What Are the Odds on DNA Painter – The Best DNA Genealogy Tool Ever!?!  – A Segment of DNA
Articles Blog

What Are the Odds on DNA Painter – The Best DNA Genealogy Tool Ever!?! – A Segment of DNA


Do you want to know what I think is one
of the best, if not the best tool for analyzing DNA?
It’s called WATO and you’re gonna learn about it today. Howdy,
I’m Andy Lee with Family History Fanatics and this is a “A Segment of DNA.”
Be sure to subscribe to our channel and click on that bell if you want to be
notified about upcoming episodes. Today I’m going to tell you about the
WATO tool or “What are the odds?” and the reason why this tool is so great
is because it allows you to analyze an unknown person with the known
DNA that you share with them as well as with other people. So let’s jump over to DNA Painter
and I’ll show you how the WATO tool works. Now the WATO tool was developed by Leah Larkin and it is on Johnny Pearl’s website –
DNA Painter. And this is free to use
so sign up for an account on DNAPainter. Then you can use the WATO tool all that you want and be able to save some of that information. Now on the tool section, we go down and here is the “What are the odds?” tool and
it comes up with our canvas of what we’re going to be looking at Let me go down and share some information at the bottom here First off, if you want to learn more about this tool then go on over here to Leah Larkin’s blog and in her series
“Science the Heck Out of Your DNA” she explains all of
these statistics behind what the WATO tool is. But for those of you who aren’t
interested in knowing that much statistics. It’s okay. You can actually use
the WATO tool without understanding all that. You just need to understand a few basic principles. Now, as you’re doing this,
there’s a few things to remember and Leah mentions this in the caveats down here. First off, when we’re using the WATO tool, it is not designed to have double relationships. So if you have double cousins in some way then the WATO tool is not necessarily going to
give you the results that you want. They may be inaccurate. The other thing is 3/4 relationships. For instance, if you have a mother who
had two children from different fathers but those fathers were brothers that
would be 3/4 siblings. and that is another relationship That is not really handled by the WATO tool. And finally, matches that are known to have a lot of endogamy. The reason why this is a problem is that
it increases the overall DNA even though that relationship may be far distant. And so you’re going to get inaccurate results if you have a lot of endogamy for that match But knowing that, let’s go on up and let’s start using the WATO tool Now I’m going to use an
example today that is a real example that I use in some of my research on my family. I’ve changed the names so that
all the living people are not identified But the first thing we want to do is we want to create a tree of known matches
and how they relate to each other. So you need to start off with at least two
matches that you know how they relate to each other and then we can build
the hypothesis from there So in this case, I actually have
a group of four matches that that I know how they relate to each other. So I can start with one person and
what I can do is I can just type this person’s name in and I’m going to type
in the first person’s name. What I’m going to do right now is
I’m going to put the amount of DNA that they share with my target person in parentheses And from there now I can
start adding parents and children. In this case, I’m just going to add a parent. And now the parent shows up. Then I can change that name. I can then add their parent and I can change that name. And I can add their parent. At some point, you’re going to
get back to a common ancestor. Now that may be a common couple or it may be a single person. In the case of where you have half-siblings that you have your matches descended from. So in this case. I’m actually going to put in the common couple that
I know all these matches are related to. So this is James and Christina
So there’s my first line. So this is my first match that I have
with my target unknown person. And now I can start to build my second match with this target unknown person. Now the other matches are all related through a sibling of this Benjamin. So what I can do is I can go to the parents and I can add another child and this child is Kristina. And then Kristina had three children,
so I’m gonna add three children. And each one of these children all have
one match as one of their descendants. So I’m going to edit these. I got Herber.
I got Jesse. And I have Frank. Now each one of these has a child. And each one of these also have a child
which ends up being the match So I’m going to just leave that child name as child. I’m gonna add their children and put in the match information. So for instance, right here I have Jane, and right here I have John, and then right here I have George. All right, now the reason why I’m putting the
amount of centimorgans in parenthesis is just so that I can be changing some of
these amounts to show you how different things work. But this is the basic tree that you’re first doing You have your matches that you have how much they match with the target person and you have how those matches are related to each other. I’m going to name this based off of my target person. In this case my target person’s name is Elizabeth. So this is a tree for Elizabeth. So now next is what I want to do is I want to put in the amount of
shared DNA that each one of these people has with Elizabeth. And so when I go to the menu there’s enter the shared CM and I can enter that number in So in this case is 155.7.
Now you can enter in the amount of centimorgans or if you have the percentages for
instance from 23andMe or from MyHeritage you can also enter the percentage
there. And Jane has 33.8 and John has 32.9 and George has 36.2. So I’ve put in all of the amounts of shared DNA that they have with this Elizabeth,
this unknown person. And you’ll notice Elizabeth is not
on this map yet, so now it’s time to actually put Elizabeth on this map. Now how are you going to know where to put Elizabeth? Well usually, what I like to do
is I like to go and I find the highest amount of shared DNA,
so, in this case, this would be Jack and then I put that number
into the shared cM project tool and that will give you the list of different relationships and then I’ll go
