RLP 130: How to Evaluate an Ancestry Thrulines Hypothesis

Nicole (1s):
This is Research Like a Pro episode 130. How to Evaluate a Thrulines Hypothesis. Welcome to Welcome to Research Like a Pro a Genealogy Podcast about taking your research to the next level, hosted by Nicole Dyer and Diana Elder accredited genealogy professional. Diana and Nicole are the mother-daughter team at FamilyLocket.com and the creators of the Amazon bestselling book, The Research Like a Pro a Genealogists Guide. I’m Nicole co-host of the podcast join Diana and me as we discuss how to stay organized, make progress in our research and solve difficult cases. Let’s go, Hi everyone, and welcome to the show.

Nicole (46s):
I’m Nicole Dyer cohost, and I’m here with accredited genealogist and my mother Diana Elder. Hi, Diana. Welcome to 2021. It’s a new year.

Diana (55s):
Hi, Nicole. So exciting to have a new year, and we are all hopeful that this is going to be a much better year than 2020, right?

Nicole (1m 4s):
Absolutely.

Diana (1m 5s):
2020 was not what we expected, but there were some silver linings that gave us opportunities to maybe do some things we hadn’t planned on doing it’s fun though, because we’ve been working on our Research Like a Pro with DNA book, and we’re almost done with that and getting ready to get to the final stages before publishing. So that’s exciting.

Nicole (1m 26s):
Yes, we can’t wait. I’ve been working on kind of revising my Airtable template for a DNA research log. It seems like every time I use that, I come up with a new way to simplify it. And this last time we were using it with the study group, there were a lot of tables we were using and a lot of linking between tables. And I just realized, you know, we don’t need a separate table for test takers and for DNA matches, we just need one table for all the people in the project. And some of them will be assigned the role of test taker, you know, if you manage their kit, and then the others are matches. And then sometimes we might ask a match to share, share the results with this so we can change their role to test taker and somebody whose kit we manage right there in the people table.

Nicole (2m 10s):
So

Diana (2m 10s):
I like that change. Yeah. I agree that sometimes we just have to work with these tools and then we find better ways to do them. Well, we are excited because we’re getting ready to start a new study group and this will be just using traditional or documentary research, this is not our DNA study group, and if you’re interested in learning more about that, make sure you join our special study group email list. And we also are accepting applications for mentors. So if you have done Research Like a Pro, either on your own or in a past study group, or have done the e-course and you would love to be in charge of a small group of people that are also taking the study group, be sure you apply to be a mentor.

Diana (2m 58s):
You get a complimentary registration, and really you are just leading a discussion. You don’t have to be the expert because you bring all questions back to Nicole and to me, and we answer those in the larger session. You’re just there as kind of leading the small group, and it’s really a great experience. All those who have done it in the past have loved it. So we invite anyone who’s interested to check out the information on that on the website. And then if you are looking to get our coupons for any of our products, make sure you join our newsletter, where we send those out periodically. This month we will be launching the Research Like a Pro with DNA ecourse.

Diana (3m 40s):
So everyone whose been wanting to take the course, but couldn’t commit to the study group in the fall when we have that, now you have the opportunity to take the course and get all the same lectures and the materials. So we’re excited to make that available for you very, very soon.

Nicole (3m 56s):
Well, today we have a listener spotlight that we received in a Facebook message from one of our listeners. And he said, recently, I discovered your podcast on quarantine as I gave up the gym for long walks, I really enjoy your topics. While I consider myself an expert non-certified genealogist with 20 years experience and a passion for DNA analysis, you’ve given me new ideas, strategies, and websites to further my research, I would like to request more podcasts, show topics on DNA analysis, especially on more advanced topics in the area. Thank you for your work on the podcast. So we are going to be talking about one of those topics today, which is evaluation of a Thruline hypothesis.

Nicole (4m 37s):
So thank you for the suggestion. And we’ll certainly be adding more DNA topics like that in the future.

Diana (4m 44s):
So let’s talk about Thrulines. You probably have all sorts of interesting DNA Thrulines, the Ancestry DNA, and you might be wondering how reliable it is or if it’s even possible, and so we’re going to give you lots of ideas and some examples today. First of all, let’s talk about how it works. Thrulines is based on an algorithm that compares your DNA matches, their trees, and all the searchable trees in the Ancestry public member tree database. So if the algorithm can make your tree and your DNA matches tree connect somewhere, that hypothesis will show up in Thrulines.

