A more accurate YDNA haplogroup from 23andme tests

While trying to find information about my family's Y DNA haplogroup, I ran across a post on a message board stating that the SNP testing at 23andme was accurate but that they used the 2010 tree. Hmm, I remembered reading a great blog about extra stuff you could do, and I went back to it. I thought I would blog about it so that other 23andme user's would know how to get the latest  haplogroup based on the 2015 tree. To be honest, I do not know if there are snp's not tested by 23andme that determine haplogroups.

1. Download your raw data from 23andme.
This is found under the Browse raw data option on the upper right hand drop down menu. Once the data is downloaded, you need to extract the zipped file.

2. Download the 23andme to y snp program.
This wonderful little program is found here.

3. Open the 23andme to y snp program and then open your raw data file from within the program. It will analyze the data. Save this new information to a txt file from within the program. I usually save it in the same file as the raw data that is unzipped so I can find it again.

4. Download the chrome extension, ISOGG Y tree add on, a link to it is found here.

5. Open the ysnp file you just saved. Select edit, select all and copy.

6. On Chrome, go to the three lines in upper right corner, and click on it. Select settings in the drop down menu. Then select extensions.Find the ISOGG Y tree add on in the list. Under it there will be a link for options. When you select this you will get a new window. A sample is already in slot 1. You can add up to 9 more people's y snp data. Put the name of the person and paste the ysnp file data you copied from the text file. Close the window.

7. Go to the ISOGG Y tree. You can find it here. You will see letters representing the first letter of the haplogroup. There are also links for the Y tree from previous years. The add on will work with all of them.
Click the letter that corresponds with the haplogroup you were assigned by 23andme.

8. Once you get to the haplogroup page, the add on will load. In the upper left hand corner when it is done you will find a drop down screen. Select the name you want to run the program on. When it stops saying loading, scroll down the page. The SNP's will be highlighted RED or Green. Green means you have the SNP, Red means you don't. Simply put, the last green highlighted snp on the page is the one that corresponds to the haplogroup to it's left, and that's the haplogroup your dna is at this time.

As a result of this, my mom's paternal line went from R1b1b2a1a1d at 23andme to R1b1a2a1a1c2 in the 2015 table.

Hope this helps.

Discovering a lineage, with help from the X chromosome

As I have mentioned before, Mom's half first cousin recently tested. While I was waiting for his results from 23andme to show matches, I uploaded his raw data to Gedmatch. A few days later I was surprised to see his third highest match, 138 cM autosomal and 54.1 cM on the X chromosome with an estimated difference in generations on gedmatch of 3.4.

I emailed the match and requested a tree. Before receiving it though he had already confirmed he was of Hungarian lineage, which I also knew was the nationality of both of SH's grandparents. Unfortunately though, all SH knew was his grandfather was John Pado and his grandmother was Agnes Odor. From ancestry.com I knew what year Agnes arrived, and that on her death certificate her father's name was listed as John Odor.
Her immigration arrival gave also where she was from in Hungary, Gyovar.

I already knew that any connection we found in the tree's would have to come not only from SH's maternal side, but from either a connection to the mother of  John Pado, the mother of Agnes Odor, or the mother of John Pado, because an X chromosome match ruled out anyone who was not an X donator. I also verified that the name Janos was the Hungarian equivalent for John. Since I was unable to find, and still cannot find, the immigration record for certain of John (Janos) Joseph Pado, I looked at what I did know, the village of Gyovar.

When I received the tree of SH's match, I looked it over. Bingo, he had  in his tree the Kaposi family who intermarried with the Horvath family in his tree as his Great Grandparents and they were from Gyovar. Since the Great Grandparents were the right generation for the estimated match, I then went to ancestry.com and searched for Pados in Hungarian Baptism Records. I found Janos Pados whose wife was Rozalia Kaposi in several in the village of Gyovar.

My next step was to see if Anna Kaposi and Rozalia Kaposi were sisters, and indeed I found that Vendelin Kaposi and Julianna Holtain had baptism records for Anna Kaposi and Rozalia Kaposi in Gyovar, and that they were of the right age to match the baptisms of each girls children. Thus SH's great grandmother Rozalia Kaposi was the sister of his match's great grandmother, making them 3rd cousins, and giving SH and his sister a little bit more of their immigrant ancestor's history.

Ancient DNA and me

Over on Gedmatch you can compare your autosomal DNA raw data with ancient DNA. More about the types of ancient DNA and some other nifty tools you can use can be found here.

