Saturday, May 26, 2012

I Found My Third Cousin Today at 23andMe!

(I wrote this on 5/18, but hadn't gotten a chance to post it yet.)

Over on the Rootsweb Autosomal DNA List a debate is raging in regard to the chances of randomly matching an actual third cousin in the autosomal DNA databases at 23andMe or Family Tree DNA. Some feel that this is extremely unlikely and a number of our top atDNA genetic genealogists are weighing in on the subject. I have been following this intriguing discussion for a couple of days now and hoping that the astronomical odds quoted are overly pessimistic. With this is mind, I was especially excited to randomly discover and confirm an actual third cousin match today from 23andMe's Relative Finder.

Honestly, I am so busy helping other people with their autosomal DNA research that I rarely have time for my own. However, today I was quickly scanning my paternal uncle's new Relative Finder matches and I saw a predicted 2nd-3rd cousin.


Happily, this cousin accepted my invitation to compare our DNA and told me that he had the Moore surname in his family tree. When I looked up his unusual name on Facebook, I found that we had a mutual friend - the wife of my third cousin once removed through my Moore great great grandparents. She quickly filled me in on my match's genealogy and I discovered that we are, indeed, third cousins sharing our mutual great great grandparents Calvin and Mary/Martha (Armstrong) Moore. Of course, we don't share that much DNA at that level, only .43% of our DNA matches each other (3rd cousins would be expected to share about .781%). My sisters are also well below the expected amount at .52% and .30% and my second cousin from this line shares only .31% with him. My father's two siblings who are his second cousins once removed share 1.13% and .93% (expected ~1.563%). These numbers are a little lower than expected, but this isn't altogether surprising since autosomal DNA inheritance is so random and, especially after the second cousin level, I have found it to be pretty inconsistent. Regardless, the prediction by 23andMe was right on.

Since I don't share that much DNA with this third cousin, it was fairly easy to overlook him in my match list:
As you can see, I hadn't even sent him an invitation. The fact that I had an actual third cousin hiding in my match list, parading as "nothing special", gives me hope. So, please don't give up on autosomal DNA matching. I believe that our cousins are there just waiting to be found. If you can - test your parents, their siblings and your grandparents if you are lucky enough. If I didn't have my uncle's DNA in the database, in my haste, I probably would have overlooked this promising match and missed out on adding these three segments to my chromosome map.

Given this opportunity, let's look a little closer at my family comparisons with this cousin. The chart below graphically illustrates the DNA that my sisters and I share with our 3rd cousin. Each gray bar represents a pair of chromosomes. The colored bands represent where each of my sisters and I have a matching segment of DNA with our cousin. As you can see, my sisters and I each inherited unique patterns of our great great grandparents' DNA. Much of the shared DNA is on Chromosome 8, especially at the end where all three of us inherited the same small DNA segment from our Moore 2nd great grandparents.

Third Cousin Comparisons - Click to enlarge

Now, let's look at the inheritance pattern of two second cousins once removed.  This graph shows the areas where my dad's siblings match this Moore cousin. They share twice as many segments as two of us third cousins and three times as many as my sister who has the least DNA in common with them. Notice again, how little overlap there is between the two and that there is very little matching on Chromosome 8. This means that my dad's inheritance pattern was equally as unique since we sisters all show considerable matching on that specific chromosome with our Moore cousin.

2nd Cousin Once Removed Comparisons - Click to enlarge

Next, we can see that our second cousin (also a third cousin to this newly discovered Moore cousin) inherited completely different DNA from our shared 2nd great grandparents than my sister and I. Also, she only has two segments in common with him (light blue), but one is larger than any of those my sisters and I share with him. Her overall sharing is lower at only 23 cM versus 32 cM for me and 39 cM for my sister.  She is more in line with my other sister (not shown here) who only shares two segments and 22 cM with this cousin.

More Third Cousin Comparisons (Click to Enlarge)

As I have explained before, I tested my father's siblings since he is deceased, hoping to be able to use them as a proxy for his DNA. Since siblings only share an average of 50% of the DNA, you can see this plan does not always work out. My uncle (dark blue) shares almost completely different DNA with this cousin than my sister and I do.  The only commonality with this "new" Moore cousin lies on Chromosome 11 between my uncle and sister (illustrated by light blue and dark blue bars).

One 2nd Cousin Once Removed versus Two 3rd Cousin Comparisons

The same thing holds true for my paternal aunt (dark blue bars below). You can see all three of us match this cousin only on a small segment on Chromosome 8 (stacked colored bars). Since the dark blue on the chart below does not overlap any of the other DNA matches that my sister and I have with our Moore cousin, it is evident that my father's pattern of inheritance from these specific ancestors was quite diverse from that of both of his siblings.

