It is very interesting to look at the overlapping DNA segments of one’s matches in order to figure out where they may have come from and how they might be related. It also helps with tracking these relationships and comparing the results from different sources. Personally I know when someone matches my 2nd cousin Dick that the relationship is on dad’s father’s line. My next task will be to start diagramming the possible relationships based on who matches whom among all our new distant cousins.*
To find the DNA segments where you match another person at 23andme that you are sharing with you put your cursor on “My Results” in the very top menu bar and then click on “Ancestry Tools” at the bottom of the second column of selections to get the page that lists Family Inheritence: Advanced. Within this function you select the person to compare to on the left and all those to do it with on the right. So typically I take a new person and compare them to me, my brother and my Dad; then various cousins. You need to select the table version to get the numbers shown in the spreadsheet below.
Here is how I track overlapping segments. I make one spreadsheet for each person I am looking at, sorted by chromosome, segment start, segment end and length. I use the same columns as the 23andme table view of my genome shares but add one column at the beginning marked with a P or M (paternal or maternal side match) whenever I know it. Then I add a column for the most recent common ancestor (MRCA), one for known relationship (when found), and another one for notes where I put comments like “same match with Kitty, no match to Dick, matches Jane elsewhere).
So as I find more and more DNA matches with Norwegian-American relatives who tell me “Oh we have an Andressen or a Larsen too,” how do I explain what little use that is? I try saying, “Well in Norway before the late 1800s most people used their father’s name (so Jonsen means the son of Jon) and perhaps their farm name for a surname. So please tell me what locality your family was from and the farm name if you know it.” Plus Per might be Peder in a different document and other names have other variations. Still one can’t complain, there are an enormous number of Norwegian records online and the records go back to at least the 1500s in most localities.
I have found two resources which explain the naming practices quite well:
John Føllesdal also has an excellent website hosted by ancestry.com explaining how to research your ancestors in Norway.
Well I thought I had found a real cousin on the German Jewish side due to a few common surnames but no luck finding the relationship yet. Sadly there is a large match on our X but the common surnames are not on a branch where X could come from. Of course one of the problems in German Jewish genealogy is that all but a few prominent families had no fixed surnames (they used their father’s name) until 1813/14.
Even worse she mainly matches me on segments that are what I call “Ashkenazi Pile Ups” or locations where there are well over 30 people matching me but not my Dad for more than 5cm. By comparison I notice only one such pile up on my 100% Norwegian father’s matches at chromosome 9, at about 80,000. But that will be the topic for another post.
These are the three pile ups my new distant cousin matches:
Chromosome 2: 45 matches for this segment at 23andme
Chromosome 4: 80 matches for this segment at 23andme
some start at 18.1 and some end at 25.0
Chromosome X: about 30 matches for me, 50 for my brother…. hmmm
a few are longer than this
For those of you who are wondering where to find this data on 23andme you can download all the segments that match yours with the name of the donor (most will be anonymous) by going to “ancestry labs” under ‘My Results” and clicking on “Countries of Ancestry.” Scroll down the page to the long blue button where you can download a CSV of all your matches. [updated 27 dec 2013 – n.b. you can get this data and more by using the http://DNAgedcom.com site to do the downloads]
There are a few more pile ups in my and my brother’s matches than these …
So in the images in my last post about GEDmatch you may have noticed that my Dad has less Mediterranean, Siberian, and Southwest Asian than I did. Perhaps you are wondering if there is a way to see what I got from my mother? Separating what you got from which parent is called phasing and you need to have at least one parent tested and uploaded to use this function at GEDmatch. The Genetic Genealogist has a good explanation of phasing in this blog post – http://www.thegeneticgenealogist.com/2012/06/07/gedmatch-com-adds-phasing-tool/
Here is what I got from my mother although GEDmatch cautions that the phased data may not be all that accurate.
Eurogenes 12b does not have as many Northern breakouts but does include Finnish which is of interest since Dad has several Finnish matches so I collected all the Finnish percentages. As you can see Mom had some too and Shipley got more than I did:
One of the best tools around for genetic genealogy is the GEDmatch site which allows you to upload your raw data from whatever service you used and compare it to everyone’s data at GEDmatch in many different ways. But my favorite tool is the pretty pictures of your ancestry mixtures (called admixtures). Here is mine using the Eurogenes K12 calculator which seems best for us Northern/Scandinavian folk:
What does it all mean? A discussion of these populations is here at the Eurogenes blog.