So I tested all five of my profiles at gedmatch. Not one snp matched, even on the lowest setting on comparison directly. You get a graph on the other function there, but all it looks like to me is a whole bunch of orange lines. So I decided to use the ancient calculator that Feliz Immanuel wrote and test our DNA that way.

Here are the results. I arranged them sort of higher to lower matching. The number is the percent of DNA that matches.

Ancient DNA	ME	JM	SH	DAD	MOM
Ne 1	20.09	21.46	21.04	20.16	21.34
usht ishm	16.64	17.04	17.09	16.99	17.93
loschbour	16.92	16.32	17.36	17.5	15.34
linearbandkeramik	20.35	20.04	20.07	21.16	19.69
clovis	12.59	13.52	11.3	12	14.03
br2	22.65	21.62	20.94	21.73	23.29
kostenki14	5.7	5.35	4.83	5.12	5.34
motala 12	6.12	5.66	5.26	6.16	5.31
australian aboriginal	1.01	1.1	2.24	0.85	1.29
br1	1.44	1.8	2.05	1.75	2.25
co1	2.6	2.42	3.17	3.31	3.07
hinxton4	1.03	1.23	1.6	1.28	1.11
IR 1	3.34	2.98	2.79	4.4	3.37
KO1	3.24	3.27	4	3.44	3.18
LaBranaarintero	3.9	3.26	4.75	3.57	3.41
Malta	2.47	2.03	2.16	2.23	1.89
NE 5	1.89	2.05	2.6	2.38	2.13
NE 6	2.48	2.32	2.99	3.48	2.67
NE 7	2.38	2.64	2.71	2.39	2.54
Paleoeskimo	2.58	1.85	1.62	2.85	2.22
gokhem2	0.57	0.43	0.76	1.46	0.88
hinxton 1	0.12	0.09	0.3	0.28	0.12
hinxton 2	0.19	0.31	0.27	0.22	0.71
hinxton 3	0.36	0.24	0.24	0.34	0.64
hinxton 5	0.19	0.14	0.48	0.28	0.47
vi33.16 neanderthal	0	0.12	0.1	0	0.04
vi33.25 neanderthal	0.05	0	0.11	0	0.05
vi33.26 neanderthal	0	0.04	0.19	0	0.15

I would love to know if there are any averages for the above tests, and also if the above samples test higher with one than another in comparison. 

Disappointment with Ancestry Admixture or Composition Results

Since many members of my groups on facebook research possible Native American ancestry, it is not unusual for me to see posts about how disappointed they are that there is not any, or very minuscule, amounts of Native American showing up on their admixture or composition results.

To be honest, there are times I honestly think that they got the results they were supposed to, however, there is a very good reason that someone does not have the results they think they should. It is called recombination and it is how our genes are passed down.

To illustrate this fact, I am going to use examples from my mom, her cousin SH and her nephew JM with genealogically proven third cousins once removed. Genetically speaking a third cousin has the estimated DNA of 0.781 percent or 53 cM,  a third cousin once removed has  the estimated amount of DNA around 0.391 percent or 26.56 cM shared, and a fourth cousin (for JM) should share about 0.195 percent or 13.28 cM of DNA.

Mom has matches with two sisters and one daughter on her Hager side. The sisters are 3rd cousins  to my mother and SH and the daughter is a 3rd cousin once removed. (for JM that means third cousin once removed and fourth cousin).There is another Hager descendant also of the same degree they match as well.

Mom matches one sister at 145 cM, the other at 130 cM, both third cousins, the daughter at 65 cM and the last at 9.9 cM both third cousins once removed.
SH matches one sister at 40 cM, the other at 74 cM, the daughter at 39 cM and the last at 5.2 cM
JM matches the one sister at 29 cM, the other at 0 cM, both third cousins once removed, andthe daughter at 22 cM and the last at 7.7 cM, both fourth cousins.

Another example is a genealogically documented third cousin once removed on the Collins family.
Mom matches him at 38 cM, SH at 134 cM and JM not at all.

So let's say, hypothetically speaking, that the native DNA was represented by one of these ancestral common ancestors DNA. As you can tell by the numbers, the amount passed on varies significantly, even though the genealogical distance is the same for SH and my Mom. In the case of JM, when he has no shared DNA it means he didn't receive any of that genetic material. If this portion of the DNA represented the person who was passing on the Native DNA it would mean yep, he's not going to show a percentage even close to that of his biological Aunt.

