the brown cluster. Cluster 8 has connections with the pink and blue Maternal Clusters. Ned is the top match in Cluster 8. I had thought that my match with him went back to the Pilgrims on my paternal side, which seems to go against the side matches shown in gray.

Sarah in Cluster 9

Sarah is a familiar name in Cluster 9. She is my third cousin once removed:

However, that is almost impossible to tell from her FTDNA tree:

Her Margaret Shreak should be Margaret Shroek.

Clusters 10-13

Here there is no maternal or paternal designation. I’m on my own to figure these out:

I recognize Tracey as the first match in Cluster 12. She has a tree which I have built out in the past:

I have many matches which go back to the name of Lougheed in Ireland. I suppose this would be on my Spratt Line which I know the least about – or possibly Clarke.

Amy is also in that small Cluster 12:

Amy and Tracey are no doubt close relatives as they both share the ancestor of William Alexander Parr born 1939.

Summary and Conclusions

  • Autocluster is a good way to look at your matches as it puts them in groups with similar matches. It helps make sure you haven’t missed any important DNA matchs.
  • AncestryDNA autoclusters are more helpful, but they have not been available for a while. Perhaps FTDNA has tweaked their matching criteria which has also made FTDNA autoclustering more useful.
  • With thirteen clusters, one might expect these matches to be at around the third cousin level. Sarah was, but others are unidentifiable, so likely go out further.
  • The gray connections between clusters can be confusing, as they seem to indicate connections between maternal and paternal clusters in many cases.
  • When checking for connections, you have to look at locations. In my case, where I am missing information on my Clarke and Spratt ancestors from Sligo, finding common Sligo ancestry with my DNA matches can be a hint.

 

Posted on

Denis’ DNA Match On My Wife’s French Canadian Side

I was looking at my wife’s Aunt Lorraine’s DNA matches at Gedmatch recently. I found that she matched Denis:

I was also able to find Denis at Ancestry:

His tree was modest, but was something that I could work with.

Finding a Common Ancestor Between Denis and My Wife’s Family

The next step is to build my own tree for Denis. My assumption is that the match is on the Lefevre side as my wife’s aunt’s mother was a LeFevre. As soon as I enter Denis’ parents into my tree, I start getting hints.

I see that my wife’s second great grandfather was Edmond Lefevre born in Saint-Antoine-de-Tilly, so I feel I am on the right track.

Here is the baptismal record for Denis’ father Ernest:

Ernest’s father was Odilon. We further learn that Ernest’s mother’s name was Alma LeFleur.

Odilon LeFevre Born 1874

Next I would like to focus in on Odilon:

Here he is in 1881:

This appears to be three generations of LeFevre’s: Lazare, Joseph and Odilon. Here is Lazare from my wife’s grandmother’s tree:

Linking Joseph to Lazare

I next just need to show that Joseph was the son of Lazare. Here is a portion of Joseph’s marriage record:

This establishes that Joseph was the son of Lazare, “Notaire”. From what I understand, a notaire would be a sort of paralegal.

That means that we have a connection. I have a small LeFevre tree already, but it needs updating:

Let’s add Lazare and Denis:

According to this tree, Lorraine and Denis are third cousins. There may be other connections further back, but I will disregard those connections for the time being.

Mapping Denis’ DNA to My Wife’s Aunt Lorraine

The match between Denis and my wife’s Aunt Lorraine is either from Lazare LeFevre or Adelaide Bouré. We don’t know which without further DNA analysis. Here is what I have for Lorraine on her maternal side so far:

Lorraine’s DNA is 37% mapped out on her maternal side and 22% totally mapped (maternal and paternal). This would show as pink above. Denis bumped up Lorraine to 38% mapped maternally. Here is Denis’ contribution highlighted in gray:

He brings new information on Chromosomes 3, 5, 11, 12 and 16.

On Chromosome 11, it is truly new information. On Chromosome 12, it is already covered by RL, Sandra and Caroline.

Painting On My Father-in-law

I’m a bit behind on Richard as I don’t have a category for Lazare LeFevre and Adelaide Bouré. Richard has a more modest match with Denis:

When I paint this on to Richard’s DNA map, I see something interresting on Chromosome 5:

This shows that Richard’s DNA match with Denis is overlapping with Richard’s DNA match with Michelle. This is where I have a problem with my key. I had thought that Michelle’s common ancestor was Methot, but it is actually with Joseph Martin LeFevre born 1874. I need to differentiate my colors better. I changed Methot to yellow, so that it will be less confusing in the future;

Adjusting My Tree

Now that I know where Denis belongs in my tree, I can add him in:

That should help AncestryDNA figure out the ThruLines.

