23andMe Automatically Creates Tree Using New Technology – Relationship Triangulation

23andMe has introduced a new tree feature that automatically creates a “tree” for you based on your predicted relationships to others and their predicted relationships to each other. 23andMe hasn’t coined a term for this, but I’m calling it Relationship Triangulation.

Let’s look at traditional forms of triangulation and how Relationship Triangulation is different.

Triangulation

Segment Triangulation – In traditional triangulation for genetic genealogists, we match the same (reasonably sized) segments between a minimum of 3 not-closely-related people to assign that segment to a common ancestor. Of course, in this scenario, you need to know who your ancestors are – at least some of them.

Once enough people test and match on that segment, hopefully at least a few will be able to identify a common ancestor or minimally, an ancestral line. You can read about this type of triangulation here and a more detailed article here.

Tree Triangulation – Genealogists use what I refer to as “tree triangulation” when segment data isn’t available, like at Ancestry, or when they are seeking to determine parentage.

Adoptees use this method of triangulation where they look at the trees of their closest matches, hoping to discover a common ancestor in their close matches’ trees – because that points the way to a descendant of those ancestors who is their biological parent.

I’ve described this technique in the article, Identifying Unknown Parents and Individuals Using DNA Matching.

Relationship Triangulation

What 23andMe is doing with relationship triangulation is different yet.

They are using the same techniques used by genetic genealogists manually to try to place their matches in their trees. In essence, we perform the following steps:

  1. Look at the predicted relationship provided by the testing company in order to get an estimation of where that person might fit in our tree – in other words, how far back in the tree the common ancestor will be found. First cousins share grandparents, second cousins share great-grandparents, etc. Additionally, I utilize the relationship table at DNAPainter to view alternative relationships based on total shared DNA.
  2. Look at who we and our match matches in common. For example, if the match also matches my first cousin on my father’s side, there’s a high probability that’s where the person fits on my tree. If they also match with my second and third cousin on my father’s side, especially if they match on the same segment (traditional segment triangulation) then I not only know which “side” they match on, I can push the match back several generations to the common ancestor or ancestral couple I share with the third cousin.
  3. If I also have detailed haplogroup information about the ancestors in my family tree, and my match also has haplogroup information, I can use that to rule out or include lines as potential placements. 23andMe provides haplogroup estimates at a high level, not at a detailed level, but they can still serve to rule out various lineage placements on the tree when compared with other people. For example, if a person’s mitochondrial haplogroup is J1c, they can’t be the child of a female ancestor whose haplogroup is H1a. This is exactly why I have constructed a DNA Pedigree Chart.

What is 23andMe Doing?

I don’t have any inside information about exactly how 23andMe is constructing this tree, but they published the following. The difference, of course, is that they have automated the process and they don’t have customers’ trees to work with.

This information is found on your account at 23andMe under the 3 dots at the upper right-hand side of the tree itself, which I’ll show you how to access in a moment.

23andMe tree options.png

23andMe tree about.png

23andMe tree predicting.png

23andMe will probably provide additional information and make adjustments as this feature comes out of beta – including the ability to modify relationships which are inaccurate – and several are.

Beta is Required

In order to utilize the new tree feature, you must enable Beta for your account, which you can find under your name in the Setting area.

23andMe FamilySearch settings

If you need help, I wrote detailed instructions for enabling Beta in this article.

Accessing Your Tree

You can access the new tree feature in 2 ways.

23andMe DNA Relatives 2.png

At the top, under Ancestry, select DNA Relatives

23andme DNA Relatives selection.png

Scroll to the very bottom.

23andMe Family Tree beta.png

To see your tree, click on the Family Tree link.

The second way to access your tree is under DNA Relatives. When you click on any specific person, if they are in your tree, you’ll see the following:

23andMe tree cousin.png

The “View your Family Tree” button takes you to your family tree.

Your Squiggly Relationship Triangulation Tree

Please click to enlarge this tree in the image below. The font for names is small, but at 23andMe, you can zoom in on various sections.

23andMe tree

Click to enlarge

When you first see your tree, it looks a bit odd and takes a some getting used to because it looks different than the trees genealogists are used to working with.

Here’s what the blue box contents say:

23andMe tree blue.png

23andMe tree blue 2.png

23andMe tree blue 3.png

I tried to see if there was a way to add people in my tree. When you click, the node enlarges, but there isn’t anything to be done yet.

23andMe tree closeup.png

How Is 23andMe Creating Relationship Trees?

23andMe is constructing my family tree based on how I’m related to people, and how they are related to each other as well. I wouldn’t be surprised if they are also using their basic haplogroup information.

Let’s look at how this works.

My two children have tested. I did not tell 23andMe that they are my children, but because both of their mitochondrial haplogroups are the same, and I’m related to both of them in a parent/child relationship, and they are related to each other as siblings, I have to be their mother. That’s the only combination of relationships that works for the three of us.

If one of them was my parent, they would not be related to each other as siblings.

One could be my parent, and one my child, but then, again, they would not be related to each other as siblings.