Another 2nd Cousin Once Removed versus Two 3rd Cousin Comparisons

Since Calvin and Mary Moore are the sole common ancestors shared with this newly discovered cousin, then we can deduce that all of the shared DNA in those charts above originates with them. For me this means that this specific DNA is inherited from their son Willard Calvin Moore, my great grandfather. I can now identify these three little spots - one on Chromosome 1 and two on Chromosome 8 - as a gift from Willard to me.

Willard Calvin Moore (1877 - 1934)

[**Update - I have been receiving lots of inquiries about Mary "Martha" Armstrong Moore, so I will post her basics here. She was born on 2 May 1836 in Newport, Ontario, Canada to recent English immigrants Thomas and Dorothy (Hudspith) Armstrong from Northumberland, England. Mary had 13 siblings. She married Calvin Benjamin Moore in Ontario in 1854 and shortly thereafter moved to Michigan where she had, at least, eight children before dying in 1878 of consumption when my great grandfather Willard was only one year old. I would love to find an Armstrong cousin, so if this sounds familiar, please drop me a line!]

27 comments:

  1. Thanks for sharing, CeCe! Such good news stories are very encouraging. Now if only we could get the service providers to tag shared DNA segments with the names of our ancestors we inherited them from. :)

    ReplyDelete
    Replies
    1. Hi Helen,
      We are working on that. If they don't, we will!
      Thanks for reading and commenting,
      CeCe

      Delete
  2. Very nice find and write up. You are lucky to have so many close relatives to test to able to triangulate your results.

    Jim

    ReplyDelete
    Replies
    1. Yes, I am. It makes a huge difference. Thanks for commenting, Jim!

      Delete
  3. I agree with Jim on all accounts. This past week was not quite as exciting as yours however I got two 4th cousins in two days. One from FTDNA and the other from 23andme. Both paper trail confirmable 4th cousins once removed. That's the closest I've come at 23andme. One was a person previously unknown to me. The other was someone's father that I didn't even know was testing. I had corresponded with her erratically over the years.

    I also had some new Family Finder matches one shares 3 segments and .59%. No response yet.

    I do believe as more and more people get tested the odds will increase dramatically. I also wish I had older generation members to test (all deceased) as there's a huge fall off with my children. Many of my 22 cM or better matches either don't match them at all, or become 9 cM matches with my children. What that means is that many of those smaller matches I've been tending to ignore---may represent 5thcousins which are traceable.

    And like CeCe I have tested at all three (no results yet from Ancestry) to increase the odds (and drive me crazy).

    Kelly

    ReplyDelete
    Replies
    1. Kelly,
      That is super exciting! I think that is at least as good as my week. I agree with you and sure hope that we are starting to hit that sweet spot where a lot more of us - especially adoptees - get closer, traceable matches.
      As demonstrated by this find, having those older generations certainly does make a big difference. I hope this inspires other people to get them tested. I'm sorry that you don't have the opportunity, Kelly.
      I'll be interested to hear what you think about Ancestry's BGA. Expect too much Scandinavian and perhaps too little British Isles. That seems to be extremely common. I found a "confirmed" 9th cousin there yesterday on my Gillett line, but it kind of feels like "so what" since I have no way to triangulate to see if that is in reality the correct common ancestor and, if so, no way to add it to my chromosome map.
      Thanks for reading and commenting!
      CeCe

      Delete
  4. Sure wish my closest 23andMe matches would even bother responding... Congrats on the good match.

    ReplyDelete
    Replies
    1. Judy,
      You never know - one of mine from 2 years ago just responded! Perhaps as testing becomes more mainstream, people will check back in more often and become more active.
      I also have a lot of good ones that haven't responded still, but as I get more, those ones hurt less and less.
      Thanks for the comment!
      CeCe

      Delete
  5. Yay for finding close cousins! I've found a number of 3rd cousins at different levels of "removed." Comparing them to my Great-Aunt's DNA, our closest matches convert into her 3rd cousin and a 2nd cousin 2x removed. The matches are out there.

    ReplyDelete
    Replies
    1. That is great news, Valerie. I agree with you that the matches are out there waiting to be found! Thanks for your comment.

      Delete
  6. My best find so far on FTDNA has been what I believe is a 5th cousin - I have not heard back from this person, yet, but he includes three distinctive surnames that pinpoint the family for me, and the surnames that he does not include lead me to believe we have a set of gggg-grandparents in common.