Speaking of ThruLines

Here are Lorraine’s ThruLines for Lazare LeFevre:

There is one match on the Joseph Line with Brad.

More Painting for My Wife’s Aunt Susy

I haven’t been paying as much attention to Aunt Susy’s DNA Map:

However, she has nice colors. Here is her match with Denis at Gedmatch:

I’ll need to add Lazare in to DNAPainter. Here is Aunt Susy’s maternal side:

DNA Painter gave me a boring color, but a chose a brighter color for Lazare and his wife. Susy was 33% painted maternally and 21% overall. Denis bumped these numbers up to 36% on the maternal side and 22% overall.

Denis and My Wife

Unfortunately, my wife and Denis do not share a lot of DNA.

Denis and Marie are third cousins once removed. That should translate to 4.5 generations to their common ancestors. That means that Denis and Marie share a little less than average for that relationship.

Painting this on to Marie’s map does not add any missing areas:

However, it does confirm the match with Caroline and others.

More on Caroline

I should have added Caroline to my LeFevre DNA tree. Caroline is at MyHeritage:

Charles Lefebvre shows as 1891. I checked my old blogs and see that the tree had “before 1891. Here is Lazare in 1851:

Charles is 9, so was born around 1842. The birthplace for all is Ville Quebec which I take to be Quebec City. I made a new tree with Caroline in it:

I’m sure that this tree could get quite big. This is what Marie has for ThruLines at AncestryDNA:

Outside of the Edmond Lefevre Line, Marie has 9 matches from 5 siblings of Edmond. And when I expand the Charles Line, I find Caroline in the photo:

Marie J G above is Marie Josephine Gratia who went by Gratia.

Summary and Conclusions

  • I started by looking at the DNA of Denis. I painted his DNA onto my wife’s map as well as my wife’s father’s and two aunts’ maps.
  • Marie’s father had a smaller match with Denis compared to the matches of his two sisters. Marie’s match was even smaller.
  • I made a DNA tree for LeFevre/Lefebvre
  • This tree included Caroline who I had mapped previously. She also appears on my wife’s ThruLines at Ancestry.
  • I was able to map Marie’s family’s DNA connection to Caroline because she also has her DNA at MyHeritage.
Posted on

A Newfoundland DNA Match with Barbara on My Wife’s Side

I was perusing my mother-in-law Joan’s Gedmatch results recently and noticed that she had a match with Barbara. This match is a little over 200 days old, so fairly recent. Here is Joan’s match with Barbara at Gedmatch:

Barbara also shows as a match to Joan at Ancestry and Ancestry suggest a common ancestor between the two:

My guess is that Joan and Barbara could share other common ancestors. However, this DNA match is important due to the lack of vital records for this part of Newfoundland. I already have a fairly extensive family trees of Upshalls who match by DNA:

The Newfoundland branch descends from Sarah Upshall. The George Upshall Line from Sarah is in the 4th and 5th columns. Here I add in Barbara without checking Ancestry’s genealogy:

Barbara is on the bottom left in green. I use green to indicate that detailed chromosome information is available for her.

DNA Painter

Because there is detailed information for Barbara’s match with Joan, I can use DNA Painter. This maps out Joan’s DNA. Here is some of the DNA that Barbara has added to Joan’s map:

Unfortunately, the program gave Peter Upshall a lighter color. It is better to have lighter colors for the more recent ancestors. However, I can adjust that in the program. At this point, Joan is 36% painted over all. That breaks down to:

  • 28% paternal
  • 44% maternal

Barbara matches on Joan’s maternal side. I’m pretty sure that Barbara added to the maternal and overall amount of DNA that Joan has identified on her map. Here is Joan’s Chromosome Map as shown by DNA Painter:

The top row for each chromosome is paternal and the bottom row is maternal.

More Painting: Aunt Esther

My mother-in-law’s Aunt Esther has this match with Barbara:

Right now Esther is 35% painted, but 53% painted on her paternal Upshall side:

I had already added Barbara to Esther’s DNA Map. Here she is on Chromosome 1:

Painter Joan’s Sister Elaine

Elaine has a more modest match with Barbara:

This adds a new set of ancestors for Elaine, going back to Peter Upshall born in 1800:

This raised Elaine’s maternally painted side from 30% to 31%.

Painting My Wife, Marie

I would have thought my wife would have had a bigger DNA match with Barbara based on her mother Joan’s match:

This match gives confimation that Karen is indeed matching on the Upshall side.

Summary and Conclusions

Barbara is another piece in the puzzle for confirming Upshall genealogy by DNA and was helpful in the DNA mapping of my wife’s family.