Diana (5m 29s):
And of course, if anything is based on trees, we’re going to have problems. And this is one of the main problems, because a lot of the public member trees have errors. If just one person puts in an incorrect relationship, that’s often copied by multiple other people in their trees. And because Thrulines is a computer algorithm, not a genealogist, it does not analyze those relationships and sometimes merges identities that should not be merged. And the other problem is that we have thousands of matches in the ancestry DNA database, and some of those are likely false.

Diana (6m 10s):
Blaine Bettinger has a great blog post all about this, it’s titled the Danger of Distant Matches and he published it on his blog, the Genetic Genealogist on 6, January, 2017, that has been four years ago that he published that. But I think a lot of us still do not realize that some of these small matches could be false. And because Thrulines pulls up a lot of those really small matches under 15 centimorgans, we have to be really careful with using those as evidence in our genealogy.

Nicole (6m 46s):
So true. So I recently came across a Thrulines estimate that I really wanted to check to see if it was accurate or not. And it was for our cousin who had shared her results with me. And so I wasn’t very familiar with the other branches of her tree that we didn’t share. And when I looked at this Thruline, it was taking a match who I had thought was related on the line that we share, and the Thruline was saying that this match was in a different side of her tree. And so I was worried that I had done the DNA analysis wrong. So I wanted to check this Thruline to see if it was accurate or if it was just kind of a messed up one. So I was pretty worried at first, because previous to this, the only experience I had had with Thrulines was that they were either empty and didn’t have a lot of evidence, so they weren’t very useful or they were accurate.

Nicole (7m 34s):
And that’s kind of how it would look to me and the people whose DNA I had used my own DNA and my grandparents’ DNA. So after doing that, I came up with a little checklist of criteria to determine if a Thrulines estimate is reliable or not. So what it came up with is that if it’s a reliable Thrulines, then most of the matches will share over 15 centimorgans. And then if most of the matches are under 15 centimorgans, many could be false matches sharing pseudo segments with you instead of an actual segment. And of course Ancestry doesn’t show you matching segments, but if they are small, you know, under 15 and especially segments under a 10, those could be a result of the maternal and paternal chromosome and weaving back and forth between those.

Nicole (8m 19s):
That can be a problem. Another criteria to look into is if the matches are grouped to the correct parent, and this is only possible if you have one of your parents tested. So for this particular cousin who I was looking at, her mom had tested as well, but I didn’t have access to her mom’s results. I only had hers, but if you do have one of your parents tested, then Ancestry puts a grouping on your matches over 20 centimorgans that says mother’s side. Or if you had your father tested father’s side, if you just had one parent, it will only tell you the one that tested. So it’ll say mother’s side, and then on the matches that are over 20 centimorgans, that don’t have the mother’s side grouping.

Nicole (9m 2s):
you can infer that they’re on the father’s side. So that’s a really useful thing, but a lot of you won’t have that. So you’ll have to use the other criteria.

Diana (9m 11s):
Another thing that you can do to see if it’s reliable is to see if those matches that are coming down through Thrulines form a genetic network. And so what I like to do is to click on each one of them and then see their shared matches. And you also will want to be seeing that there are other people in that network that match each other and form what we call these genetic networks or clusters. So if you know your family tree, well, you know, your cousins and you’ve already identified some of your matches, your closer matches, you should be seeing them in those shared matchless as well. And if you’re not, then that could be a huge red flag.

Diana (9m 53s):
Another thing is that you want to always pair your DNA analysis with the documentary evidence. And so the names, the dates, the places for the known ancestor and the potential ancestor or potential siblings, all look feasible. You know, you shouldn’t have something really glaringly wrong. So you want to take a look at all of those things, make sure that things feel right and look right and look possible.

Nicole (10m 22s):
Some red flags to watch out for are a common surname. So if the Thruline is for somebody with the surname Smith or Jones or any of common surnames, there’s an increased risk for erroneous merges of identities within the Thrulines. And so the Thrulines algorithm looks at a bunch of different individuals and trees, and they have the same name and maybe lived around the same time they could get merged, especially if there isn’t very much information about that person in their profile. Another red flag to watch out for is if the potential ancestor lived before 1800, there’s an increased risk of errors in those user submitted trees. And also there’s more possible descendants of that person to find them on your thousands of matches.