    ReplyDelete
    Replies
    1. Hi Greta,
      That sounds like a great find. It demonstrates that even more distant cousins can be found using atDNA.
      I sure hope he answers you soon.
      Thanks for reading and for your comment.
      CeCe

      Delete
  7. Congrats!

    My Father has a predicted 3rd cousin (range 3-4th cousin) with 0.82% (61 cM) over 4 segments shared. This gives me hope of finding how they are actually connected.
    Of course this cousin needs to reply first - since Feb 2012.

    ReplyDelete
    Replies
    1. Thanks Paul. February isn't too bad. Hopefully, he/she will check back in soon and answer your request. I have had many people answer after quite some time.

      Delete
  8. Finding a third cousin through DNA testing is great, CeCe!
    I still have not heard from my 5% match.

    ReplyDelete
    Replies
    1. 5%!? Now, that is a close relative!! I sure hope they answer you soon. Do you have any idea who it could be?

      Delete
  9. Great post! I am a lucky one...found my birthfather via a third cousin on 23andme, earlier this year. All I had was an approximate spelling and the 23andme results for my birthmother and myself. Got a few names that meant nothing to me from a possible 3rd or 4th cousin. Put those names into ancestry.com and WOW, there was a family tree that matched. Now we're in contact and I'm getting to know my new family in Ireland! It works!

    ReplyDelete
    Replies
    1. That is so exciting, Suzzy!! Congrats!
      Can you please send me an email: yourgeneticgenealogist@gmail.com? I would like to hear more details.
      Thanks for reading and for sharing your inspiring story.
      CeCe

      Delete
  10. With my limited experience and knowledge, I have been fortunate enough to find two Proctor cousins. One is a fifth twice removed and one is a third cousin. We were able to verify the relationships with family trees. I have also found a couple Skinner cousins. If I knee what I was doing, I might be dangerous!

    Pat Skinner

    ReplyDelete
    Replies
    1. Pat, that is awesome! You must be doing something right! :-)
      CeCe

      Delete
  11. Ok I have a problem I am adopted and I know my mother line very well. I did the DNA matching and found a strong line with 2 third cousin and many fourth. I believe this has to be from my father side but I have no idea who he is .What is the best way to narrow or chart. so that I can find him PLEASE HELP <3

    ReplyDelete
    Replies
    1. Sorry that I missed this, DeeAnn. Have you joined the DNAAdoption group? We have a methodology there that will help you. You can also find it at DNAadoption.com.
      Good luck and please let me know what happens!
      CeCe

      Delete
  12. My wife has been using the 23andme to try and trace back to her father who was an anonymous sperm donor and recently was matched to 3rd and 5th cousins with only small percentages around .25-.43% can anyone tell me if those are good numbers can they be her relatives I'm not well versed in this subject and I want to support her in this but I don't want to see her get hurt chasing false hopes or leads any thoughts or suggestions

    ReplyDelete
    Replies
    1. Hi Mark,
      They are definitely her relatives, but they are pretty distantly related (probably 4th cousin and further). She should learn the methodology that we use for the adoptees over on the DNAAdoption group (linked on the right side of this blog). You can read about it at DNAAdoption.com. We are having great success with it.
      Best of luck on her search and thank you for being a supportive spouse! Please let me know how it goes.
      CeCe

      Delete
  13. Sure would be nice to have a graphic of your family tree so we could better see your relationship with these two relatives.

    ReplyDelete
  14. Hi,
    I received my 23 and Me report. My reason for doing the testing is in effort to find out who my father was/is. I'm a female in my 50's. I found out recently that the man I believed to be my father is not. The situation is very complicated. My mother is deceased. I found out after she died, via DNA testing.
    I collected several squares of blood-stained gauze from the medical appts I accompanied my 'dad' to, along with my cheek swab. Two different DNA testing companies confirmed, at 100%, that he is not genetically related to me. I have no means of acquiring DNA to parse my maternal dna. I am unable to enlist the help and support of siblings/family members. So with that: Yesterday 23 and me sent me my report. I have 4, second degree cousins listed as matches, sharing between 11 to 15 segments. Some with names, (which I do not recognize from my maternal side), and, some listed as anonymous.
    Can I assume that finding 'second cousins' is a hope-giving start toward the goal of finding out the name of my biological father? I'm truly at a loss as to how to utilize all this overwhelming information in finding out who my father was/is.
    I sent brief, introductory message to the 2nd cousin matches that just, basically, ask if they would be willing to have dialogue re their family tree. I didn't include anything else. Any advisement would be so very greatly appreciated. *I'm also awaiting results from Ancestry.

    ReplyDelete