 

 

Posted on

A Fourth U106 Whitson BigY Test

A new BigY test is in for Pete’s brother Patrick. In my previous Blog on Whitson U106 YDNA, I was hoping that Patrick would answer some of the questions we had on previous testing in this area. Right now I don’t see any differences between the old Block tree and the new one. Here is the old one:

This shows Pete matching Tom and Norton under a block called R-FT137411. This block contains 23 SNPs. Here is the new Block Tree from the viewpoint of Patrick:

The only difference appears to be that there are now four in this group. I also see that the average number of Private Variants for R-BY62217 has gone down by one for some reason.

According to SNP Tracker, BY62217 appears to have stayed in Germany:

The Whitson Group headed over to England at some time during the Medieval Period:

Private Variants and Non-Matching Variants

The question that came up previously in my blogs on the U106 Whitsons was why were there so few Private Variants. The common ancestors between Pete (and his brother Patrick) and Norton is in the 1800’s. The common ancestor between Pete and Tom was in the 1700’s. Between the four of them, there was only one Private Variant with Norton. FTDNA normally pulls from the the Private Variants for new branches. However, new branches also require matching with someone else. With only one Private Variant, there could be no matching.

That is when I started looking at the Non-Matching Variants. My understanding is that these could be there due to testing in non-reliable regions of YDNA or incomplete test resutls. In my previous Blog, I came up with this chart for Non-Matching Variants between the three testers:

However, looking at this now brings up some questions. Norton had a private variant at 5014418. This is an unnamed variant. I think that Private Variants are generally unnamed locations. However, it seemed like Pete should have two named private Variants at BY44298 and BY55572. Further, if Patrick is positive for these two SNPs I would think that Pete and Patrick would be on a new branch for these two SNPs.

Patrick’s Results: BY44298 and BY55572

Here is what Patrick shows for BY55298:

This shows that the results for Patrick were not conclusive:

Patrick had only one read and it didn’t show a mutation. That was unfortunate, as it is likely that Patrick, as Pete’s brother, should have the same mutation for BY44298.

Next, I look at BY55572:

Again, there is a similar situation:

These two SNPs are in close proximity to each other on the Y Chromosome. However, I don’t know which regiosn are troublesome vs. easy to read. Here is Pat on my spreadsheet:

I think that if the test was better, then it should have shown Pat positive for these two SNPs.

In additions, Pat shows very little in the way of non-matching variants:

Patrick’s only Non-Matching Variant was Norton’s Private Variant.

Patrick’s Results for the Previous Non-Matching Variants

Next, I can fill in the rest of my Excel Spreadsheet for Patrick:

Where NT means not tested. That means that these are disapppointing results. Pete and Norton got results for all the SNPs on the list Tom had results for two SNPs.. Patrick had results for none of the SNPs. Most of the ? results only had one read. I was hoping that Patrick would have matches to many of Pete’s non-matching variants. Logic would dictate that Pete and Patrick should have most if not all the same variants as they are brothers.

Summary and Conclusions

  • Previous testing has shown that Pete was positive for 10 SNPs that Norton was negative for
  • Subsequent testing showed that Tom was negative for two of those SNPs and had inconclusive test results for the other eight. However, there has to be two people positive for a SNP for a new branch to form
  • When Pete’s brother Patrick tested, I had thought that he would have matching SNPs with Pete which would form a new YDNA Branch. However, Patrick’s results were inconclusive for 8 of his SNPs that Pete was positive for and for two, that region of YDNA was not covered.
  • With the information I have seen, the Whitson Block tree may remain the same. However, it is possible that the manual review has not been completed and FTDNA may see more detailed information from the test results which could shed some more light and give a new branch for Pete and Patrick. If the manual review is done, I would recommend asking why Tom and Patrick’s results did not cover the important Non-Matching Variants between Pete and Norton. I don’t know if Pete and Norton just had really good results or if Tom and Patrick had unusually bad test results.
Posted on

Reinhold’s DNA Match and the Hirschenhof Connection

Reinhold matches my Mom by DNA at AncestryDNA. The match is not huge at 38 cM, but Ancestry shows common ancestors. Here is the first common ancestor tree that Ancestry shows:

I see one problem in the tree in that Reinhold’s grandmother shows as being born in 1970. I’m not sure why. Reinhold’s tree has Veronika born in 1903. Here are other likely common ancestors:

Here are those same names highlighted on my mother’s tree:

The closer three connections are on my mother’s paternal grandfather’s side (Rathfelder). The further connection is on my mother’s paternal grandmother’s side (Gangnus).