Furthermore, my son would carry my mitochondrial haplogroup, but I would not carry my father’s mitochondrial haplogroup, so my relationship to him cannot be that he is my father. It would be very unlikely for a father’s mitochondrial haplogroup to match a child by happenstance although that’s not impossible, especially with only partial haplogroups with no full sequence testing. If you’re interested in the difference, you can read more about that here.

Lastly, my children would be related to a subset of the people I’m related to, but (in general) sharing less DNA, so a generation removed. If one of them was my parent, I would be related to a subset of the people they are related to.

23andMe is also looking at who shares DNA with other relatives.

23andMe tree relatives.png

In this DNA Relatives chart comparing me with cousin Laura, you can see that she and I share a common segment of DNA with my twin and with “J”, meaning we have segment triangulation, although my V4 kit doesn’t count for triangulation because it’s a duplicate. Additionally, Laura and I both match Trista, but Laura, Trista and I don’t share a common DNA segment between the three of us.

Laura, “J,” Trista and my V4 test are all shown in my tree. We’ll analyze those matches momentarily.

My Two Results

You’ll notice that I have two tests at 23andMe. I took a V3 test and another V4 test when it was introduced to see if my results were the same. 23andMe accurately calculated this as a “twin” relationship.

In general, the results are close, but not exactly the same. Interestingly, my V3 tree has 2 more cousins than my V4 tree. Of course, the need to “fit people in” on the tree means the layout looks different between the two trees, so I’m only going to work with one set of results in order to reduce confusion.

It’s confusing to look at both trees side by side, because these trees are auto-generated to “best fit” the branches with the available screen space – so the branches are not in the same place on both trees. As more cousins are added to your tree, the layout will change.

Paths

For each person in your created tree, you can click on the person you match and the path from them to you highlights.

I have to laugh, because there is a quilt pattern named “Drunkard’s Path,” and it’s much straighter than this path from my second cousin, Patricia, to me.

23andMe tree path.png

Of course, today there are no names on these “?” ancestor nodes but I know who they are.

Let’s take a look to see if these placements are accurate.

Identifying Nodes on the Tree

Based on where my matches are placed, I determined the identity of the “connecting couple” nodes on my tree and utilized Snagit to mark up the 23andMe tree, below.

You can see that I’ve labeled my mother and father, based on what I know about the various people that I match. For example, I know where my Ferverda cousins reside on my tree, so those cousins must be on my mother’s side.

Conversely, my Vannoy cousins must be on my father’s side.

In my case, my mother is on the left and my father on the right, which is backwards to the normal genealogist pedigree way of thinking. However, without any tree information at all, 23andMe is flying completely blind.

23andMe tree markup

Click to enlarge

I analyzed each of my 18 matches that 23andMe placed on my tree, of my 1553 total matches at 23andMe.

Accurate Placements

The gold stars indicate accurate placements. There are a total of 7 gold stars, BUT, of those 7, one is my own second kit which clearly is my “identical twin.” Two are my children whose accounts I manage. Of the balance, there are:

  • one 1C1R
  • one second cousin
  • two third cousins

There are a total of 4 accurate placements that are not immediate family members whose kits I manage.

Why might a placement not be accurate? 23andMe says the following:

23andMe unexpected.png

This is exactly why I utilize the DNAPainter segment tool to evaluate different possibilities for the matching total centiMorgans.

Unknown Placements

Green boxes on my tree are people who are unknown, but based on their common matches with known cousins, they are in the right general area of my tree. I wish I could tell more.

With 23andMe’s historic lack of tree support, and few people having added surnames or recently, ancestors through Family Search, it’s impossible to determine how these matches descend from common ancestors. The best I can do based on common matches and shared DNA is to determine and assign “sides” and sometimes roughly how many generations back to the common ancestor.

Hopefully, once people can actually identify ancestors on their trees, this may well improve. I’m concerned that so many years have passed with no tree support that many people aren’t signing in anymore, will never notice the tree feature and won’t add anything. I hope I’m wrong.

There are only 4 people in my tree whose relationship to me is unknown. Perhaps I should send them a message. I do know approximately where they belong on the tree, so I could at least ask some leading questions with surnames I think they might recognize.

Uncertain Placement

The gold box is an uncertain placement. Based on common matches, this person descends from Ollie Bolton’s parents’ line, not from Joel Vannoy and Phebe Crumley. However, Diane also matches Patricia M who descends from my mother’s line, so she could be doubly related to me. Those lines are not even from the same state, so this person is a bit of an anomaly.

This causes me to wonder how 23andMe will handle situations where a match does descend from both sides of a person’s tree.

I should send her a message too.

Inaccurate Placements

The red boxes indicate placements that are known to be incorrect, and why they are incorrect. When the placement needs to be moved to another connecting node couple, I’ve drawn a red arrow.

I’ve numbered the red boxes so we can discuss each one.

In box #1, Cheryl is identified correctly as to her relationship with me, as is “J,” but their relationship to each other is inaccurate. They are half siblings, not first cousins.

In boxes #2 and #3, these people are shown descending from my grandparents, when in reality, they descend from my great-grandparents. Of course, this makes their relationship to me inaccurate too.

