Genetic Genealogy Questions Answered: Live DNA Q&A – SCGS Jamboree 2019
Amber: Hi everyone and welcome to another LegacyTree Live DNA Q&A. We are broadcasting from sunny Burbank, California at the SCGS Jamboree 50th Birthday Bash. My name is Amber Brown I'm the marketing director for Legacy Tree Genealogists, and I'm here with Paul Woodbury, our in-house DNA expert, and he's answering your questions that users submitted on legacytree.com/live and we'll try to get to as many questions as we can in the next 30 minutes. We will also take a limited number of questions live, so if you have them, drop them in the comments below and we'll try to get to as many as we can.
So you ready Paul?
Paul: I'm ready.
Amber: OK, let's do this. So our first question came from Kathleen. Kathleen if you're listening, drop us a like. She says, “I had my DNA tested and received my DNA results, but not sure what to do next to discover my ancestry.” And I think that's such a common issue that users face. “I have an ancestry chart somewhat completed but I need to know how to continue my search in Poland.”
Paul: Awesome. This is a great question. And it's one that we often get, like Amber mentioned. And taking your DNA from just beginning your search and doing your DNA test, to actually incorporating that as part of your family history research is a really important element, a really important next step. To do that, what you want to do is move beyond those ethnicity estimates, and you want to begin looking at your match list. You begin with those that share the highest amount of DNA with you. And you begin to search and try and figure out how you might be related to those people. You can do that by looking at their family trees that might be associated with the DNA test results. You can do that by reaching out to them and requesting more information regarding a family tree if they haven't included that information with their DNA test results. And after that, you begin to ask questions about their family tree and you try and figure out how you are related to them because if they show up in your match list, you're going to share a common ancestor. And based on how much DNA you share, that ancestor could be your grandparents, your great-grandparents, or, if they share a lower amount of DNA with you, then they could be your seventh great grandparent's, eighth great grandparent or very far distant.
So taking that approach, you'll slowly begin to see how these genetic cousins are fitting in to your family tree, and you'll be able to confirm your own proposed genealogy that you've started and you might be able to make some discoveries regarding your family tree.
The way that those discoveries work are, you're going to find genetic cousins
who are related to each other, and who have a known ancestor with each other but who have no known relationship to you. And by analyzing the extended family of that particular common ancestor between your genetic cousins, you can sometimes break through a brick wall in your own family tree by identifying collateral relatives who are among your own ancestors. So it's a little bit different than what we usually do in genealogy, a lot of the times we're starting with what we know and we go to the parents, we go to the grandparents and we're working back.
With genetic genealogy as we incorporate DNA evidence into your research, a lot of the times what you're actually doing is you're identifying those more distant ancestors first, and then tracing back down to see how their descendants fit into your family tree.
Amber: Perfect, so good luck with that Kathleen. If you need help, let us know.
OK this question comes from Minnie. She said, “I did a DNA test last year with the hope of finding out who my biological father is. I've had so many matches, but I don't have a clue how to figure out what's what, I'm at a dead end.”
Paul: OK, and this is somewhere where we would love to help.
At Legacy Tree Genealogists, we have a whole team of genetic researchers who specialize in using DNA evidence for unknown parentage cases, adoption cases, instances where you have an ancestor in your recent family history that is unknown to you.
And the approach that you're going to want to take in this instance is first, you're going to want to organize your match list with your autosomal DNA and you're going to want to figure out which of your genetic cousins are related to you through your known family. Who are the individuals who are related to your mother? Are there individuals in your match list that are related to your mother? And using those individuals, you can begin to filter your match list. Anybody who shares DNA with those known maternal relatives, you can rule out from your analysis, and that will leave you essentially, with a whole pool of genetic cousins, people who share DNA with you who are also related to you through your unknown father.
At that point, you'll begin to break that pool of matches down further, based on their relationships to each other. Start with your closest likely paternal relative. And search for all of the people that share DNA with him or her. And begin to cluster those individuals based on their genetic relationships to each other. Then go to the next unknown paternal relative and search for everyone who matches that individual and build another cluster that way.
Once you've built several of these clusters of people who are genetically related to each other and who are not related to your known maternal relatives, then, you begin to identify how they are related to each other. We know that there's a genetic connection somehow. So you build out their family trees, or you look at the family trees that they've attached to their test results, and you search for patterns. Search for common surnames. Search for common ancestors that they may have in common. You'll search for locations where they were living. And you build out their family trees to see how those people are related to each other.
And what that will do is it will generate a whole list of what I call ancestral candidates, people who are likely among your ancestors, or who had close relatives, like a brother or a first cousin who were among your ancestors.
Once you've identified these ancestral candidates, you can then search for how their descendants interacted, intermarried with each other to trace down through the generations, until you find one person or a set of candidates who could be your father whose genealogy accounts for all of the expected ancestors that you've identified, and who was in the right place at the right time, and of the right description to be your biological father.
Amber: Great advice, and that's something that we deal with a lot so great question.
OK we have a live question from Brian O'Keefe, he says “any hints on how to get matches to reply to messages? It's difficult getting replies from anyone.”
And Brian, we actually get this question frequently as well. And we wrote a blog article all about that. So you can check it out on our blog at legacytree.com/blog, and I'll also drop a link in the comments, but did you want to briefly mention anything else?
Paul: Yeah, I would love to. So a lot of the times, we reach out to our genetic matches and it can kind of feel like using a shotgun. You just try and shoot as much as available out there. And so you send out this message and you blast it to all of your relatives, and you don't get a response. (laughing) And one of the ways that I have found can make collaboration and requests for communication a little bit more effective, is to do a little bit of homework beforehand. Find out as much as you can about that genetic match that you're going to reach out to. Find out about their approximate age. Find out about why they may have tested for genealogy. Try and build their family tree yourself before reaching out to them.