Let’s see if I can find all these people on Reinhold’s tree. He is two generations removed from my mother for six of the common ancestors and one generation removed for Niclas/Riedel.

The Rathfelder/Bittenbinder Connection

I have Biedenbender rather than Bittenbinder but I am sure they are the same. Here is Reinhold’s tree:

This couple are the 5th great-grandparents of Reinhold. I could check on Reinhold’s tree, but I have no reason to doubt it. According to Reinhold’s tree, his paternal grandmother Veronika Simon was born in Riga in 1903. My grandfather Alexander Rathfelder was born in Riga, Latvia 10 years earlier in 1893. This was a tumutuous time and place to be born. Veronika’s mother was born in Hirschenhof, Latvia in 1880. This was a German Colony in the heart of Latvia. These Germans tended to intermarry with each other within that Colony.

I name both Rathfelder and Biedender as the DNA shared between Reinhold and my family only came from one or the other of the two. It would be difficult to find out from which side the DNA came from or even if the DNA shared is from either as there are so many other possibilities.

Mertz/Muth Connection

Again, from Reinhold’s tree:

Reinhold’s fourth great-grandparents were Gustav Mertz and Anna Maria Rathfelder, but we are connected throught the parents of these two or at Reinhold’s fifth great-grandparent level.

Gangnus/Muller Connection

I’m glad that Reinhold built such a good tree. This is a bit of a who’s who in my mom’s Hirchenhof genealogy. This is on Reinhold’s Spengel line. Interestingly, my mom’s grandmother was a Gangnus, but this Gangnus is on my mom’s paternal grandfather side.

What’s more, if I have my genealogy right, my mother had two Gangnus 3rd great-grandparents named Gangnus. I have that Philipp Julius Gustav Gangnus and Philipp Jacob Gangnus were brothers.  Philip Jacob was on my mother’s Gangnus side.

One Generation Back to Niclas/Riedel

It is less likely that the DNA match is on either of these two people, but it is still possilble:

On Reinhold’s tree, this couple is on his Simon side. In my mother’s tree, this match is on her Gangnus grandmother’s side.

Adding Reinhold to My Rathfelder DNA Tree

This is a tree that combines DNA matches with genealogical matches. This is what I have now:

Reinhold is a 4th cousin twice removed, so he will be two levels below Astrid and Ingrid:

Reinhold’s name did not get centered for some reason.

A Merz/Muth DNA Family Tree

This is a new tree for me:

Turns out I can copy and paste from the previous tree, but the formatting is a bit odd. Reinhold’s tree has a birth date for Johann Peter Mertz at the top, so I borrowed that from there. As far as I know, Reinhold is the first one who I have found that shares these two ancestors of Merz and Muth.

Gangnus/Muller Tree

Unlike the Merz/Muth Tree, I already have an extensive Gangnus tree:

Rienhold should come in on the left side with Philipp Julius Gustavis Gangnus.  He descends from Sophia Gangnus born in 1783:

This gets a bit complicated the way I have it. I don’t think that Ancestry picked up the fact (if I have it right) that Reinhold is a 4th cousin twice removed and 5th cousin twice removed on the Gangnus Line. This tree is quite robust. There are many branches which in my view is desirable.

Niclas/Riedel Tree

This tree is also a first:

This tree also has the good type of branching in three different generations that I like to see.

Summary and Conclusions

  • Reinhold matches my family by DNA and by many common ancestors.
  • From what I can tell, Reinhold shares at least 10 common ancestors with me and my family.
  • I added Reinhold to my Rathfelder/Bittenbinder and Gangnus/Clausing Tree
  • I created a new Merz/Muth and Niclas/Riedel DNA Tree to show the relationship between Reinhold and my family.
  • It is no surprise that Reinhold has DNA matches with my family considering all the common ancestors we share.

 

Posted on

A Frazer STR Tree

A new 111 STR match to the Frazer YDNA group has me thinking about Frazer STR trees.  I haven’t seen the specific STR results for the new Frazer match, but we have five testers who have already taken the BigY test. That means that we already have the skeleton of a STR tree which is the SNP tree. Here is the current Block Tree from the perspective of my cousin Paul:

This goes up to the level of R-YP6489. I can portray that like this:

These SNPs tend to set this tree in stone. We know the genealogy of the first four testers. Y151390 is from the James Line. James Frazer was born in the early 1700’s. Y5682 is from James’ older brother Archibald. We are not sure how Frazier fits in by genealogy other than the fact that the connection is further back in time. – probably in the 1400’s in Scotland.