In box #4, my half-grand-niece, meaning my half-sister’s granddaughter is shown as descending through another child of my grandparents, when in fact, she descends from my father and his first wife. My father needs another spouse in the chart and the relationship needs to be calculated accordingly.

In box #5, both individuals are actually one more generation further down the tree than they are shown, so the relationship needs to be recalculated.

Summary

How did 23andMe do with their automated Relationship Triangulation tree construction?

23andMe summary.png

  • One third, 33%, are inaccurate..
  • The close family matches, 17%, are accurate, but really don’t count – those were freebies.
  • There are exactly as many unknown as accurate, 22%.
  • 6% are uncertain, meaning I can’t tell if this person is accurately placed or not, because the matches are confusing.
  • Of my total 1553 matches, 1.16% were placed in a tree. I wonder if 23andMe has drawn an arbitrary line at 3rd cousins, at least for now.

Given that this tree was entirely constructed by 23andMe without any genealogical foreknowledge, based only upon genetic relationships to me along with genetic relationships of my matches to each other – this isn’t bad at all. It’s certainly a start.

Adoptees must think they’ve died and gone to the happy hereafter, because all other vendors’ tree support requires you to actually HAVE a tree of some description. Of course, adoptees and people seeking an unknown parent don’t have trees for their unknown parents.

23andMe creates a tree from scratch for you based only on genetic relationships – Relationship Triangulation.

While only one third of these matches are accurately placed today, all were at least on the correct side of the tree, with one confusing possible exception. Genealogists already know that things like pedigree collapse and endogamy will complicate efforts like this.

My mother’s side was accurately identified to appropriate great-grandparents, but my father’s side wasn’t as clean, especially where “half” relationships are involved.

It would be interesting to combine the Leeds Method or Genetic Affairs clustering with the 23andMe technology and see what those results look like.

23andMe would be in a much better position had they never obsoleted tree support years ago. They have never been a “genealogy” company, with their focus always having been on medical genetics. Genealogists, with their incessant need to know were the perfect people to attract to test. 23andMe has always given us “just enough,” but never trees or the heavy duty tools we need.

Perhaps the tide has turned and this is their way of reintroducing a hybrid genetic+genealogy tree.

The Future

In the short-term future, 23andMe is going to add the ability to define and modify relationships on this tree, which will help them refine and improve their machine learning tools. This will also help with their medical research initiatives because clearly, the companies they partner with want to know specifically, not generally, about the heritability of medical traits.

Trees encourage genealogists to provide 23andMe with family information for their medical research, while trees clearly benefit genealogists too.

In the longer term, for genealogists, let’s hope that the marriage of genetics, machine learning, trees and technology produces helpful tools.

This Relationship Triangulation tree is interesting and fascinating, if not yet terribly useful. This technology holds a lot of promise and every innovation begins with a imperfect tenuous first step.

Who knows what the future will bring, either at 23andMe or perhaps as other vendors integrate this same type of technology.

You can’t play if you don’t test and 23andMe does not accept uploads.

If you would like to take the 23andMe genealogy only test, click here or if you want the 23andMe genealogy + health test, click here.

If you have already tested, check your account for your new tree and tell me what you think!

____________________________________________________________

Disclosure

I receive a small contribution when you click on some of the links to vendors in my articles. This does NOT increase the price you pay but helps me to keep the lights on and this informational blog free for everyone. Please click on the links in the articles or to the vendors below if you are purchasing products or DNA testing.

Thank you so much.

DNA Purchases and Free Transfers

Genealogy Services

Genealogy Research

58 thoughts on “23andMe Automatically Creates Tree Using New Technology – Relationship Triangulation

      • That’s good to know I’ve been wondering why I couldn’t get the menu items you mention in your article. Very keen to see what they say! Thanks again Roberta for letting us know about this development!

      • My triangulation tool is working now, so in my case it took about 24 hours to generate.
        However, I was surprised to learn that the tool needs at least two tests to work: a grandparent and a grandchild. In my case, my grandchildren took the FTDNA test, so it would cost $199 each to have them take the 23&Me test in order for the tool to work for me.
        Sigh…….

        • Kay, are you talking about GrandTree?

          I dont have any grandchlldren (or grandparents) and my Beta Family Tree is working fine, albeit with misplaced matches.

          • Ah-ha – Yes, I was referring to GrandTree, which I had not noticed before!
            When I first tried to check “Your Family Tree,” it was still generating, so when I checked again several hours later, I saw both trees listed and didn’t remember that the new one is called “Your Family Tree.”
            I just checked YFTree, and it looks fine!
            Thanks!

  1. Thanks for this blog… It contains much of the information I was wondering about. The tree is unique and has led to some new discoveries and ways of working my matches. Yes, there are corrections needed, but I trust there will be a way to work through those.
    Kudos to 23 and Me…. and to you for explaining it all. Thanks, Roberta!