Nicole (11m 8s):
So if you take a person who had thousands of descendants who lived far back in time, you could probably find some of your matches within the database who can trace their tree back to that person, whether or not you’re actually related to that match through that ancestor and that’s the common ancestor is another story,

Diana (11m 25s):
Right. I think we can all picture family trees, get further out and you get so many ancestors. It’s something that I have found that it’s common is that you are related to that person, but it’s on a different branch of the family tree. So the DNA might be accurate and connecting you, but Thrulines is connected you through the wrong ancestor. And that’s a big problem.

Nicole (11m 47s):
Yeah. Another obvious problem that you can see in a Thrulines estimate is when they give you a potential ancestor and the only child of that ancestor is your known ancestor. And so all the DNA matches descend through your known ancestor. So that’s kind of silly because it doesn’t actually give any evidence of that new potential ancestor. It just provides evidence for that one generation down that you already know. So that’s the example that I wanted to share first was one of my Thrulines and this is on my dad’s side, to our ancestor, Martha Sylvias. So when I clicked on this Thruline to Martha, it was the type that is just my known ancestor, Jessie Estelle Ross, and then her mother is Martha M Sylvias.

Nicole (12m 32s):
And I already have that in my tree, but there are no DNA matches descending from siblings of Jessie Estelle Ross, through other children of Martha Sylvias. So it’s not that useful of a Thruline, right? Because it’s the same Thruline as the one for Jesse Ross, and there’s no siblings to Jesse Ross, so it doesn’t provide any extra evidence of Martha being the mother of Jesse. Although I have that in my tree and I have documentary evidence, there just aren’t any DNA matches who have descended from Martha’s other children. So when I saw that and I kind of looked at my grandfather’s matches, I noticed that there really aren’t any descendants from Martha M Sylvias, but if we go back a generation to her parents, Jesse Sylvia, he does have a Thruline and there are some matches descending from Martha’s brother Oscar Silvias.

Nicole (13m 22s):
So we have some matches descending from the Sylvias side, but just not through Martha. So I thought that was interesting. And I think that usually indicates that there’s not very many descendants or that not many descendants have taken the Ancestry test. So I could go and find somebody from that descendancy family to test, but there’s none there for now.

Diana (13m 42s):
I think that’s such an interesting example. And we do see that, especially when we’re working, not only in our own lines, but in our client work, sometimes we’ll just see a cluster of people and they’re all within the same family. So you can tell that somebody took a DNA test and it said, Hey, you should take a DNA test. And, you know, they got everybody around them and their family to test that you have all these people with the sending from this one individual that have tested, and then you have nothing from others. And we really do have to rely just on the vagaries of testing and who out there has tested, unless we do some targeted testing. Like you suggested. So just because you’re not getting anything doesn’t mean it’s not accurate. Like you said, with the Sylvias connection, you have documentary evidence.

Diana (14m 25s):
So you didn’t feel like this discounted that, right? It just, you felt like there were not enough people who had tested

Nicole (14m 32s):
First. I was wondering if it was a missattributed parentage or adoption, but then when I found matches to one generation further back, I just realized that some of Martha’s children died young or didn’t have any descendants. And the other descendants hadn’t tested

Diana (14m 46s):
Really good example there. Let’s give another, this one is for my Cynthia Dillard and I’ve talked a lot about her on the podcast. I recently discounted that her father was the George w Dillard that I had spent so much time researching, but we found family Bible pages that said, no, she’s not part of that family. But then there was a Thrulines hypothesis of a completely different man, and his name was Hopson Milner. So not even the same surname, this popped up in the Thrulines of some of the other tests that I have access to, my second cousins, and I decided it was time to just do a project and figure out who is this Hopson Milner?

Diana (15m 27s):
Why is he coming through as Cynthia Dillard’s father? And I want either to prove it or disprove it. Now, when I took a first glance at the family, you know, he was born appropriate time period, 1794 in Virginia died 1872 in Harris county, Georgia, just over the border from where Cynthia was living. So time and place fit. Okay. And he did have a lot of children and Cynthia could have fit into the family. So I decided I just needed to look at the DNA. And especially at these Thrulines while I had my two second cousins and my own Thrulines that I used. And when I looked at them, just like Nicole said, all the matches really came through Cynthia.