through and just start putting these in. So let me just start adding some
different relationships. Now the first one, since this is
actually pretty close over here, I’m going to add a child to Carl. And I’m going to maybe make that a grandson. And so here I want to use this
as my first hypothesis. And you can see that this
hypothesis has a score of zero right now which means it’s not really
possible. That it’s too close of a relationship for it to be that person. Well then what I’m going to do is I’m going to go back over to this child.
Now this is the brother of Jack. And I’m gonna define a half relationship
I’m gonna say this is actually a half-sibling of Jack. Now when I say it’s a half-sibling of Jack, all of a sudden that hypothesis becomes possible. Now it’s not good enough to just have one hypothesis because the one hypothesis means that it’s either possible or it’s not possible. What you want to have is
you want to have many different hypotheses so then you can ask
to compare which hypothesis is the most likely. So let me go and start putting in
some other hypotheses So for instance, we did a child of Carl. So let’s do another child of Benjamin
and we’re going to add children to that. And then we’re going to
start putting hypotheses in. Now we can put as many hypotheses as we want. So let me just add these two right here as hypotheses. So this would be Jack’s grandfather, another son or daughter from that grandfather, their grandson or
great-grandson. And you can see all of a sudden these numbers are starting to change a little bit. Hypothesis one has a score of 7. Hypothesis three has a score of 68,570. And hypothesis two has a score of 1. So what does this mean? Well, let’s go down and we can get
to the explanation of the hypotheses and you can see this is calculated that hey
hypothesis three this means that this location of that match it’s about 9, 796 times more likely than the next hypothesis. So if this was the only possibilities then what you can do is you can really focus
on looking at hypothesis three in other words, the great-great grandchild of
Benjamin through one of his sibling or through one of Carl’s siblings. Now the thing to always remember with the WATO tool, and one thing that the
Shared cM Project isn’t always going to show, particularly when you get out to
more distant relationships, is half relationships. There’s always the possibility that
somebody descended through a half relationship. So what I can do is I can add another child to Benjamin. And now I’m going to define a half relationship. I’m going to say that this person is a half-sibling of both Carl and this other child. Now if you don’t have DNA, you don’t have to put all of the children. So for instance, Benjamin may have had
Carl and John and Fred and George, but since I only had DNA from Carl, he’s the only one listed
because John and Fred and George they’re all represented by this other child
right here. Now, this half-sibling might not be known to the family. So let’s add him and let’s go and add a couple of children. And then let’s start making some hypotheses. So let’s make this one a hypotheses. And let’s make this one a hypotheses. And we can see that we can actually add another child here. And we can make this one a hypotheses. Now the numbers have changed even more. So we have our hypotheses 3 used to be the most likely. But as we go down, we can see
that hey hypotheses 5 is now the most likely. And it’s twice as likely as what
hypothesis 3 was And hypothesis 3 is twice as likely as hypothesis 4. So let’s go take a look so hypothesis 5 is the most likely at this point. So this is actually through a half relationship through Jack’s grandfather. And then hypothesis 3. And then we go to hypothesis 6 So we’re starting to see
where some of these might. But again, we’ve only gone just a couple of
generations. And if we actually go back and we look at this we can see, well this person related through all these others. What if there was another half
sibling? So let’s add another child and let’s add another half sibling. So we’re going to define the relationship as a half sibling to this Benjamin and
Christina. And then we’re going to add some children. All right. So let’s start looking at what
possible hypotheses could be. So I’m just going to add some different hypotheses. I’ll add another child and add use that as hypotheses. So I’ve already got up to nine hypotheses. And so you can see as you keep on adding more and more hypotheses, it gets the separation of which ones are most likely a lot, from the ones which are least likely. So for instance, this hypothesis two it has a
score of 1. It is extremely unlikely because some of these others are scores
are in the hundreds of thousands. Okay. So let me add something else here. Let’s go to one of these children and let’s add a line that we’re gonna look at this as the hypotheses.
There’s sometimes where a hypotheses is going to be zero. That means it’s really not possible based on the odds. Because this person Elizabeth,
who was represented by each one of these hypotheses, she has to match Jack, Jane, John, and George all at these amounts of centimorgans in order for that relationship to exist. And so if you come up with any hypotheses that are zero, we can eliminate those right out. I’m gonna just eliminate this entire
line right here. So this is now, where we can start using this in order for us to
research. What we’ve done is we’ve narrowed down to the most likely locations, on this family tree, where this unknown match fits. Now this unknown match maybe somebody who is adopted and so they don’t know who their biological
parents are. It may be that they’re just starting in genealogy and maybe they only have one or two generations of information available. Or it may be that they’re just not answering you and you’re trying to figure out how they fit
into your tree. So once you’ve figured out this, then I would actually start on
the most likely and try to narrow it down. Now some of these might be able to
be narrowed down and how can we do that? There’s a couple of ways. First off, is I look at it is that generation even possible? So let me take a look at
hypothesis number seven. Now it has a score of 20,000, which is, you know, seven
times less likely than the most likely, but it’s still somewhat high. But could this person actually be that person? Well, if I know that Jack died a few years ago
and he was 95 when he died So let’s say that, you know,