  2. I do not have a Beta Tree option in any menu! I have checked in settings, Ancestry and DNA rRelatives and the Family and Friends heading. The menu you show is under Family and Friends, but tgere is no Beta Tree in my list of options in the menu.Do you have to be part of the Beta testing program before being offered this Beta Tree option? Is it only for people who have signed up to.be a Beta tester?

    I would like to know more about being a Beta tester before signing up for the program.

    • Yes, the article says that you have to be a Beta, and also provides an image and link with detailed instructions for how to be a Beta. There is more informatoin when you sign up.

  3. fun to play with it. i found it surprising that they were able to accurately assign matches to the correct maternal and paternal sides (though there are other errors which i will correct when that function is added. it is pretty beta – i couldn’t assign the correct birthplaces for many of my parents/grandparents/great-grandparents.

  4. Thanks for bringing this new update to our attention! I had signed up for Beta participation a while back (I think based also on one of your blog posts), but I wasn’t yet aware that this feature was yet live until this post. I too had many nodes placed incorrectly (i.e. several nodes placed under great grandparents that should have instead been under 2nd great grandparents). Also, one thing they didn’t get right is that they are using the relationships that they **think** are correct, BUT in several cases I had already given them the correct relationship by making the correction in the match (for instance, they thought a match was a second cousin, once removed but is actually a second cousin). One would think that they would use the corrected relationship we have already given them rather than their original guestimate of the relationship. Also, at some point, I hope they give us the ability to make corrections to the trees they have developed. That would not only help them to improve their algorithms but would also aid in better identifying and matching where other people fit in other trees. As an example, I purposely did not fill in the information for one set of my great grandparents because they had just as many nodes under that set of great grandparents who actually descended from them as they did nodes for people who descended from one of their parents (my 2nd great grandparents) and I feared that by giving them the names of these great grandparents, they would use that information as a “fact” that the others also descended from them when they did not.

  5. Thanks for letting us know about this, Roberta. 23andMe is the only one of the 4 major testers I’ve not tested at yet – for many of the reasons you’ve pointed out over the years I’ve been reading your blog.

    But, I must admit, this “Relationship Triangulation Tree” is intriguing. You surmised that 23andMe has maybe drawn an arbitrary line at 3rd cousins, at least for now. For me, that’s not far back enough yet for me to entertain the thought of taking their test just to see this resulting “tree.”

    I’ve been doing genealogy since the 1970’s and don’t really need help on 3rd cousins and closer. But if and when they were to take the arbitrary line back to the 4th cousin level (and maybe more), then I’d say that could be a VERY interesting prospect and I’d seriously consider taking their test just to see it.

    • I’m hoping they expand it since I need triangulation at the 4th grandparent level and beyond. It has a lot of potential. I’m waiting on a kit to arrive and can’t wait to see the results.

  6. Roberta
    Of course I had to go and look. Nothing to write home about in my case. Apart from my mother, my son and a cousin once removed, nothing is linked – just a vague link to a whole bunch of third cousins, none of which seem related to each other and all assigned to my mother. So I log in as my mother – and her tree shows some but not all of mine and her tree does show some other second and third cousin that are linked (related?) to each other. Some dots have question marks allowing me to enter the personal info of the close relatives – like entering my family tree manually once again…

    In the case of my son, he has me, my mother and this second cousin – nothing beyond.

    I would say very BETA…

  7. I’m actually excited about the possibilities with this feature, but right now I’m seeing what I view as a rather glaring problem. My closest relative beyond my father, my siblings, my daughter, and my siblings’ children and grandson is a *maternal* 2nd cousin. If I filter for “father’s side” this cousin does not appear; and if I look at my father’s DNA match list, he also does not appear.

    Despite that, 23andMe has drawn a line connecting my father through an unnamed parent to a pair of (unnamed) grandparents, then down through a couple of generations to my 2nd cousin. So my father’s grandparents are shown as my 2nd cousins great grandparents. That would indeed make him my 2nd cousin, but the problem is that he is clearly not on my father’s side. If I filter for “not father’s side” (since my mother did not test), my 2nd cousin does appear.

    What’s more, 23andMe has drawn in lines that connect the parent who isn’t my father (my untested mother) to ancestors who connect in turn to relatives who are shown as “relatives in common” to both my 2nd cousin and me — but not my father.

    In other words, it should be very clear to ancestry that this 2nd cousin is a maternal relative and not connected on my father’s side. My father does happen to have the same name as my eldest brother, except for the distinction of “Sr.” versus “Jr.”, but 23andMe shows both my father and my brother in the correct positions in the tree. So I have no idea what has caused this error.

    I hope this will be quickly resolved, because it doesn’t give me much confidence about 23andMe’s ability to make predictions using DNA. As the tree now shows the connection, my 2nd cousin would be my father’s 1st cousin once removed — yet he shares 0 DNA with this person. (My brother with the same name as my father shares 265 cM in 12 segments with our cousin, while I share 207 cM in 11 segments.)

    That said, once they get such silly little bugs worked out I think this will become a very helpful feature.

  8. My family tree is on FindMyPast but I can’t find any information about how to link it to 23andMe. I’ve tried copying a link to the FindMyPast webpage for the tree but this is rejected by 23andMe.