Diana (16m 10s):
So they were proving, Cynthia was my ancestor and the ancestor of my second cousins. But for my Thrulines, there were absolutely no siblings of Cynthia, supposedly Milner, that came through that for my two second cousins, they did have a couple of siblings that had some DNA coming through them. And so I’d like to those one was for little Barry Milner, who was supposedly the brother of Cynthia. And then there were some matches that came through Nancy Milner, who were the siblings of Cynthia. And if these were true siblings and these were true matches, then that would point that Hobson Milner really was her father.

Diana (16m 51s):
So I evaluated each one of those. And I found that the majority of them actually had a common ancestor with whichever second cousin it was, that was different. And so what I did was I clicked on the DNA match and then I clicked on their shared matches, and among those shared matches, several of them would have the little thing, common ancestor pop-up or they’d have a family tree and I was able to see that the genetic network of that DNA match and my second cousin was not through the Dillard’s or Milner’s, it was a completely different line. I was able to disprove every single one of those.

Diana (17m 32s):
One of them was a false positive. I think it was only eight centimorgans and it only had three other shared matches. And none of them were to any of the known descendants of Cynthia. So I decided that one was likely a false positive. The other thing that I had was the name of a DNA match, who was two generations closer than us on the family tree to Cynthia. And he would have inherited more DNA and likely matched all of those people as well, that he didn’t, he wasn’t in any of the shared match lists. And so that actually helped me also to say, you know what? These are in different networks completely.

Diana (18m 13s):
They are not in Cynthia Dillard’s network of her birth family. So after I discounted all of that, then I went to work on the documentary evidence and I researched the family of Hopson Milner. And I found the record that likely was the source of all of this confusion. There was a marriage in Harris county, Georgia, which is right there on the border of Alabama and it was a Cyntha Milner, married to a William E Milner. And I think what happened was someone years ago, who didn’t have access to as many records as we have, decided that this Cyntha was the spouse of Thomas B Royston and either married Thomas B Royston after this marriage, I don’t know.

Diana (18m 57s):
I think whoever it was got them confused, but I was able to find this Cynthia Milner and her husband, William Milner, they were living just a county away from my Thomas and Cynthia Royston in both the 1850s and 1860 census. So it just showed that two identities had been merged, that of Cynthia Milner and Cynthia Dillard. And so this Hopson Milner was definitely not the father. DNA did not show it and the documentary evidence to not show him.

Nicole (19m 26s):
That’s so interesting. It always comes back to these cases of identities being merged, I feel like.

Diana (19m 33s):
It does. Yeah. Same first name, really similar locations, very similar ages, Cynthia and Cynthia. Right. They have to be the same person, but they weren’t.

Nicole (19m 44s):
Yeah. It’s interesting that you could find that each of the matches were in different genetic networks and didn’t match each other and were in a network with a different common ancestor. So that’s really helpful to be able to see that. Okay.

Diana (19m 57s):
Yeah, it was nice to have a methodology for it. I think sometimes we look at our Thrulines, we just don’t know what to do with it. So it was really helpful having done it as a project and really taking each one and analyzing it separately, that was very helpful. And then writing up a report to prove it, because now I have this report and I’m going to send it to all the people that have this incorrectly in their family trace so we can get that cleared up, hopefully. Yeah.

Nicole (20m 24s):
And that’s how you can get the Thruline to go away. Yup. If everyone removed that weird thing from their tree.

Diana (20m 31s):
Yes.

Nicole (20m 32s):
All right. So here’s another example. So back to my cousin that I mentioned at the beginning, I’ll call her Diane, Deanna, not Diana that’s you. So Deanna had another Thruline that was for a potential ancestor to evaluate and his, the ancestor was named Isaac Newton Wilson. So I thought I would evaluate this one, using my new checklist that I made and see what I could find out if this was a good guess or not. The first criteria that I put on to this Thruline to see if it passed the test was if there were any siblings to the known ancestor, the known ancestor of Deanna was Arthesa Wilson. And supposedly Arthesa was a daughter of Isaac Newton Wilson.

