You know you have an obsessive personality when a cousin’s DNA results come in and you put off as many plans as you can for the next 3 days in order to explore them. At least I have some observations to report on the new Ancestry chip as well as more data for my study of the Wold family.
Wold line cousins Kitty, Ed, MM
According to Ancestry, this new chip has dropped some less interesting SNPs and replaced them with medically relevant ones as well as ones more useful for determining ancestry composition. The details are at http://blogs.ancestry.com/techroots/customer-testing-begins-on-new-ancestrydna-chip/
GEDmatch only tokenized 455K of the 700K SNPs from that new chip. However when I imported the raw data into a spreadsheet I saw that there were 668,961 lines of data as opposed to the previous 701,495 (then subtract 20 for the header), so not that different a number. New is chromosome 26 which is for the mitochondrial DNA.
My Wold cousin MM is the cousin whose doorstep I arrived on, Ancestry kit in hand, because I really really wanted her results. Those of you who have been to my Triangulation talk or read the article here may have noticed that I had no other cousins tested who are descended from my great-grandmother’s brother Charlie, the one presumed to be Kristine’s great-great-grandad. MM’s test has rectified that although she is descended from a different wife of Charlie’s than Kristine.
Since MM is the half first cousin of Kristine’s grandfather, she and Kristine are half first cousins twice removed. Do they match at the expected level?
According to Blaine’s chart, the average match for a first cousin once removed is 208 cM with a range of 27-415. Therefore a half first cousin once removed should match for half that amount. Here are the results for cousin MM versus Kristine at GEDmatch
| Chr | Start Location | End Location | Centimorgans (cM) | SNPs |
| 4 | 169,929,384 | 185,625,481 | 26.6 | 1,813 |
| 5 | 93,013,557 | 124,667,367 | 26.9 | 2,885 |
| 11 | 12,871,394 | 37,102,227 | 33.6 | 2,779 |
| 20 | 58,729,743 | 62,364,868 | 11.3 | 614 |
Largest segment = 33.6 cM
Total of segments > 7 cM = 98.5 cM
4 matching segments
Just about the expected amount. Additionally, I find that larger chunks like they have are a good indication of a closer relationship. So we now have another proof that Kristine is a Wold. When I compared cousin MM to all the other Wolds tested however there was quite a wide variation. Here are the totals of segments greater then 7cM for those cousins she matched (most all of them matched)
total cMs relationship who 98.7 half 1st 2R Kristine 63.8 2nd 2R Mike 146.7 2nd 1R Katy 51.2 2nd Kittys Dad 32.3 2nd 1R Kittys brother 18.3 2nd 1R Kitty 35.2 2nd 1R George 80.4 2nd 1R John 19.6 3rd 2R LW
These numbers demonstrate yet again the randomness of DNA inheritance once you get past close family. Katy seems to have more than her share of Wold DNA as she matches everyone at somewhat higher than expected levels.
The ranges are crazy they are making my head hurt!!! In searching for my grandmothers grandfather I was looking for someone who fit in the 1C1R category. I actively hunted down a great granddaughter of one of the men in question and she tested for me but didn’t fit this range. Meanwhile on my grandmothers test was a match that was at 144 cms so I had not paid that much attention to it but looking at this same match with her three siblings one had 349, the other 269 and the last 265. I was like wow I think I have it!
Kim,
DNa inheritance gets pretty random after close family but it does sound like you found them, well done!
Only after reading this did I discover that the picture wasn’t cousin Katy! Am still a bit stunned 🙂
Millie, so sorry, I am still very grateful you got Katy to test, Her DNA has been invaluable for studying the Wold DNA and solving this mystery
This demonstrates the bi-lateral usefulness of DNA in family research.
I think we all have had DNA matches with unknown individuals who after we contact them and delve into their paper tree find a matching ancestor. I would say this is quite common.
In this case the reverse process occurred. A paper trail filled with hints of a possible relationship, from changed surnames, incorrect spelling, unknown mother and unsure locations, pointed to a possible Father/Son relationship between Charlie WOLD and George WALD…….A relationship that in the end we had high levels of confidence in despite all these variations, yet there was always that unsure area. DNA analysis confirmed that the sketchy paper trail was correct and limited any doubts we had…..
Now with this new DNA result from a first cousin twice removed we have even more confidence in this WOLD/WALD relationship…..
So there you go, one can use DNA results as a source of possible family members or one can use it to confirm a suspected relationship with questionable paper trail….it works both ways….
Thanks to Dennis’ persistence in tracking the Wald/Wold trail, we found out what happened to the first child of my grandfather. I think somehow we knew it was a boy, but what happened to him was a mystery. Looking forward to meeting more of that family -Ed and I have been meeting at times, (see photo) and this last time was with Kitty also.
funny how many Kristine, Kerstin. Kristina, and also the number 26 shows up in my family (Grandpa Wold fathered 26, according to the birth records we found) Just think how many cousiins we have yet to find!)
MM
So dear cousins, I have just realized that MM descends from a different wife of Charlie’s than Kristine so they are only half first cousins once removed. This means that the shared DNA amount is exactly right.
