Sunday, September 19, 2021

How to Group Your DNA Matches to Help Break Down Brick Walls

How do you break down a brick wall with DNA? It's what everyone wants to know - after all, what is the point of getting a DNA test if the ethnicity report is unreliable? Everyone says the true value of the test is in your DNA matches, but how do you utilize them to actually be useful in your research? To break down brick walls? To do what paper research couldn't?

This sort of ties in with my instructions on how to find unknown biological ancestors with DNA, though that was targeted more at NPE or adoption situations. However, the same basic process and workflow can be applied to breaking down brick walls. In the past, I've detailed specific cases where I've used DNA to break down a brick wall, but some of them are a little unique - every situation might be a little different, and therefore might require a bit of a different process. But here's the basics. 

In my post about finding unknown biological ancestors, in Step 1, it says, "Look for your closest DNA match that you can't identify as being from another known branch of your tree."

But wait - how do we even get to the point of finding a match you can't identify? You do that by identifying and grouping as many matches as you can. This is how my workflow goes, it works best for me, your mileage may vary, but in my experience, this is how most people do it in some way or form. Some maybe use a spreadsheet and the "Leeds Method", but ultimately, it's just a matter of grouping your matches by what branch of your tree they belong to, and since AncestryDNA have a built in grouping tool, I find that works best for me.

Grouping your matches.

Step 1: Create a group for each "branch" of your tree. Which branches? I recommend a group for each of your sixteen 2nd great grandparents, unless any of those 2nd great grandparents were from the same specific location, or endogamous population, because they will be difficult to tell apart. For example, my 2nd great grandparents who both came from the same tiny town in Italy called Monteroduni got grouped together because I have no other branches from there, and since the town is so endogamous, it would be difficult to always tell them apart. So I just have one group for "Monteroduni". Don't group by broader locations, like country. I did that by grouping my other 2nd great grandparents together because they were both from Norway, but now I regret that because they came from totally different parts of Norway, so there's no endogamy between them. So although I recommend a group for each 2nd great grandparent, depending on your ancestry, you may want to sometimes group them differently. 

16 groups does mean that it will fill up a lot of your available groups, AncestryDNA only allows you a maximum of 24, so you will only have 8 groups left to do with whatever you want. So like I say, you may want to group them differently, but this is what worked best for me.

Step 2: Start at the top of your match list and work your way down. Do you recognize your top match? Or can you see from their tree (if they have one) what ancestor you share? Is there a ThruLines/common ancestor hint for them that you can verify? If you already know the match or can identify how you're related to them, mark the branch you share by adding them to a group you've created for that branch. Do not assume a shared surname alone is the source of your shared DNA, it must be an actual common ancestor.

You may also want to add a note of your common ancestors, so you can see who they are more easily, and also so you know there's identified common ancestors (though I also have a group for MRCA - matches that have identified a most recent common ancestor).

My top matches are all my Italian cousins, you can see how
I've grouped them and added our MRCA to notes


Step 3: Do the same for the next match, and the next - keep going until you can't identify a match. When that happens, look at your Shared Matches with that person. Are any of them the people you've already identified with a common ancestor? If so, they are likely also from the same branch (especially if there's more than one match they share from the same ancestor/branch), so add them to that same group. 

I don't know my MRCA with Bettye because she hasn't added a tree,
but I can tell she's from my Smith branch because she matches
several people who are confirmed Smith descendants

If they have a tree, even a tiny one, build on it until you can find the connection to the branch you know they are likely from (focus on lines that come from the same/nearby location). If you can't find a common ancestor, that's okay, leave them in that group and you can come back to them another time.

Step 4: Keep doing this, ideally for all your estimated 4th cousins and closer (20+ cM). That's a lot, I know (I currently have 1,048 matches that share 20+ cM with me). It takes time, it's a lot of work, but in the end you'll wind up with 3 types of matches: those with identified common ancestors, those who likely come from an identified branch, and those you have no clue how you're related, not even a potential branch.

What to do with these groups? 