  9. Thank you for another great tip, Roberta! I was disappointed to find a number of wrongly assigned matches; i.e., maternal vs. paternal. They are clearly not using the same techniques as Gedmatch does for one to one and triangulation. I’d also love for them to provide a feedback mechanism, especially for tree-building. If I have only a year of birth or death, they change the year to 12/31 of previous year, for example. Also, I much prefer the traditional date format of day month year to mo/da/year. I look forward to being able to move people to their correct tree positions and to add new people. I’m surprised that they bring so few people in from one’s FamilySearch tree. It’s a lot of work to identify each node beyond parents and grandparents.

  10. I turned on the beta in 3 accounts but this feature only became accessible to one of them. Are they not showing it to kits with early chip versions?

    Good news is the one that shows reflects assignments by side I made by comparing kits with yes in the share list combined with looking at specific segments, The relationships on the side I had been actively researching were accurate.

  11. Roberta, were you by chance using an iPad or other tablet when you tried to add a relative. I had this problem at first, then realized the input form was “hidden” in the upper left corner – just the bottom margin was showing. I used the zoom feature to bring it into play and continued adding relatives.

    HTH.
    Kathleen Cooper

  12. I signed up for 23andMe beta tree builder. I noticed right away that it doesn’t handle a man having children by two women correctly, at least in my case..

    My grandfather had my uncle and my dad and a few years later he had a son by another woman. The descendants of this woman are shown correctly, but the grandson of my uncle is shown as descending from my great grandfather.

    My parents, grandparents and all aunts and uncles are deceased.

      • Yes, apparently for purposes of 23andMe’s tree, half relationships do not exist — even though 23andMe explicitly mentions them in its explanation of how they determine relationship.

        “These relationships are predicted using DNA shared among many relatives, and the relative ages between them. For example, a relative may be predicted as a half-sibling if they are within about 10 years of your own age.”

        My wife and her half brother are only 19 months apart — so less than **2** years, let alone 10. 23andMe is still insisting that her brother must be her uncle. The thing is, we’ve already told them what the relationship is — half siblings.

        To be fair, it can be difficult to tell the difference in a DNA comparison between uncle versus niece or half brother versus half sister. For instance, my wife and her half brother share 1926 cM in 51 segments. I share 2113 cM in 51 segments with my brother’s daughter.

        But suppose 23andMe had access to a list of relatives common to both. Oh, let’s suppose it might be called something like “Relatives in Common”. If they had something like that, they’d be able to take advantage of the fact that siblings — including half siblings — are related to all of their shared relatives in the same way, with the exception of descendants. The only difference between half and full siblings here is that for half siblings, their shared relatives are all on the same side.

        By contrast, uncle and niece are from two different generations. Thus for most relatives, an uncle will be a generation closer than the niece will be. (Exceptions would include the niece’s siblings and descendants.)

        So **if** 23andMe had something like this, they could just look and see that my wife and her half brother have relatives who are all the same degree, except for my wife’s daughter. In 10 pages of shared relatives, with almost no exceptions is someone listed as a “1st cousin” for one, not also listed as a “1st cousin” for the other — and typically with a very similar percentage of DNA shared. Often, when the percentage is different, it’s actually a little higher for my wife than for her brother.

        Let’s just look at the 1st page, omitting my wife’s daughter. The first column is the relative’s relationship to my wife, the second is the relative’s relationship to her half brother.

        JB — 1st cousin, 8.35% / 1st cousin, 8.99%
        LP — 2nd cousin, 4.36% / 2nd cousin, 2.81%
        PR — 3rd cousin, 1.59% / 3rd cousin, 0.96%
        EE — 3rd cousin, 1.36% / 3rd cousin, 1.10%
        BK — 3rd cousin, 1.04% / 3rd cousin, 0.89%
        JD — 3rd cousin, 0.99% / 3rd cousin, 0.70%
        MP — 3rd cousin, 0.88% / 3rd cousin, 0.79%
        WSJ — 3rd cousin, 0.81% / 5th cousin, 0.17%
        RB – 3rd cousin, 0.81% / 4th cousin, 0.48%

        By contrast, let’s look at the Relatives in Common between me and one of my (full) nieces. The first column is the relative’s relationship to me, the second is the relative’s relationship to my niece.

        RB — 2nd cousin (2.78%) / 3rd cousin (1.39%)
        LQ — 3rd cousin (2.13%) / 3rd cousin (0.60%)
        JW — 2nd cousin 1xR (2.11%) / 3rd cousin (1.07%)
        EW — 2nd cousin 1xR (2.02%) / 3rd cousin (0.97%)
        MM — 3rd cousin (1.66%) / 3rd cousin (0.93%)
        IW — 3rd cousin (1.63%) / 4th cousin (0.38%)
        JR — 3rd cousin (1.62%) / 4th cousin (0.31%)
        MH — half 2nd cousin (1.48%) / 3rd cousin (0.87%)
        SM — 3rd cousin (1.44%) / 3rd cousin (1.00%)

        I could have chosen either of my tested nieces or either of my tested nephews, and the results would have been pretty much the same. My point is, because of the generational difference — with absolutely no need to reference age — it’s easy to tell whether someone is likely to be a half sibling or a full uncle or aunt. The relationship of any cousin on the same side as the uncle or aunt should be closer than to the niece or nephew. My 1st cousins are my niece’s and nephew’s 1st cousins once removed; my 1st cousins once removed are their 2nd cousins; and so on.