he was born a hundred years ago in 1919, and I know that Elizabeth, remember that’s who all these
hypotheses represent, I know that Elizabeth is 25 years old. Well then between James and Christina and Jack and James and Katrina and
Elizabeth is only three generations and so how likely is it that there is
somebody who is a hundred and somebody who is 25 in that same generation? It’s really unlikely. Now if you happen to know some more information about these other children or you know some more information about Elizabeth’s parents, as far as when they are born, that might help you to narrow down or to
eliminate some of these hypotheses altogether. And if you can do that, then you can shrink down where this person might be. So let’s say this based on ages
that I have eliminated hypothesis. I can go through and I can remove that
hypothesis And so that one’s gone, and I just have these ones left. And likewise maybe I can go through and I can say you know hypothesis six isn’t
impossible because Elizabeth is, you know, 80 years old. She’s almost as old as Jack, but There’s not going to be that many generations between Jack’s grandfather
and Elizabeth in this case. So let me remove that hypothesis as well. And I can actually delete that child. So I’ve narrowed it down now to just 7 of
these hypotheses and still hypothesis five is the most popular one And then that’s followed by three. And so that is where I would start my research is trying to figure out who those children and their descendants are or if there’s any half-siblings. Now you can do that by
finding other matches that may match in the same place. You might find some other matches that are closer that you can start adding in. So, for instance, let’s go
through and let me say that I actually found a child of this person, who had a descendant,
who had some DNA. And we’re gonna call this person Andy, because I like my name.
but it’s not really me. Now if I add the amount of shared cM that
Andy has with Elizabeth, then these hypothesis numbers are going
to change. And let’s in this case say that I add a 155,
I actually share as much as Jack. You’ll notice now that there’s a couple of
these hypotheses that are no longer possible, even though before they were very possible. So let me go back and let me
change that back to zero. And you’ll notice that hypothesis 3 is actually the second most probable hypothesis. So Elizabeth might be in this hypothesis 5
or she might be in this hypothesis 3. But when I add the centimorgans for this
other match in. And let’s say that it’s a 155. Hypothesis 3 is no longer possible. So I can actually eliminate that one And in this case, now hypothesis 5 is by far the most likely. It’s 10 times more likely than
hypothesis 6 down here at the bottom. And we can see that in the table. Every time that you make it change this,
this table is going to update and help tell you where it is that these are like Oh, I’m sorry! It wasn’t hypothesis 6.
It was hypothesis 4 at 1,479, that is the next most likely. What you enter in for matches matters,
and more importantly, if you have a known match that doesn’t share DNA. that is important also. If we add in another child.
Let’s add in another child to this Herbert And let’s call this person Caleb, that would be my son, but that’s not who this is. Now we have DNA from Caleb.
And Caleb, he matches this family. we know he fits into this family,
but Caleb doesn’t match Elizabeth at all. So I’m going to put in zero centimorgans
for the amount of DNA. Now that may matter depending on how closely related this person is. So for instance, I put 0 cMs in here, and these numbers didn’t change at all On the other hand, if I change and Andy to 0 CMs, you can see that the numbers did change. All the sudden, hypothesis 5 is up to 411,000, and six is still high, and seven now is one of the highest ones as well. But not only that, these are
far higher than what the other one is. So if you know that somebody doesn’t share DNA with somebody then that is really important. So as you’re building your WATO chart,
make sure you’re remembering you’re comparing your matches to one
single person. Now the great thing about this is, it doesn’t matter if that
information is coming from any of the different web sites as long as you have
a match with that. One thing I would say is if you are using Family Tree DNA
information, make sure that you put in a corrected amount. Now why do I say that? Let’s go and let’s take a look at an example here. If I changed this Andy match to what I might get from let’s say 23andMe (35 cMs). We can see that hypothesis 5 has a score of 62,544. And the top ones is hypothesis 5, hypothesis 6, hypothesis 7. Now the one thing that
Family Tree DNA does that other companies don’t do is Family Tree DNA
adds in some of those small segments. And so that can add up to
a significant amount. Now I’m gonna just change this up, to let’s say 95 which is
completely possible I have several matches from Family Tree DNA that really
share a low number of centimorgans so because of all those little tiny
segments it adds up to a lot. Well once I add that in these scores
changed. Now all the sudden hypothesis 5 is 17, 659. Now, in this case, the hypothesis 7 is lower than hypothesis 4 but hypothesis 5 and 6 are still in the same spot. Now that’s just because I have one match that had a difference in that.
If you have several matches that have that you may be getting inaccurate results.
A lot of the WATO information is from analysis that Ancestry has done and so
reporting all these tiny segments in with the DNA really throws off what
those relationships are. But if you’re using information where you have a match that you have information from MyHeritage, or 23andMe, or AncestryDNA, or you
know even Living DNA, or you’ve gotten it off of GEDmatch, then that informationis usually within that six or seven centimorgans and just looking at those
segments that are larger than that. And so I just put those information in just
as it is. So this was just a quick introduction to the WATO tool. I hope that this is going to be helpful for you. If you haven’t use this and go ahead and
try it again. This is a free tool and it is excellent for identifying and
narrowing down how people might be related to you If you have any questions,
about the WAT0 tool, put it in the comments below and I’ll try to answer it. If you like this video be sure to give it a thumbs up and share it with all your friends.