This is where there will be some overlap in my instructions on finding an unknown biological ancestor. Look at the closest match that you haven't even been able to group into a certain likely branch (or a common ancestor). Even if they don't have a tree, that's okay - look at your Shared Matches with them and open any match that has a viable tree. Compare the trees - do any of them share an ancestor with each other that you don't recognize? If so, research that ancestor and build a tree for them, you may find it links up with yours somehow, maybe even by breaking down a brick wall, or that it leads to an NPE - when someone's parent(s) is/are not their biological parent(s).

Additionally, you can look at your closest match that you haven't identified a common ancestor with, but you have grouped them into a likely branch. If they have a tree, again, build on it, and keep researching until you can find a connection. See my case example of Emma Elizabeth Sherwood.

This method of grouping your matches to single out the ones you can't identify at all can help lead you to some enlightening revelations, but they tend to be rather random. You don't know what you're going to find, you don't know which brick wall it might break down. Even the matches you can group into a likely branch but you're still searching for the common ancestor might surprise you - in my example of Emma Elizabeth Sherwood (above), I knew the match was related to my Mills branch (Emma's husband), but I had no idea it would finally break down the Sherwood brick wall that had been blocking me for 12 years.

Other methods.

There's other methods of breaking down a brick wall with DNA, ones that are more targeted for a specific brick wall, but they heavily rely on the surname you're looking for not being a very common one. You basically just search your matches trees for the surname you're looking for, and then compare the trees of the matches in the results, looking for a common ancestor among them. It can work well when the name isn't common, because it's likely most of the matches in the results will be the ones you're looking for. But the more common the name is, the more matches there will be in the results that aren't related to the branch you're looking for. That's why this never worked with Emma Elizabeth Sherwood (in my above example), because Sherwood was too common of a surname, I only found her family by using the more random grouping method and not knowing where an unknown match would lead me.

The surname search method would be much more effective if AncestryDNA would offer a very simple feature: the option to search for a surname within a specific location. At the moment, you can search for a surname or location, but not a surname in a location. So you can search for Smith OR Christian County, Kentucky, and you can search for them both at the same time, but it will include results for match's trees that have either the surname Smith, OR the location Christian County, Kentucky. And even if the tree includes both, it's not necessarily for the same branch or ancestor, it might be their Jones branch that's from Christian County, Kentucky, while their Smith branch is from Pennsylvania. For common surnames, we need a way to narrow it down, and the best way to do that is by looking for surnames within a specific location. At the moment, we can only do that manually by searching for a surname, and going through each match in the results to see for ourselves if that branch is from the right location. If so, then we can look for a specific common ancestor. It's very time consuming, and the more common the surname is, the less realistic it is to go through all those matches manually, yet there's a very simple way to make it easier, if AncestryDNA would just listen to their customers.

The surname search works a lot better if it's not a common surname. I successfully used this method with the surname Deaves, and also a suspected maiden name of Brannin.

You can also search by just location, but this only really works if your ancestors are from a very small, unique town, especially where there's endogamy. In my above example about my 2nd great grandparents who came from a tiny Italian town called Monteroduni, it's safe to say that the town is so small and endogamous that anyone who has ancestry from Monteroduni is probably related. Certainly, any DNA match of mine that has ancestry from Monteroduni, it's safe to say that's very probably how we are related. So I can very easily search my matches trees for the location of Monteroduni and even if I can't find a common ancestor between us, most likely that's probably where our common ancestors were from. Brick walls are difficult with endogamy though, so that might be the most I'll ever be able to determine. Searching by location may not break down any brick walls in your tree, but it does help you identify and sort your matches into groups/branches, which can help you find other unknown matches that may lead to a brick wall.

Like I say, sometimes breaking down a brick wall with DNA can be unique to the situation. Sometimes you have to think about what you're looking for, and consider the best way to come at the problem. But this should give you the basics to get you started. Feel free to share your success stories!