        (I have a half uncle who is *younger* than I am, and my older brothers have a full uncle who is younger than they are — and he’s less than a year and a half older than I am.)

        • In my case the half family is shown in the right generation. The full family is one generation too far back.

          Do they have a place to report problems?

          • Just to clarify, this 2nd cousin IS in the DNA Match lists of all of my siblings and my daughter. It’s just that he is clearly a **maternal** relative with no connection to my father. As a matter of fact, his paternal grandfather and my maternal grandfather were brothers.

          • Actually, I did report my problem to customer service. I anticipated that the response might be that it’s a beta feature, and there might be some “bugs”.

            Their *actual* response was nothing so reasonable.

            “I understand you are seeing some changes to your predicted relationships in. It’s important to note that our relationship prediction algorithm has been updated, so the predictions you are seeing are the most accurate predictions we can provide at this time.”

            Although I talked about the problem in an earlier comment, I’ll mention it again here.

            My father, who tested at 23andMe and is linked to me and to all of my five siblings, is shown as linked to a 2nd cousin of mine. This person in **not** in my father’s DNA Match list, nor in “Relatives in Common” between either my father and me or between my father and any one else in the group — so not with any of my siblings. He also isn’t a “Relative in Common” with either of my brother’s tested children, with either of my sister’s tested children, or with my sister’s grandson.

            There’s zero connection. Nada. Zilch. Except … 23andMe shows my father in his proper connection to me as my father, then it has a line running up from my father to his (unnamed) parents, and then from one of those parents to his grandparents on that side.

            From this set of my father’s grandparents — paternal great grandparents to me — a line runs downward to one of my 2nd cousin’s grandparents, then to one of my 2nd cousins parents, and then to my 2nd cousin.

            Another line runs down from this same grandparent of my 2nd cousin to a sibling of my 2nd cousin’s father, then to that person’s child, then to a grandchild. This person is a 2nd cousin once removed to me, and a 1st cousin once removed to my 2nd cousin. But no relation at all to my father, and there is no shared DNA with my father.

            I even sent a screenshot of the tree, as requested, and it clearly shows the connection to my father.

            And for what it’s worth, when I joined elected to participate in the beta, all 10 of the profiles in my account ended up being included. So we now all have trees. I am the only one of this with this problem. My 2nd cousin is not included as a relative of my father in my father’s tree, no in any of my siblings’ trees, nor in my daughter’s tree.

            I’m afraid my reply to “Customer Service” might have been just a tad on the snarky side. What she probably *should* have told me is, “Hmm. That’s odd. Probably a glitch. Please give it some time.”

            Just goes to show that when you don’t really know the answer, you probably shouldn’t try to give an answer anyway.

          • I have found 23andMe to be very arrogant in the past too. Thanks for sharing.

  13. I think someone else may have mentioned this, but it appears 23andMe is ignoring an obvious means of determining the side someone is on — at least in the case of males.

    If you’re male, the only possible source of your X chromosome is your mother. So unless your parents are related — which of course is sometimes the case, but most often not — if you share a significant segment with someone on the X chromosome, then you’re likely maternally related. Certainly, at least in part.

    As it happens, most of my X chromosome came from my maternal grandfather. (All except 20 cM around the centromere.) Therefore, it came from his mother — the great grandmother I share with my 2nd cousin.

    Unfortunately, this is no use with my 2nd cousin himself, since he is descended from her through three males in a row. Even two males in a row will not pass on an X chromosome. However, I also have a 2nd cousin once removed who happens to be my 2nd cousin’s 1st cousin once removed. That is, my 2nd cousin’s father was a 1st cousin to my 2nd cousin once removed’s maternal grandmother.

    What this means is that there aren’t two males in a row with my 2nd cousin once removed. The path is from my great grandmother (his 2nd great grandmother) to the son who is my 2nd cousin’s paternal grandfather and my 2nd cousin once removed’s maternal great grandfather, to my 2C1R’s grandmother, to his mother, to him. On my side, the path is from my great grandmother to my grandfather to my mother to me.

    So it wouldn’t be at all unlikely for this 2C1R and I to share X chromosome DNA — and in fact we do. We share a 31.45 cM segment that includes 2,103 SNPs. That’s definitely a significant segment, and it confirms that he is related to me on my mother’s side.

    In spite of this, 23andMe shows my 2nd cousin once removed as being related to me on my *father’s* side. They do this even though my 2C1R is not in my father’s DNA Relatives list, just as my 2nd cousin is not.

    I have tried to explain to 23andMe that this can only represent a GLITCH, and not a legitimate function of their Family Tree algorithm. But I’m not sure they’re listening.