Here is that same group in the Fraser and Septs YDNA Project:

Here I wrote down the Mode which is the third row above. This is generally thought to be the ancestral value:

However, I questioned DYS710. The values were:

Frazier had a 32. The first row is supposed to be the minimum and shows as 33. Hmm. It would make more sense ifthe ancestral was 33. Then Rick and Paul on the Archibald Line would have mutated up and Frazier would have mutated down.

For a similar reason, I would like to use 35-39 for CDY even though it is the value appearing least often:

If I have written down all the deviations from the modified mode correctly, I have this:

Green is the Archibald Line of Frazers. Blue is the James Line and reddish is the older Frazier/Frazer Line.

When I put those STRs onto the SNP or Frazer Haplotree, I get this:

I didn’t have room for all the Frazier/Frasher STR changes over the last 500 years, so I left those out. I do note that Frazier and Rick had what is called a parallel STR mutation at DYS444. That means that Rick and Frazier don’t share the same branch by that mutation but that they both had those mutations independently of each other.

Why Is a STR Tree Important

This tree is mostly important for Frazer relatives who have not taken the BigY test. This should show how they are related to various Frazers. Once I get the new 111 STR results for the latest Frazer relative named Dingman, I will be able to add him to the STR tree. He likely to be aligned on Rick and Paul’s side, but the common ancestor is believed to be here:

Because the connection is so high in the tree and because Jonathan has no known mutations since about 1690, that would explain why Dingman shows a closer STR match to Jonathan:

In the tree above, there is some uncertainty as to which son of Archibald Frazer and Mary Lilley Dingman descends from.

A New Frazer STR Tree

Here is the new Frazer STR Tree with the new Dingman STR results which just came in:

One major contribution of DIngman is with DYS710. He has a value of 34 along with Rick and Paul where the other three testers have 33. That means that a DYS710 of 34 represents the Archibald Line of Frazers (as opposed to the James Line). That means that either Archibald Frazer born about 1715 or his son Archibald born about 1743 had this mutation.

Checking the Numbers

Above, I mention that there should be 5 STR value differences between Jonathan and Dingman, but there appear to be 6. For the shared ancestor of Jonathan and Rodney (Thomas Henry Frazer born 1836), I had an extra STR. If I take that out, it adds up:

I left my mistake in above to show how easy it is to make mistakes in these trees. I also had the wrong STR name for Paul which I corrrected above. Hopefully there are no more mistakes. Now it adds up as there is one mutation between Jonathan and Archibald Frazer of 1690 and four between Dingman and Archibald Frazer of 1690. That adds up to a geneatic distance of 5.

Next there should be a GD of 7 between Jonathan and Frazier. That means I missed a Frazier STR.

DYS710

It turns out I missed DYS710 for Frazier, the STR I made a big deal of above. It turns out that DYS710 is an important defining STR for Frazer Genealogy:

  • DYS710 = 33 – Ancestral to Frazers/Fraziers (probably) and defines the James Line today
  • DYS710 = 34 – The Archibald Frazer Line (as opposed to the James Line)
  • DYS710 = 32 – Defines at least part of the Frazier Line. We don’t know how far back. Unfortunately, this marker is the first STR in the 111 STR test, so it will only tell us about those who have taken the 111 STR test.

DYS710 is a good defining STR because it is slow moving. CDY is in the differrent lines, but it is a fast mover, so not the best STR to define a Line.

This image shows DYS710 in lighter blue as it is the first SNP in the 111 panel (after the 67 Panel of STRs). I have an arrow pointing to CDY. The faster moving STRs are shown in a maroon color.

The tree also shows a consistant amount of STR differences going down the tree from top to bottowm between Dingman, Rick and Paul. They each have four STR changes in the tree. The older Frazier line has 6 changes. Jonathan and Rodney only have one and two STR mutations respectively. My guess is that the James Line has fewer than average STR mutations and that the Archibald Line has more than expected.

Where Should Dingman Fit In As Far As SNPs?

Assuming the chart above is correct, Dingman will be positive for FT421618 as he descends from Archibald Frazer born 1690. He will be negative for Y151390 which is in the James Frazer Line.  We don’t know if  Dingman will be positive for Y58652 even though he is from the Archibald Frazer Line:

Y85652 is representative of a block of three SNPs.  These three SNPs are also representative of James Frazer born about 1804. Dingman descends from Archibald Frazer born around 1743. That means that any SNPs that Dingman would not have any SNPs came about between that Archibald and that James. My guess is that he could be positive for all three or more likely he will be positive for one or two of the SNPs and negative for the other one or two.