Tuesday, September 7, 2021

ThruLines is not the enemy

I see a lot of skepticism out there about ThruLines, and some of it is warranted, because it is based on family trees, which can have errors that get copied multiple times. But that doesn't mean you should dismiss ThruLines entirely, there are ways to get reliable use out of it, and not just by finding records that confirm them. There are ways to use DNA to find biological relatives or break down brick walls in your tree even when there's no written records of the lineage, and ThruLines is just one tool that can help you do this.


It's basically a matter of probabilities. The more people you match who are descended from multiple siblings of your ancestor, especially when all those descendants all or mostly match each other to form a cluster, the less likely it becomes that it's an error. When the matches mostly all match each other to form a cluster, you know they are all related and descended from the same branch/ancestor - you just need to identify which branch/ancestor, which is where trees and ThruLines come in. Each sibling that those matches descend from would have to be an error for trees/ThruLines to be wrong, so the more siblings you match descendants of, the more likely the trees are accurate. If you match 20 people (who mostly all match each other too) descended from 5 siblings of your ancestor, what are the chances there's been an error in the trees for each of those 5 siblings, plus your own ancestor? Extremely unlikely. In the example above (click to enlarge), there's 41 matches descended from 8 siblings of Elizabeth Mertz, so for this all to be wrong, there would have to be 9 different errors. This amount of evidence is really very conclusive, and I can probably confirm this family now.

Even assuming there's only one error and those siblings are indeed siblings to each other, but your ancestor is the lone error, and not actually their sibling, what are the chances you would match that many people from a certain family, if you weren't related to that family somehow? Using the example above again, what are the chances I match 41 people descended from those 8 siblings, if Elizabeth Mertz is not one of their siblings? Again, it's very unlikely - and the only way this would be possible is if there was a lot of endogamy involved, but even so, it would still be pointing you towards a specific population you're likely descended from (and matching surnames from the same endogamous population means you're probably related to that specific family somehow), so you don't want to dismiss it.

Granted, it doesn't confirm who exactly the parents of those siblings are, only that they are indeed siblings. For that, you'd have to go up another generation and do the same thing - look for people descended from siblings of the alleged father and mother. In the example above, it doesn't really confirm that Phillip Mertz is the father of Elizabeth and all her siblings, only that they are siblings from the same parent(s), whoever that may be. But for now, it's probably safe to add Phillip Mertz at least as a placeholder until more research can be done (it really is okay to add speculative data to your tree as long as you know it's speculative!).

In the example below, you can see how this ThruLines doesn't confirm descent from Benjamin Butler - the 6 DNA matches are descendants of children of David Butler, so this really doesn't confirm this potential ancestor at all.

And there's other limitations, mainly the fact that the Shared Matches tool (which is the only way to confirm if matches match each other and form a cluster) only includes estimated 4th cousins or closer (20+ cM). AncestryDNA really need to provide something more comprehensive. They say it's limited to 20+ cM because it would tax the server too much if they expanded it to include all matches. But at the very least, they could expand it to 15+ cM segments, which have a 100% chance of being identical by descent. That would still exclude most matches (8-15 cM) and therefore not be as taxing on the server, but include all matches that have a 100% chance of being IBD, which would make ThruLines so much more useful and reliable. At the moment, they are excluding hundreds, even thousands of IBD matches from the Shared Matches tool, which is extremely debilitating. Alternately, they could offer another tool that would be less taxing on the server - a simple one-to-one comparison. Pop in two match usernames, which would tell us whether those two matches match each other or not. Very simple, not very taxing, but it would get the job done.

Even so, it's still possible to get reliable usage out of ThruLines. Remember, ThruLines is only automating a process that people used to manually do (and still do when the relationship exceeds ThruLines' 5th great grandparent limit). If it weren't possible to use DNA to confirm relationships when there is no written record of it available, what use would DNA be, and how do you think all these NPEs are being discovered? While it's true that you do have to watch out for tree errors being replicated in ThruLines, if you understand how DNA and ThruLines work, there is useful data you can get out of it. To often, I see people who seem to completely dismiss ThruLines, as though it's not reliable at all, but you're only hindering your own research by thinking that.