  14. Thanks for pointing us to this new feature! I gave it a try. What I get is that my matches are assigned to the correct grandparent’s family (well, three grandparents of the four are shown with matches, anyhow), but my grandparents are mismatched. In other words, my paternal grandfather is matched to my maternal grandmother and vice versa. The left side of the screen has my father’s maternal relatives (and nobody’s paternal), and the right side of the screen has my father’s paternal relatives and my mother’s maternal.

    There are, of course, other glitches like wrong generations and so on, but this is the main problem for me because it means I can’t fill in names on my grandmothers’ sides. It already makes it look like my maternal 1C1R is descended from my paternal grandmother…

  15. Love my short experince with that beta tree guessing at things. A number of close matches have never gotten back with messages, and the guesses provided for them pretty much answer my questions.

  16. prior to the update, my matches were assigned to the correct maternal/paternal sides. not any more – some have been switched, so i am reluctant to add my grandparents back. i guess we just need to wait until their system allows corrections.

  17. I keep getting the page that reads:

    Growing, growing…oops.
    We’re having trouble loading your tree.

    Our team is already on the case, so check back in a few days. Thank you for your patience!

    Anyone else? I had built up the tree. Then it changed and I built it up again, and now no tree.

    • My tree has been like that for about a week. I still have my sister and father’s tree working but I didn’t do much with them and I won’t now that I’ve seen what happened to mine. I know this can’t be easy, they don’t know our families and if you have parents who have cousins marrying cousins of the other parent, it’s hard to figure out where people belong. I told them that unless I can connect my mother to my father’s matches, you and our matches won’t ever see the big picture. It’s a tree made for Halloween. Very scary. 😉

      • Like most people, I do have some consanguineous marriages in my tree — with the closest being an ancestor who married the daughter of one of his 1st cousins. (I really can’t imagine any of my 1st cousins as my mother-in-law.)

        But, neither my parents nor either set of grandparents appear to be related to each other in any way. You have to go back to my great grandparents’ generation to start seeing any sort of cousin marriages.

        Even if my parents did happen to be slightly related — and sooner or later, we’re *all* related — you’d think the algorithm would be designed to “notice” if a parent it was connecting through failed to share any DNA with the person they were connecting.

        For now we’ll excuse these sorts of things as “glitches”, but in programming a glitch is simply a result that cannot be readily explained. It **does** have a cause, whether it’s faulty data or faulty programming, we just may not be able to find it easily.

          • I don’t know and they haven’t said. Not sure they know it’s maternal and paternal, just different sides.

          • I don’t think they know beyond that there are two sides (we say which is maternal or paternal if we haven’t tested a parent there). I think the reason they have swapped sides on my maternal and paternal Norwegians is that none of my first cousins have tested at 23andMe, so there is nothing to clue them in on which two of my four possible clear groups go together. So they guessed wrong, partly because while a 1C1R did test, they have not identified any matches at all for my maternal grandfather’s relatives, just the other three grandparents. So as they didn’t bring in the genuine existing matches on grandpa’s side, they can’t tell that the 1C1R is descended from him as well as from one of my Norwegian grandmas.

  18. I found another piece of nonsense when looking at the beta Family Tree of one of the other profiles in my account. The tree attached one of his relatives as a 2nd cousin, but the relative shares 705 cM with him.

    When I clicked on the relative’s circle and then “See other possible relationship”, I received the explanation that “Statistically, Valerie is most likely your 2nd cousin …”

    Someone should ask where they heck they found whatever statistics they’re using. At Ancestry, I have two 1st cousins whom I’ve known since I was a kid. Their sharing with me is 831 cM and 690 cM. Sharing 705 cM with someone does NOT reflect a relationship of *2nd* cousin. It could possibly be a 1st cousin once removed, or a half 1st cousin — but definitely NOT a 2nd cousin.

    This relative’s son is also shown in the tree, as a likely “2nd cousin once removed”. And, again, the “explanation” claims this is the most likely relationship. And of course, if his mother is my profile’s 2nd cousin, then 2nd cousin once removed would be correct. However, here the amount of shared DNA is 332 cM.

    That is NOT what you expect for a 2nd cousin once removed. This amount of sharing is indicative of *at least* a 2nd cousin (with no removeds), or perhaps a 1st cousin once removed.

    Again, I actually do have a 1st cousin once removed at Ancestry. In fact, six of them. The highest sharing is 633 cM, and the lowest is 344 cM.

    Once again, 23andMe’s Family Tree is not only not yet ready for prime time, it isn’t even ready to be a beta release.

    • I do have high hopes for the future of 23andMe’s Family Tree feature, I just think they may have released it to beta too early. I *hope* that won’t prove to be a fatal mistake for the feature.

  19. Thank you for another in-depth analysis of a new feature. I was relieved to read you too had a female on the left and male on the right which seems backwards to us. At first I thought I wasn’t entering my information correctly. My tree is such a mess, I can’t wait for the ability to make changes. I left feedback early on and once is all that’s wanted because I’m unable to leave more feedback with what I’ve learned since. I applaud 23andme for putting this out in beta and for attempting to create a tree based on the information they have.