I will correct the article now!
Hi Ms. Cooper,
I haven’t taken a DNA test but I really want to. But, my paternal great uncle did. This paternal great-uncle is my father’s father’s brother. He was told he was Y haplogroup C3 (now C2, I think because someone said they changed the name in 2016). I was wondering if I would have this paternal Haplogroup too. Thank you so much!
Jacob – your Y test is expected to be the same as your great-uncle although 1 or 2 mutations can happen since you are 2 generations away but the expectation is an exact match.
There are several different types of DNA tests so while your Y will be the same, your autosomal results will be quite different. Read the DNA basics page here for more on that.
Thank you so much for your reply! So, if there happened to me a mutation or two would I still have Haplogroup C3, just with minor differences of the Y-Chromosome?
Yes the same haplogroup is expected. Any changes would be on the more genealogically relevant STRs. It is the SNPs that determine haplogroup. See the newbie FAQ page here for more on that
Haplogroups reach back thousands of years. Read more about C here http://www.eupedia.com/europe/origins_haplogroups_europe.shtml#T
I received this from Christi in a comment on another post
For those of you wondering, chromosome 26 is what they call the mitochondrial DNA (mtDNA) and chromosome 24 is the Y while chromosome 25 is the PAR region where X and Y combine
Kitty, I noticed something interesting with the new ancestry chip. I just got the results for my 1st cousin – he has 727 4th cousin or closer matches. I tested his sister some time ago (old chip) she has only 360 4th or closer. I also have about 360. Wondering what you think and what others have experienced.
Interesting Jan, I have not noticed this for my kits. My cousin’s new chip kit has 78 fourth cousins or closer. Three of her grandparents are scandinavian immigrants (or children of).
The variance I see among my kits in 4th cousins appears to be entirely based on the number of colonial ancestors. My brother has only 40 4ths or closer (only one grandparent born in the USA and that one a child of immigrants), a cousin in Oslo has 61, my 1st cousin whose father is Pennsylvania Dutch has only 116 while a 3rd once removed with some colonial has 461. A client from the south has a whopping 1485!
So my guess is the brother has just inherited different DNA which is more common in the gene pool.
I recently had a family member test at Ancestry and his test was run using the Ancestry V2 chip. I downloaded his raw data and compared it to my raw data (which was done some time ago using a V1 chip). The raw data file header describes which array was used. Here is what I found using Excel to analyze the two files:
SNP Total on V1 = 701478
SNP Total on V2 = 668942
V1-V2 Difference = 32536 (32.5k or 4.6% fewer SNPs in the V2 data file)
SNPs in V1 file that are MISSING from V2 file = 276254
SNPs that are the same in the V1 and V2 data files = 425224
This is consistent with the data reported by mskitty and Christi in an earlier post.
I do not know which of those SNPs are used for *matching*, i.e., which are significant in the matching algorithms and then whether any of those have been removed or changed.
No conclusions or inferences drawn yet.
Greg, thank you so much for these numbers
I’m redoing my test just for comparison sake. Both my wife and I show 4% Irish on the ethnicity estimate. My son took the V2 test and he shows 12% Irish.(The 12% is more accurate based on my paper trail) They obviously have fine tuned their ethnicity snps.
I’m a junkie now. Anestry V1 and 2, FTDNA and 23and me along with Big Y, Y111 and mtDNA.
Griz,
Welcome to the club! next step is to talk as many cousins into testing as you can! Did you see this wonderful article at the Huff? http://www.huffingtonpost.com/entry/i-have-an-dnaddiction_us_57a8c531e4b02251db3fbcca
When I look at some “shared marches” on Ancestry, I do not understand why sometimes all the members of one “shared match” group do not always share with all the members if the group. Hope I’m clear about this. For example, let’s say that my record shows that I share a match with Smith, Jones, and Brown. But if I go to Smith’s record, he might show only shared matches with me and Brown. Why not with Jones? Thanks for all your help.
Having a group of shared matches with someone does not guarantee you share the same common ancestor. The matches may be from different parts of your trees.
Even if you do share an ancestor there is no guarantee that everyone you and the match share with will also share with each other. The randomness of DNA inheritance can mean that Jones and Smith did not happen to get any of the same DNA from that ancestor so they do not match each other even though they match the rest of you.
This scenario is true of more distant cousin matches like 4th or further out. Unlikely to happen with 2nd cousins or even 3rds.
There are a few references to being able to upload raw 23andMe V4 data to DNAland, download it from DNALand which will have filled in some of the “missing” data with null responses, then doing some form of “massaging” to the data. It would then be upload able to FTDNA’s autosomal transfer. But I can’t find any specific info on which file from DNALand to d/l (they have several imputed forms.) nor what this massaging is?
Then, of course, one has to wonder if the new Ancestry data could be uploaded in a similar manner.
I wondered if you, or some of your readers, might point me at more information? They only link to anything that I have found, is on a 23andme forum, which I do not have access to (a cousin tested there, but I have not)