Nicole (21m 14s):
And in the Thruline, there were several possible siblings to evaluate that would be siblings of Arthesa with DNA matches descending from them. So it passed the first test that there were actually other siblings there with DNA matches coming down that would provide evidence of that other being Isaac Wilson. So the next thing I wanted to check was if most of the matches shared more than 15 centimorgans. So I found, that all of the matches except one were over 15 centimorgans. So only one of them was less than it was 13. And so then I decided to take some of those matches and check them and see if they were paternal matches because this was on Deanna’s paternal side.

Nicole (21m 56s):
It was supposedly through her father whose grandmother was, Arthesa Wilson. So since Deanna’s mother had tested, I was looking to see if these matches did not have the mother’s side indicator on them, because if they said mother’s side, that would make the Thruline probably wrong or that they matched on both mother’s and father’s side, I guess. So when I clicked on all the matches, I saw that they did not have any group assigned. So right under the name, you noticing a match of Deanna and Cat predicted relationship, fourth to sixth cousins sharing 22 centimorgans across one segment, and right under that at the top of this Ancestry match page, it says either mother’s side or no group assigned because her father hasn’t tested.

Nicole (22m 37s):
So it did say no group assigned on all of these matches. So then it passed another test. So since it didn’t say mother side, we can infer it was on the father’s side, just like the Thruline was showing. And the next thing I wanted to test was if they were all in the same genetic network to each other, like if Cat and the other matches matched each other, as well as Deanna, I had already made a network graph using Gephi for Deanna. So I just pulled up the network graph and searched through it for these different people. And I found that Cat was in this coral colored cluster kind of toward the bottom of the graph. And then I put in the other names and found that Joe and Jen and now, and all these other people were in the same network, which was wonderful to see that they were all matching each other and other people who likely descended from this Wilson side of the family.

Nicole (23m 22s):
So my conclusion for this hypothesis was that it is a good Thruline hypothesis. That Isaac Wilson is a good candidate for the father of Deanna’s great-grandmother Arthesa Wilson. And further research could prove that it’s the case

Diana (23m 36s):
That’s so neat when it kind of comes together. Now, I think we probably should mention the fact that Ancestry DNA does not assign the mother or the father under 20 centimorgans. Is that correct?

Nicole (23m 50s):
Yes. Thanks for bringing that up. And that’ll be important in my next example, too, because when I first evaluated this next example, I didn’t know that I just figured the Ancestry knows if everyone in your match list matches your mom or dad. So why wouldn’t they just tell you, but I was wrong. They don’t tell you for matches under 20 centimorgans, which is the same cutoff for shared matching. So if you’re expecting to see a mother’s side designation on one of your matches, and you’re not seeing it, you can’t infer that that’s the father’s side or a false match like I was doing, because they’re just not telling you you’d have to go access your mom’s matches and look at the profile page of that DNA match, and then use your dropdown list at the top and say, compare this person to my mom’s kit.

Nicole (24m 38s):
And then it will tell you if they match. So that’s kind of frustrating, but it’s good to know.

Diana (24m 43s):
It is good to know because otherwise you could go off on the wrong path of research.

Nicole (24m 48s):
Yes. So important to really understand what the company is reporting. It reminds me of FamilyTreeDNA reporting segments under seven centimorgans and thinking, oh, this match it’s so close. But then when you take out all the seven centimorgans segments and lower, then they really only share like 20 centimorgans on one segment so they’re much more distant match than they appear. Right. All right. Well, let’s do our last example here. This is the example from Deanna’s Thrulines that really had me worried because I thought, oh no, if this is the common ancestor for Deanna in this match, then my whole brief argument is really thrown off. So I really wanted to examine this and figure out if it held any weight. So this was a Thruline for Deanna on her mother’s side to an ancestor named Jeremiah Green.

Nicole (25m 34s):
So what do you think, is Green and kind of a common surname or not?

Diana (25m 38s):
I would say so.

Nicole (25m 39s):
I think it is. And then the other thing that I noticed is that the daughter of Jeremiah Green, who supposedly Deanna came through was Martha Green, who married a Harris. And so some of her children were supposedly named Harris. Was Harris a common name?

Diana (25m 57s):
We know that one because we have a Harris it’s so common and so difficult to read.

Nicole (26m 2s):
Yeah. So now looking at this, that’s a red flag to me that these are kind of common names that could easily be merged at first glance, though, it did seem to pass one of my criteria, which is that there are several siblings to the known ancestor. Jeremiah Green is showing a bunch of different children who matches come through to DNA matches for DNA matches 2, 6, 4 from all these different children of Jeremiah Green. So you can see why I was worried. So then my next criteria to look at was the number of centimorgans these people share. So in all the cases, this is like six generations back from the DNA match to the common ancestors, like seven back to Jeremiah Green.