It's obvious that there is some kind of sticky DNA going on in some of my mom's family, she is sharing 1.7-1.9 percent of DNA with her Hager cousins, but only 0.5 percent with her Collins one, and on the flip side, her first cousin is sharing .0.5-0.9 percent with his Hager cousins and 1.8 percent with his Collins cousin.In both cases, the lower percentage is more the average, so the larger matches while possible, are not typical.

Ancestry composition is passed the same way as this DNA, it is a recombination draw of the lottery in what gets passed on to each generation. So how is it my Mom actually shows a percentage of Native DNA close to her actual degree of blood based on genealogy and the Dawes? Luck for the most part, when you stop and think that her last full blood ancestors on the Trahern only side were born in the 1770's, with mixed bloods intermarrying until her great great grandfather married a full blood. Even at that, Mom's great grandmother could only be 11/16 Native American (her father was 3/8 and her mother full).

Based on the Dawes where she claimed only 1/2, mom's Choctaw DNA should be 1/16 or 6.25 percent, if you include the reported actual, it would be11/128 or 8.6 percent . Her 23 and me results show her at 8 percent, and gedmatch admixtures average 8-10 percent based on the program.

Even though her great grandfather and her grandfather are by family history Native American, their contribution would be so minuscule that it's not even worth computing.

So if your disappointed with your admixture results, you really need to look at some of your close family and test them. First cousins, and even siblings do not always share an equal amount of DNA from all their contributors.

Blood is thicker than water, part 2

So, if testing my mother's half cousin, and her paternal relatives will assist me in finding her matches and answering questions on her tree, is there any benefit to testing on my father's?

Yes. Already 11 McCurdy cousins have tested. As you will see below, this has allowed me to determine 61.5 percent of my father's McCurdy DNA. So who would be beneficial to test?

 For everyone, we are in great hopes that a great granddaughter of Elijah McCurdy will test. If so, that may help us identify other McCurdy matches outside our own family other than the one we have found.

Dad's grandmother had a half brother whose daughter is still living. Not only is this a half first cousin once removed, her grandmother was my father's grandfather's cousin, meaning she's doubly related to Dad, and on the two least represented branches of his maternal tree, the Pyburn and the Hardy family.

We have several Hardy cousins to possibly test if they agree. They are anywhere from 3-4th cousins, and hopefully we can find a willing male who carries the Y DNA haplogroup.

The Brunson  and Franklin side has been seriously lacking in confirmed matches, despite the large percentage of mutually shared DNA Dad has that's mentioned below. Unfortunately, I don't know anyone close enough on those branches to even approach about testing.

Below is a summary of the results of research with Dad's DNA at this point and time.

Paternal DNA 3718 cM of which a total of 478.4 cM or 12.9 percent has been identified.

Combined match with two sets of most recent common ancestor (MRCA)

John L Barnes and Eleanor Baker and Jacob Barnett and Amanda Thomas 359 cM or 9.7 percent of the DNA from both sets of families as below, in addition to below matches.

50 percent or 1859 cM consisting of Barnes, Baker, Owens, and Copeland families

MRCA John Sanders Barnes and Sarah Owens 34.6 cM or 0.9 percent

MRCA Willoughby Baker and Rachel Copeland 69.2 cM or 1.9 percent

50 percent or 1859 cM consisting of Brunson, Franklin, Thomas, and Merrill families

MRCA Jacob Merrill/Holland Merrill 62 cM or 1.7 percent

MRCA Isaac Brunson and Margaret Oldys 6.4 or 0.1 percent

Brunson unconfirmed MRCA 10.4 cM or 0.28 percent

Maternal DNA 3718 cM of which a total 321.1 cM or 8.6 percent has been identified.

50 percent or 1859 cM consisting of McCurdy, Sunday, Beck, Pyburn and Chitty families

MRCA James Chitty and Mary Brown 35.9 cM or 0.97 percent

MRCA Elijah McCurdy and Barbara Sunday 109.3 cM or 2.9 percent

MRCA William M McCurdy and Amanda Beck 67.2 cM or 1.8 percent

MRCA William M. McCurdy or Elijah McCurdy/Barbara Sunday 33.7 cM or 0.9 percent

MRCA John Sunday or Elijah McCurdy 13 cM or 0.3 percent

MRCA McCurdy immigrant 10.4 cM or 0.27 percent

MRCA John Sunday 11.9 cM or 0.32 percent

MRCA Barbara Sunday (x chromosome) 39.7 cM or 1.1 percent (representing 21.8 percent of X chromosome DNA)

Note, possible some of Elijah McCurdy and Barbara Sunday matches may be matching the Beck family and not the McCurdy (100.3 cM match with this person, some confirmed with other matches).