Another interesting thing is that the Archibald Branch of Frazers seems to have produced 3 SNPs in 4 generations.  That represents the period between about 1715 and 1804 or about 89 years. That is quite a bit considering that some people use an averege number of one SNP every 84 years. However, having these extra SNP is important for future fine tuning of which SNP goes with which Family under the Archibald Branch of the Frazers of Roscommon Ireland.

Where Did I Go Wrong in My Last Blog?

In my previous Blog, I had predicted that the common ancestor between Dingman  and the Frazers would precede Archibald Frazer by some time. This prediction was wrong. It is difficult to make these predictions based on STRs. Now that I know Dingman’s genealogy, it is easier to place him in the SNP tree and it makes the STR tree fairly easy to build.

Summary and Conclusions

  • There is some guesswork building a STR tree
  • In this case where I now understand the genealogy better and now the SNP tree for five testers, that takes away some of the uncertainty and I was able to build a reasonable STR tree
  • Dingman adds an important piece to the YDNA Frazer tree
  • If Dingman or another Archibald Frazer descendant were to take the BigY test, they would likely further fine tune the YDNA tree for the Archibald Branch of the Frazer tree.

 

Posted on

A New Dingman YDNA Match to the Frazer Group

I have been involved with Frazer DNA and specifically Frazer YDNA for quite a while. The best YDNA test at FTDNA is the BigY. The result of that test is to place you in a SNP or SNP group called a Haplogroup. The Frazers from North Roscommon are fortunate to have four BigY testers. This has not only given the Frazers their own Haplogroup, but has also split the two Frazer Lines into two Haplogroups. The best view of this is at FTDNA’s Block Tree View. Here is the view from my cousin Paul’s perspective:

I have added some notes. The first column is the James Line. The second column is the Archibald Line. We think that  the father of James and Archibald Frazer was born sometime around 1690. The SNP R-Y151390 represents the James Line. The SNP R-Y85652 represents the elder brother Archibald Frazer. The white space above these two SNPs should represent Archibald Frazer born around 1690.

Stuart and Grant YDNA

Due to the age of matches the Frazers have with Stuart and Grant, my assumption is that these early relatives were living in the same area of Scotland when people picked their surnames. Some picked Frazer, some Stuart and some Grant or others in the same area of Scotland.

Frazier YDNA

A Frazier YDNA tester came along and split the block of SNPs in half so he is now R-YP6489 where the Frazers used to be. If we take one SNP to be 84 years, then his common ancestor with our Frazers would be 252 years before Archibald born around 1690 or around 1440. I am assuming that it is more than coincidence that this match has a name very similar to Frazer. From that we may assume that the five of us descend from an unbroken line of Frazer/Fraziers going back to the middle ages.

Here is a Clan map showing how some of these Clans lived in fairly close proximity to each other:

This map includes Fraser, Grant, Stewart and Chisolm. Chilsolm has been a STR match.

The New Dingman 111 STR Match

This is how Dingman matches my second cousin Paul:

First note that STR matching is not as accurate as the SNP matching. In the SNP block tree, Frazier was more closely related than the two Stuarts. Based on this matching, one might think that it was the other way around.  The closer the match is, the more accurate the STR matching is. This shows that Rick is Paul’s closest match:

Paul and Rick are 3rd cousins once removed. Paul is a 6th cousin once removed to Rodney and Jonathan.

Where Does Dingman Fit In?

The match between Paul and Dingman does not appear to be an anomaly. I have heard from relatives of Jonathan and Rick and they both match Dingman at about the same level as Paul. FTDNA provides as guestimate as to how far back Paul’s common ancestor is with Dingman:

FTDNA gives a statistical answer in generations and percent likelihood. We see from my tree above that Paul, Rodney and Jonathan share a common ancestor back at 7 or 8 generations. I would not be surprised if Dingman shared a  common ancestor with Paul in this area:

If Dingman were to take the BigY test, we may find out that he and our Frazers have a common ancestor around 10 generations ago or around 100 years before Archibald Frazer born in 1690. This should put the common ancestor in Scotland around 1590. My other guess is that Dingman could have been a Frazer originally. A Dingman BigY test would likely split the FT421618 Block based on how he matches other Frazers with STRs.

Here is some more documentation of TIP Reports from FTDNA:

What We Don’t Know

The YDNA match to Dingman indicates that he has not provided information about his most distant male ancestor. He has not posted a family tree. In addition, I don’t think that Dingman has joined the R1a YDNA Project at FTDNA nor has he joined the Frazers and Septs Project. As he has not appeared to join these projects, we cannot tell his specific STR differences with other Frazer YDNA testers other than knowing his Genetic Distance.