Nicole (26m 43s):
So it would make that they would have small segments, but you would hope that some of them would be a little more than 15 centimorgans, but no, most of them were under 15 centimorgans. So that’s a red flag, but it’s not necessarily indicating that it’s wrong. So I needed to do a little more analysis. So the next thing I looked at was Deanna’s line back to this potential ancestor, Jeremiah Green. And actually this was on a line that Deanna did not have flushed out very far in her tree. She just had her great grandmother documented her great-grandmother Francis Lily Jones didn’t have a parent that looked like it was proven. It was just a name and a birthdate.

Nicole (27m 24s):
So Francis Lee Jones had a mother named Lucinda Harris, but the parent child link wasn’t really proven in Deanna’s tree. And I had never looked at this before. So I decided to go ahead and researched her and figured out who this parent really was, and it wasn’t Lucinda Harris, I found a different parent. The mother’s name was actually Lucy Jones. And when I did research on this, it seems like Francis Lee Jones was given her mother’s surname because she was either illegitimate or something else was going on. But on her death certificate, it showed that the name of her father was unknown, legitimate, and it was spelled wrong. So Francis Lee Jones’s mother was incorrect in Deanna’s tree.

Nicole (28m 4s):
So once I figured that out, then I could just probably say, oh, that Thruline is totally wrong, but there was some room for doubt. Maybe I got it wrong and maybe Lucinda Harris was right. And there was something I was missing. So looked to know Lucinda Harris and her family and the Thrulines and I opened the side panel of Thrulines to find where the evidence came that Lucinda Harris was the mother and that her father was this John Early Harris, who was the son of Martha Maple Green, who was the daughter of Jeremiah Green. And as I looked at those intervening generations back to Jeremiah, I found that John Early Harris, his identity had been merged with another person with the same name, John Harris, which was like the most common name you could think of.

Nicole (28m 46s):
And somebody had found a, find a great Memorial for a man named John Early Harris and applied it to their ancestor named John Harris. So there were like two people with that same birth date and death date, and the Thrulines algorithm had just merged them all into this one person. And when you open up the evidence for it on the side panel, there’s two different family trees, one showing John Early Harris with different children in different parents and another John Early Harris with another mom and different children. So you can clearly see there’s two different men living in different places with different families. And they just were merged into one. So the one that even was the Lucinda’s father, he wasn’t a son of Martha Green, so all these links were wrong.

Nicole (29m 30s):
Every single one from Francis Lily John’s to Lucinda Harris, to John Early Harris to Martha Green were all wrong. So once I figured that out, I was pretty sure that the documentary evidence was showing that this was inaccurate, but there were like 30 DNA matches in this Jeremiah Green Thruline what were they all doing there in the Thruline? And I really wanted to figure that out. So after disproving the documentary research, I wanted to examine each DNA match and really figure out why this happens. So like I said, I knew that Deanna’s mother had tested, and so I just clicked on all the matches to try to see if they had the mother’s side or father’s side designation. And since most of them were under 20 centimorgans, it just said no group assigned because ancestry doesn’t assign a group unless it over 20.

Nicole (30m 14s):
But I didn’t know that at first. So I was like, oh, these are all false matches. So some of them, I think were false matches, but others I think were just not showing the mother’s side designation. So one of the matches was nine centimorgans and it said no group assigned. And then on the shared match list, it said two of them shared matches were on the mother’s side and three were not. So I’m guessing that could be a false match because if it really was a mother’s side match, you would expect to see most of the shared matches also being on the mother’s side. Then there was a couple other matches who are all looking like they were on the mother’s side, according to the shared matches, but then they didn’t have the mother’s side designation at the top.

Nicole (30m 55s):
So that was an interesting one. And this matched shared 22 centimorgans. So it was above the threshold, so it should have been a match with Deanna’s mom, should have been showing that it was mother’s side, but it wasn’t. So that one, I had a really big suspicion that it was a false match. So that test didn’t pass. Then I decided to check my network graph. And this was very telling when I looked up all these matches from the Thruline who were over 15 centimorgans because my network graph only went from 14 centimorgans to 200 centimorgans. I found the location of all of the matches over 14 centimorgans. I found those in the network graph and every single one was in a different cluster. And several of them were kind of in the middle of the network graph or a lot of different clusters overlapped where I think maybe those were some false positive matches.