In this branch, ¼ (464  cM) each should be McCurdy, Beck, Chitty and Pyburn. Note, only Chitty DNA has been identified other than McCurdy. A total of 285.2 cM (61.5 percent) has been identified as McCurdy/Sunday branch, not including the X chromosome matches.

50 percent or 1859 cM consisting of Hardy, Nelson, and Johnson families

no matches confirmed

Blood is thicker than water

To further my use of DNA and assist in my genealogy I recently tested my mother's half first cousin, SH and her nephew JM. Both are lineal Y DNA descendants of the earliest known ancestor Steely Hager. The 23andme haplogroup is from the 2010 tree, but using the 23andme to ysnp program and then the ISOGG tree add on, I was able to confirm that both have R1b1a2a1a1c2b haplogroup on the 2015 tree, enumerated by FTDNA as R-L48/S162 in the U106 subclade.

Unfortunately, I have not found one Hager from North Carolina lineage who is in a YDNA project to see if they even have the same haplogroup. Because 23andme tests with snp's, the haplogroup is accurate, though useless to compare with the Y STR tested for the YDNA surname projects, I was hoping to find someone with the same haplogroup to look into the genealogy to confirm a connection. Hopefully one will pop up some time in the future.

That said, the advantage of testing my mother's half first cousin is they only have a mutual grandfather in common. That means the 572 cM they share in common represents for each of them 15.4 percent of the approximate 25 percent of DNA each would have inherited (61.6 percent). This has been greatly beneficial for identifying matches as either Paternal or Maternal along those segments, and not only that, but further delineating them to only one grandparent's branch is definitely easier to trace a genealogical tree.

I am hoping to further identify segments by testing my mother's second cousin, at least one of her father's first cousins, and similar relatives on my father's side. By testing a first cousin once removed or a second cousin that narrows the matches to just one pair of ancestors. By selecting individuals who share 3-6.25 percent DNA in common on average, you are potentially discovering 6-12.5 percent of one parents DNA contribution. (Keep in mind, the tables that give the average shared DNA give percentages for your entire DNA, but when testing a relative on one side, the result is actually double that in context of that particular parents match.).

So how successful have I been in the first 6 months of research on my mother's DNA matches? I decided to take a look, and here is what I found.

Paternal DNA (roughly 3719 cM total) 23.1 percent identified

50 percent or approximately 1859 cM of which a total of 5.7 percent has been identified beyond a mutual grandfather. 572 cM (15.4 percent) identified for all of this branch in common with a mutual grandfather.

Hager to include Barnett, Bagwell, Whitley, Adams, Collins, Mangum

Most recent common ancestor (MRCA) William “Dan” Hager/Eda Martha Burnett 174 cM or 4.7 percent

MRCA Daniel Collins/Talitha Adams 1 percent or 38.3 cM

Connection to Barnett (identification of MRCA not made yet) 1.9 percent or 71 cM

25 percent or approximately 929 cM total identified 272 cM or 7.3 percent

Adams to include Rogers. No identifiable Adams matches because of his unknown parentage. Additional matches to the wives whose surnames are not verified are included, though the exact connection has not been determined yet.

MRCA Henry/Mahala Rogers 170.7 cM or 4.6 percent

MRCA Dauswell/unknown wife 27.9 cMor 0.75 percent

MRCA Susannah rumored Shue (X chromosome match) 27.2 cM or 0.73 percent

Identified as a match to Rogers (27.1 cM) included above and probable Adams (47 cM) from an adoptee. Since his exact connection is the unknown portion  represents an additional 1.2 percent

 25 percent or approximately 929 cM Trahern, Riddle, Hall and Choctaw lineage.

Only three matches with no verifiable connections at this point totaling 24 cM or 0.6 percent with known Choctaw descent and no other possible common ancestors.

Maternal DNA

Approximately 3718 cM in total of which 310 cM or 8.3 percent has been identified.

50 percent or approximately 1859 cM Hinds and their early colonial ancestors and the Paxtons of unknown lineage

MRCA Francis M. Hinds/Emma Paxton 46 cM or 1.2 percent

MRCA John Corliss/ Mary Wilford 66 cM or 1.8 percent

50 percent or approximately 1859 cM Timmins, Brampton, Spicer and other English, non American lineages.

MRCA George Henry Timmins and Sarah Brampton 198 cM or 5.3 percent