Summary and Conclusions

  • Dingman is a good YDNA match to our Frazer relatives. Our Frazers have the ancestor of Archibald Frazer born around 1690 and who probably lived in North Roscommon County, Ireland.
  • Based on the 111 STR test Dingman’s common ancestor with our Frazer Line could go back to a Frazer born around 1590 in Scotland.
  • If Dingman were to take the BigY test, it would probably result in more refinement of the Frazer YDNA tree.
  • Right now we don’t have information on Dingman genealogy or the specifics of his 111 STR test other than generational distance.

 

Posted on

Whitson R-U106 BigY Non-Matching Variants and STRs

The title to this Blog is a bit specialized. There are three Whitson relatives who have taken the BigY test. This group of Whitsons are on the R-U106 general branch of humanity. From there, the Branch has narrowed down to a block of SNPs represented by the name R-FT137411.

In my previous Blog, I started looking at the three Whitson Testers’ Non-Matching Variants. So far:

  • Tom – He didn’t show any private variants. However, his date to common ancestors with Pete and Norton implies that he should have some.
  • Pete – He didn’t show any private variants, but two of his variants which don’t match with Tom and Norton could be Private Variants. The testing for Tom and Norton seemed to show that they didn’t have BY44298 and BY55572. I suggested that these two SNPs may be a new Branch for Pete once Pete’s brother’s BigY test is completed.
  • Norton – He has one Private Variant which he is clearly positive for and Tom and Pete are negative for. This Private Variant is at position 5014418.

Pete and Norton’s Non-Matching Variants

Here is the view from Norton’s perspective:

The non-matching variant between Norton and Tom represents Noton’s Private Variant. The first two named Non-Matching Variants between Norton and Pete were addressed in my previous Blog. That leaves six non-matching variants to look at between Peter and Norton.

BY151409

First, I’ll check Pete’s results:

Pete is showing positive for that SNP. That must mean that Norton is negative for this SNP:

This shows Norton as not derived for BY151409. But it also shows that this SNP is on the Y-Tree. Here are Norton’s test results:

If this SNP is on the Y-Tree, it must not be a private variant. I think that this information is either not helpful or not acurate. I don’t think that BY151409 is on the Y-Tree. Here is what YBrowse shows:

To be complete, I’ll check Tom’s results for this SNP:

FTDNA considers Tom a no-call for this SNP:

Tom only had two good reads:

A third not-so-good read had the mutation as T, when Pete’s mutation was to G, so something is off there. That could mean that this variant could be a Private Variant for Pete also.

BY151410

OK, only 7 variants to go. I will make a summary chart to keep track:

Finishing the Whitson Non-Matching Spreadsheet

It seems that Pete should have as many as 10 Private Variants which seems like a lot. That would be 11 Private Variants divided by three people or about an average of 4. However, it seems like once a variant is named it is not included in the Private Variant category. However, my understanding is that the named Private Variants are not treated the same way as unnamed Private Variants. The SNPs for Pete appear to have been named during Pete’s earlier testing. However, as far as I know, there are no other matches to these SNPs – unless they are in other Haplogroups.

I found this in a R-U106 Forum:

* There is a long, complex, and (mostly) documented description about how the data is extracted from the DNA and ends up in the BAM files, which are then parsed into the VCF+BED files we’ve been requesting for the Warehouse. I won’t go into that here. 
* This data then undergoes some extra processing. Those that match certain quality criteria are extracted and presented as a list of high-quality variants. These high-quality variants are then compared against everyone else in the tree, who is given a positive, negative or null call.
* SNPs where you are positive and others are negative (or vice versa) are used to define your non-matching variants. These are what are listed in your BigY matches.
* Your list of high-quality variants is then subject to a blacklist, removing any SNPs in problematic regions (DYZ19, the centromere and Yq12) and recurrent SNPs found elsewhere in the haplotree. These are listed as your private variants. The blacklist removes regions where complex mutations can lead to mis-mapped regions, missing data or other problems. These complex mutations can then masquerade as one or more SNPs.
Consequently, there may be many reasons why SNPs in your non-matching variants may not appear in either your private variants or your match’s private variants, and vice versa.
So I suppose my question is whether Pete’s non-matching variants include blacklisted SNPs. My sense is that they probably do.

R-U106 Whitson STRs

As suggested by Pete, the private SNPs for the three Whitson testers, do not appear to adequately describe the time to common ancestors. Pete has a common ancestry with Norton at 1835 and with Tom at 1725. Based on that, there should be roughly an average of one private SNP between Norton and Pete and two between Pete and Tom.

