Last week I watched a webinar on Legacy Family Tree Webinars that threw everything I had planned to do out the window. The webinar was,“What are the Odds? An online tool that can help solve DNA puzzles,” and it was given by Jonny Perl, the owner of the website, DNA Painter.
(The webinar is FREE to watch for all time from what I gather, but having a Legacy Family Tree Webinar subscription is always a great value.)
Last year I wrote about the fact that I got a subscription to the DNA Painter site because of the ability they gave me to create multiple trees, but this new feature on WATO (What are the Odds) is FREE.
If you’re not familiar with their tool, WATO let me tell you … or you can always listen to Jonny explain it on the webinar (his analogy was excellent). But mostly what it is is if you have a mystery in your DNA or you have a match that you don’t know how they fit in your tree, or maybe you don’t see how you fit in someone else’s tree you can use the tool to figure out the possibilities or hypothesizes. It also gives you the odds of each of those hypothesizes. Many adoptees use this tool.
The new innovation as of last week is that now you can upload a gedcom file, and after you do a bit of data entry in the tree, you can auto-generate all the possible hypotheses and what the odds of those theories are. I’m sure this was to offset some of the questions people had about setting up hypotheses because before that was a bit of a challenge.
One of the things that I’ve done for my DNA is I have a unique tree for the Beaton/Batten Mystery. I don’t want to mess with other people’s research, so this tree is private and unsearchable.
As I discover matches that relate to the mystery, I add them. This is done by identifying the common ancestor between those several matches and then building the tree forward in time and ultimately putting the DNA match in the tree. Because I can’t find that common link (yet), to all these people and to me and the rest on my family, some of the new branches are floating adrift in the Beaton/Batten tree.
So this is how the WATO (Revision 2) works. Just like before, you start with your research question and your target person, but now you add the target person’s age. (see fig 1). The age field is new, and that is super helpful so that the theories make sense. Let’s face it if I was the target, you wouldn’t want the system telling you that you could be the child of someone born in 1845.
The next step is to add the gedcom. To do this, you go to the upper right corner, and you’ll see the load button.
When it comes to a gedcom file, it doesn’t care if the people are connected, so my Beaton/Batten gedcom will work. It’s just a file with all the names and branches. So I upload my gedcom to DNA painter, and then I just have to choose the ancestor(s) I want to start with.
So I made two attempts at using the WATO rev 2. The first one I did was to see if I could understand the Beaton/Batten Mystery a bit more. So I started by making my uncle the target person. I uploaded my gedcom file, and I told the program which person I wanted to be the ancestor. I chose James Grantham. James was married several times, and I have DNA matches that connect to my family from all of these three wives. (He’s the common ancestor of several of my DNA matches).
The next step is to go into my uncle’s DNA and put the number of cM that he connects to each of the matches that are in my tree. They’re easy to find as they are the last people on each of the branches of the tree. (see Fig 2) When I add a DNA match to my tree, I always name them their DNA profile user name, then I give them the middle name DNA, then the surname, and then in the suffix box, I usually put the amount of DNA they match me. (I do this so I can find my DNA matches easily in my tree when I search, and it doesn’t seem to affect any algorithms. ) (See Fig 3)
Once I had all that information I clicked the “suggest hypothesis” button. With my uncle as the target, it came up with over 600+ possible scenarios. Wow, that’s a lot. I would never have given myself that many had I done it myself manually in Revision 1.
So I decided to try someone else. This time I choose my mom’s cousin, Marg (not her real name). I know that she’s related to me through my great grandfather, Willie Beaton. She is the granddaughter of Willie’s sister, Annie. The trouble is that I don’t even know if William and Annie are full or half-siblings. But in this case, it doesn’t matter. Because I’m looking at how Marg is connected to all the DNA matches.
I could have chosen to save my uncle’s theory, but instead, I decided that 600+ possibilities were just too much, so I cleared the hypotheses but used the same gedcom. Then I went about entering how many cM Marg is connected to each of the DNA cousins and removed any that my uncle may have had that she didn’t.
This time when I ran the analysis, it came up with 153 hypotheses. Still a significant number but maybe not as big as I think. When I look at the predictions, each is given a score. At first, you’re looking for a big number score. The highest score was 339,345, and there were actually nine with that score.
I know this is hard to read (Fig 4) but I wanted you to just see how involved this is. I think what I’m trying to figure out is an extreme case.. I’m looking for a connection in 1874.
Here’s a closer look at what I think it can be telling me and where I might go.
I know that William (Willie) was born around 1874, so all of the scenarios that show grandchildren of James, George, and Eliza don’t make sense. But what if my William (Willie) was the unknown sibling or the unknown child of one of those people? That might make some sense to check further for these people and see if I can find some documentation that can help in the research. You know… right time, right place.
This new revision is in beta right now, so I’m sure that there will be changes as people find things that don’t work, but what it does for me is helps me think about where I might go to look further.
So the next thing I decided to do was to look at one of my old WATO charts. What you can do is take something that you created in Revision 1 and see what happens in Revision 2. In the original, I wondered how Marg fit in the family. Remember I said I wasn’t sure if her grandmother was Willie’s full sibling or half-sibling?
Once I brought the Revision 1 over to Revision 2 all I had to do was add Marg’s birth year and run the hypothesis.
In the first figure, Marg’s score is only so so. But once I change William and Annie’s relationship to half-siblings, then Marg’s score becomes one of the highest. I think this may lead me to believe that William’s and Annie’s relationship is probably half-siblings.
I said I only ran two examples, but that’s not exactly true. I got excited about this new tool, and I had been working on a client’s DNA case, so I thought I’d try it out on her DNA match tree.
It didn’t give me the answer but it gave me an idea of where to search further and I went back to my tree and kept working the tree and adding more cousins. Ultimately I found the two families that were her father’s parents people. The info was always there but perhaps this new tool put new life in my search.
There are so many tools available to you now to learn more about your DNA, and I think it’s important to try them and see if they help extend your family tree. Happy hunting.