14 thoughts on “What Are the Odds on DNA Painter – The Best DNA Genealogy Tool Ever!?! – A Segment of DNA

  1. Hi how are you..

    I am trying to upload my raw dna data on my heratige

    But it always says invaild format

    I tried zip, txt thay should be supported in myheritage

    But Still the same? Is the problem in myheritage website?

  2. I wish someone would focus on creating tools for adoptee's. It seems all the tools are for people who have their mother and father tested and they can look at each other.

  3. Where are we getting the CM information? Am I using how much I share with each of the known shared matches between me and my target person? Or am I needing to get the information from my target person?

  4. I do not like this. It says things like my kids are not mine. I was there. Both birth and conception and they are. No one matches anyone with anything more than a +1. The charts are better but still………………

  5. I have just spent the last hour or so using the WATO tool trying to find out how an adoptee in the USA is related to my family which comes from England and NZ. Fortunately he has several DNA matches with various different members of my family, so matches are not a problem. I have got a possible line of descent from an ancestor, but since the Dates are all after 1920 – ie less than 100 years ago – it is going to be hard to find the details. The odds are not terribly high – so far the highest odds I have is 26. So I am not sure if this means I am doing something wrong, but I will go with what I have for now.

  6. Please i need some help…. I am still facing problems uploading my raw data on my heratige? It always says invaild kit…. I tried to unzip it as well… But still the same…..!

  7. I just want to say thank you so much for your help, education and patience with all of us and especially me. I really appreciate you taking the time to help me out and taking the step further to send me those supportive words in your email. You truly are appreciated and I wish that I could do something or send something to you for helping me with my family DNA suprise. Thank you a million times.

  8. Andy if you have a family tree DNA which has been uploaded to GEDmatch and you use the gedmatch shred match is that ok?

  9. This was great. I am trying to determine who the spouse is of a person in my husband's tree. Two other trees show this woman married to two different men. In each case their descendants share DNA with my husband. Is there a way to show a person with two spouses? Now that I've written that, I think I have answered my own question. If married to man A descendants' shared cM is in 186 and 123. If married to man B descendant shares 9cM. Can I conclude based on descendants' cM amount that man A is the husband?

  10. Hello, I have watched this video a few times and am interested in playing around with this tool. I have a few questions on what you need to set up the tree. My great aunt has a match to a person named John (example) on Ancestry DNA. I manage her kit. I do not know how John fits into our tree although from shared matches with my aunt, I know the side of the family I should focus on. First question. I want to know how John fits into our tree. Is he the Target person? Second question: There are a ton of known cousins who took this test and match to both my aunt and John. I manage some of these tests but not all of them. I can determine how many CM the tests I manage share with John. I cannot tell how many CMs the tests I do not manage share with John. In your video you said we should build this tree with people who match the target person and put in the amount of shared DNA. Does this mean I can only use this tool on the Ancestry Kits I manage that match John. Unfortunately, John did not test at another company or put his data into Gedmatch. I hope this is clear. thanks

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top