  20. I think 23andMe is realizing there’s not much money to be made in health reports. They can sell data to third parties all they want, but people just don’t care. People want accurate ethnicity estimates at a the least, and the ability to trace ancestors all over the globe for thousands of years at the most. Most people get their ethnicity estimate and never log in again, but to compete against other companies, they probably realize people can gain interest in their family history with the right tools. I’m one such person who never had any intention of testing with 23andMe until this auto tree feature. I’m waiting for a kit to arrive now and I’m excited.

  21. I have a cousin’s login and thought of another thing worth mentioning since I’ve been playing with hers. Even though it has a tendency to wrongly assign matches to ancestors (whether they’re labeled or not), it’s still an excellent way of determining which segments you share with 1st and 2nd cousins.

    For instance, something I’ve noticed with my cousin’s matches: let’s say it assigns a 1C1R to the same source as a 3C. You know the 1C1R goes to your grandparent and the 3C goes to your great great grandparent. It’s extremely likely that many of the segments shared with that 1C1R are from a great great grandparent’s line.

    Let’s say your grandparents are A & B.
    A’s parents are C & D.
    B’s parents are E & F.
    Great grandparent C’s parents are G & H.
    Great grandparent D’s parents are I & J.

    23andMe assigns the great grandchild of your grandparents A & B to great great grandparents I & J along with a 3C that is known to descend from them, specifically branching off of J. They’ve just shown a high probability where the shared segments are coming from. It’s very useful even in the beta stage.

  22. So now “Family Tree” is no longer showing my maternal 2nd cousin and two 2nd cousins once removed as being related to me through my father. This is good, because he shares 0 cM with any of the three. Unfortunately, although these relatives had been correctly placed for all five of my siblings, now my youngest sister is showing the very same error. This should make it clear that Family Tree does not compare the trees of very close relatives — full siblings — to see whether they are consistent.

    All of my siblings have *exactly* the same relatives. We don’t necessarily always share the same amount of DNA, but the same relatives can’t be on the maternal side for some of us but the paternal side for others of us. A relative may conceivably be on both sides, but if the relative is on both sides for any the relative must be on both sides for all. (Though we don’t all have to share the same amount of DNA.) What’s weird is, why would correcting *my* results have have any effect at all on my sister’s?

    Another error that has now been fixed is that my wife is no longer listed as “niece” in her half brother’s tree. The two of them have sharing of 25.9%, which can equally support either “uncle and niece” or “half siblings”. But the tree is supposed to also take ages into consideration, and my wife and her brother were born in consecutive years (he, in 1955; and she, in 1956). So there really was no good reason for the Tree to have chosen “uncle and niece”. It’s possible for an uncle and niece to be this close in age, of course; but the tree has to play the “odds”, and the odds say that with this age difference it’s more likely for them to be half siblings — which is correct.

    But, guess what? My wife’s relationship to her brother did also happen to be changed in her tree. He’s no longer listed as her uncle. Now, he’s her nephew. The sharing is of course still the same 25.9% — with an equal chance for “aunt and nephew” or “half siblings”, but the age difference is also the same. Just one year (19 months). The tree *should* have been able to easily make the correct choice, especially since it did so in my brother-in-law’s case. And he’s the older of the two, so why change him from “uncle” to “nephew”?

    What’s more, DNA shared and ages aren’t supposed to be the only factors the Tree considers. A third factor is how your relatives are related to each other. Using this factor, my brother-in-law has close relatives who share 0 cM with his supposed “niece”. I’m talking about several people as close as 1st cousin or 1st cousin once removed. Since if he were his sister’s uncle, he’d have to be her *full* uncle, then she’d be showing sharing of 0 cM with people who were supposedly her 2nd cousins.

    But the Tree had figured out a way around this, by calling them my BIL’s 2nd cousins — even someone who shared 705 cM with him! That is *not* just a 2nd cousin.

    Now that his tree shows him as half brother, the explanation is clear. These people don’t share any DNA with his half sister simply because they aren’t on the shared side (which is their mother’s side).

    Here’s the thing, though, my wife also has a 1st cousin who shares 0 cM with her supposed nephew. Keep in mind, with the amount of DNA she shares with this “nephew”, she’d have to be his *full* aunt. Just as 100% of each parents ancestors are also your ancestors, so are 100% of the ancestors of a parent’s full siblings — because they’re the *same* ancestors.

    So if someone is a 1st cousin to my wife — sharing 952cM with her — that person would be a 1st cousin once removed to her nephew. There is no chance that the “nephew” would share no DNA with such a person.

    The easy answer, of course, is that if he’s her half brother then she is his half sister. But the tree ignores their small age difference (and the fact that he’s even the older of the two), as well as the fact that each has other relatives that would contradict being either uncle and niece or aunt and nephew.

    I’m still hoping that 23andMe will fix these sorts of *stupid* mistakes. And I can them stupid for one very simple reason. They are not the result of things that the algorithm can’t know. The are things that would be *correct*, if only the algorithm adhered to its own rules.

Leave a Reply