Nicole (31m 45s):
The one match that I was really worried about that shared 106 centimorgans that was showing up in this Thruline, he was in a different cluster than all the other ones. So it made me feel more confident that the match I had been using for my proof argument was accurate. The common ancestor that I found was accurate and that this common ancestor was totally wrong. And that this group of people in the Thruline were just random mixes of other people in the database who have this Jeremiah Green in their tree and the Thrulines just merging all these identities.

Diana (32m 15s):
Wow. That’s a really interesting case study. The other thing I noticed was you had the common name of Jones in there as well. Yes. So you have Jones, Harris and Green, which any one of those would be a problem, but then combined, you can really see how you got so many different identities merged. Well, let’s just kind of recap the red flags. So the potential ancestor was born before 1800, which means they have tons and tons of descendants likely. And most of the matches were under 15 centimorgans. So kind of in that danger zone, as what I like to call it, you only had one match over 30 and only seven matches between 16 and 30 in the restaurant under 15, and the matches didn’t match the correct parent and they weren’t in the network.

Diana (33m 1s):
So I find that so interesting that it was your Gephi network graph that helped you to see the big picture. And I think that’s really helpful. We have all these different tools and we have to realize that Thrulines is just one tool and it can give us a hypothesis. It can really quickly verify something if the centimorgans are high enough. And we can really look at those trees, it can be a great tool, but we have to realize that nothing is for sure in Thrulines. And we have to be our own genetic detectives and tracking down the evidence and making sure it really fits with our conclusions.

Nicole (33m 37s):
Yeah. It just reiterates to me how important genetic networks are and finding that at least some of the people in the Thruline also match each other and they just didn’t none of the people on that Thruline match anyone else in the Thruline. So that was really telling, and you don’t need a network graph to tell you that you can just do the work manually yourself and just kind of look at each person’s shared match list. That can be kind of difficult. Sometimes if the matches are small, because if they are under 20 centimorgans and the people share are under 20 centimorgans, it won’t show up in the shared match list. But hopefully you have some that are over 20 that you could do that check on. And then if you want to make a network graph, you need to get DNA GEDCom and use that program, the subscription with that program to get the client and download your matches and your shared matches from Ancestry.

Nicole (34m 25s):
And then you can use the Gephi program to plot the monograph, which is difficult. And so I really recommend that you contact Shelly Crawford of Connected DNA and purchase a network graph from her. And she does a wonderful job. And that’s how I first got into making network graphs. I had her make me one and tried to learn from that. So I’m really grateful to Shelly Crawford for this wonderful method of looking at networks.

Diana (34m 49s):
I agree. I used one of her network graphs in a client project that solved the case. And one of the things that she gives you is also a spreadsheet. And I was able to search that and find the match I needed based on the surname and the tree make the connections. And so that really is a great tool. And I think it’s well worth the money because once you have it, then you can use it for so many different things in your family tree. It’s a great tool. Well, thank you for this great explanation of Thrulines Nicole. And I’m so glad that you figured out some of these things. I think we all see that come through and we’re not really sure how to use it. And I love that you also wrote a blog post that goes right along with this.

Diana (35m 32s):
So if anyone listening is still a little bit confused, you can go read the whole blog post and can try out the same techniques In your own Thrulines and see what you can find out.

Nicole (35m 41s):
Absolutely. Yeah. Go read the blog post. Sometimes it’s hard to explain my examples just by talking about them. You might want to go see the network graph and see what we’re talking about in the blog post. All right, everyone have a great week and we’ll talk to you again next week. Bye bye everyone. Thank you for listening. We hope that something you heard today will help you make progress in your research. If you want to learn more, purchase our book Research Like a Pro a Genealogist Guide on Amazon.com and other booksellers. You can also register for our Research Like a Pro online course or join our next study group. Learn more at FamilyLocket.com. To share your progress and ask questions join our private Facebook group by sending us your book receipt or joining our eCourse or study group.

Nicole (36m 26s):
If you like what you heard and would like to support this podcast, please subscribe, rate, and review. We hope you’ll start now to Research Like a Pro.