It would seem at this point that the STRs show some accuracy in this Whitson Line. Here is the relevant portion of the U106 Whitson STRs:

The third row is the mode of the three testers. The fifth row shows Pete’s results. His results are the same as the mode. In drawing STR trees, the Mode is considered the ancestral value. So that would mean that Pete has had no STR mutations in the 111 series since 1725. I get a tree something like this:

This shows that there are been one STR mutation between Pete and Norton since 1835 and that there have been two STR muations between Pete and Tom since 1725. I suppose that I should have put the 1725 in the red box above.

As this is a combined SNP and STR tree, I can add Norton’s Private Variant:

The confusing part are Pete’s 10 non-matching SNPs. Norton tested negative for these  and Tom had all no-calls. I will wait to see Pete’s brother’s BigY results to see if any of those SNPs come into play.

Summary and Conclusions

  • I think that I’ve taken the U106 Whitson YDNA about as far as it can go with the information I have.
  • I found it somewhat difficult to understand the nuances of non-matching variants and what they are useful for.
  • Previously, I had suggested that two Pete’s non-matching variants may become a new branch. After looking more closely at Pete’s non-matching variants, I am not so sure. I will like to see how Pete’s brother’s BigY test shakes out.
  • I summarized the testing quality for Pete’s non-matching variants in the other two Whitson BigY testers
  • I drew a combined SNP/STR tree for the U106 Whitson branch. Given the time periods Pete gave me to his common ancestors, these STRs seem to fit in better than what we have for Private SNPs right now.
  • I am looking forward to the BigY testing for Peter’s brother.

 

 

Posted on

Daley DNA and Genealogy

I am looking for specific Daley DNA and genealogy on my wife’s side. Here is my mother-in-law’s maternal side:

Elizabeth Daley was born in Canso, Nova Scotia and died in Boston during the Flu Epidemic in 1919. DNA matches have been scarce, but we will take a look.

Ancestry Daley ThruLines

An easy place to look for genealogical connections with DNA matches is with Ancestry’s ThruLines. Here is one of the Daley ThruLines as shown from the perspective of my mother-in-law Joan:

This shows that Joan has a DNA match with Katherine who likely descends from Elizabeth Daley’s sister Mary. Here is what Katherine has about her grandmother:

It looks like Mary stayed in Canso. This appears to be Mary Kelly’s mother Mary in 1911:

Here is the family in 1921:

This was not easy to figure out. Ancestry transcribes her husband’s name as Canet. However, I see something more like Everett. Julia would be Katherine’s grandmother Mary Julia. All this to say is that the ThruLines for Katherine and Joan look right.

More ThruLines Back to Lawrence Daly

This shows Edward (who my mother-in-law Joan descends from) and two siblings: Michael and Ann. I must not have Ann in my tree as she shows as green. However, the fact that there are three DNA matches to Joan tell me that Ann is the real thing.

A Daley DNA Tree

Here is the start of my Daley DNA tree:

Here is Canso, where many Daley’s lived:

Canso is in Nova Scotia below Cape Breton and not too far by water from Prince Edward Island and Newfoundland where my mother-in-law Joan’s other ancestors were from.

The Ann Daily Line

The DNA matches are to Joan. Carol has a reference to Ann Daily’s marriage which makes all this seem legit.

Here is Black Point – not far from Canso:

Here is an enlarged Daly/Daley Tree:

 

Elaine’s Daly ThruLines

Joan’s sister Elaine has some DNA matches also.

Along with the Edward Daley Line, Elaine has 4 matches in the Michael Daly Line and 4 matches on the Ann Daly LIne.

Elaine’s Michael Daly Line Matches

Elaine shows four matches. However, the ThruLines connection with Lois appears to have an error. It has Michael Daly in the tree twice for some reason. Here they are added to my Daly genealogy/DNA tree:

Elaine and the Ann Daly Line

Elaine’s match with Tara adds an additonal branch:

I had the Ann Daily Line wrong in my tree above. Here I have corrected it:

Here is the whole tree:

The could have been some marriage of cousins in this scenario as some of the surnames in these lines look familiar. I have not been able to find Daley matches where the match has uploaded to Gedmatch or tested at MyHeritage or FTDNA. Because of that, I don’t have detailed information on these Daley DNA matches.

Summary and Conclusions

  • Between Joan and Elaine, they have several DNA matches with the descendants of Lawrence Daley and Elizabeth Snyder
  • It seems like all these matches are at AncestryDNA
  • AncestryDNA does not supply specific DNA information on how those Daley/Snyder match
  • Because this specific DNA information is missing, I was not able to map the Daley/Snyder DNA matches onto Joan and Elaine’s chromosomes