In some instances, you're going to get a match named John Smith, and you're not going to be able to find out who that person is. They don't have anything attached to their match profile, they're not gonna have any clues to help you figure out who that person might be. But a lot of the time there are subtle clues in the user name, in the information that they've provided on their profile, on the attached family tree to help you figure out as much as you can about that person before reaching out.
Ideally, if it's at all possible, you should at least look at who they are sharing DNA with from your match list. If you've identified other people from your match list and you say, “well I don't know exactly how John Smith is related to me, but I know that he is a relative of all of my known genetic cousins from a particular ancestral line that I've researched,” if you can drop that surname, or drop that ancestral line, then that's giving them something to grab onto, to grasp onto and say “oh yeah, I do recognize that surname that you've mentioned to me. I do recognize the Johnson surname from North Carolina. I do recognize a particular fact” that you have shared with them as kind of a teaser to say “do you know what this is?” And that makes it much easier for them to respond “yes I recognize that, I can help.” And they feel much more empowered to be able to help rather than saying “what's your family tree like?”
A lot of people don't even know how to build a family tree or not familiar with their family tree. And so that can be off-putting to some that may not know how best to collaborate with you.
Amber: Great point.
OK so this question came from Jo. She says “since doing DNA, I found my father was not my father, how do I find my biological dad?”
And you already touched on this Paul. We discussed it in an earlier question. But I just wanted to let you know of a page that we have available, it's called legacytree.com/resources and it has a lot of great information for people that do discover these surprises in their DNA and how to handle that situation, and some different things that can assist with that.
Anything that you want to add that you didn't already cover in the previous question?
Paul: No I think, in addition to what we talked about previously, I think we often try and rely solely on DNA and, like Amber said, DNA and the research and actually identifying who your father might be is only half of the equation, or maybe even less than half.
It's an important consideration to explore what your decisions will be regarding relationships in the future, who you're going to share this information with. That's a very personal and emotional journey. And so I really do recommend that resources page.
If you do decide that you are interested in learning more about who your biological father might be, it's important to consider as much information as you can about the context of your birth, and your conception, in order to provide context and help understand who the potential candidates might be. Find out where your mother was working, who her colleagues were, what her social interactions were. A lot of the times, those key details are really important for helping us to narrow down who potential candidates might be in the end.
So don't forget, we've already talked about the DNA approach and finding out, looking at the clusters, and building family trees and finding potential ancestors, but don't forget the context of your birth, of your conception that will help you to identify the most likely candidates once you've narrowed it down to that point.
Amber: Perfect, and that page again is legacytree.com/resources.
And I've dropped a link to that in the comments.
OK, next question comes from William. “Can DNA ethnicity tests actually help with family history research?”
Paul: OK, so I get this question all the time because when I tell people that I work as a genetic genealogist, a lot of the times people immediately jump to those ethnicity tests and they think, well, I know about those ethnicity tests, but how do you get from an ethnicity test to actually identifying somebody's ancestor just based off of oh, you've got 25% German, therefore your ancestor was
(laughing) Wilhelm Schmidt.
Amber: Right?!
Paul: So…It's a little bit of a misunderstanding there. But with ethnicity estimates, yes they can be helpful but only to a certain extent. And I like to think of ethnicity estimates as one of the broadest categories of resource, or record that we actually look at. It's similar in my mind to looking at compiled family trees or looking at a bird's eye view of the entire picture, and you don't get a lot of the detail and you don't get really strong conclusions just based off of ethnicity estimates.
When you move beyond those ethnicity estimates, then we have the power to actually use genetic relationships between known and unknown genetic cousins to better figure out how you fit into the family trees of other individuals or how those individuals fit into your family tree. How you fit into each others' shared family.
And so that is really where the meat and the really important details come in helping us to solve these cases.
Ethnicity can provide some really important guidance or clues to help us in that exploration. It might provide some broad context. But in the end, we're always going to need those genetic cousins to really dig down and get to the hard and fast answers that we're actually searching for.
Amber: Great, thanks Paul.
OK this question comes from Barbara. She says “I have unknown ancestors on the paternal side which I have indicated ‘with unknown grandfather’, and ‘unknown grandmother’, ‘unknown father’, et cetera. ThruLines has connected these unknowns to specific people that I don't find otherwise in my 15,000+ tree, or list of DNA matches. I was wondering how they came to link me to these people. Is this a mistake, or are these genuine matches based on DNA matching?”
Paul: So a little bit of background here. ThruLines is a tool that was recently released by AncestryDNA, and what ThruLines does is it will take your family tree that you have attached to your test results that is either public or that is private but still searchable, and it will take that family tree, even if it's just a few generations, and it will compare it against its database of millions of family trees to try and estimate who the ancestors of the people in your family tree might be. And they do that same that same thing with the other individuals in your match list.
And sometimes you can run into problems. It is a new tool, and they're still working out kinks. It's still a work in progress but it is a fantastic clue. It's a fantastic way to get clues on how you might be related to particular genetic cousins.
Occasionally you will see some strange things happening where it will say that you're related through a particular ancestral couple. It'll say you're related through your mother and through this ancestral couple on your mother's side. Only to discover that when you look at your mother's match list, that person isn't a match to them. So they can't be related through your mother's ancestral line. So there are some things that we're still dealing with with ThruLines, but it is a good clue, or a good clue generator to help you get ideas on how you might be related to people.
As with any type of tool or with any type of compiled record, we always have to go back and look at if it makes sense, if there's additional support to support that proposed relationship or if there's a more likely other scenario.
Amber: Great, and we actually just published an article on our blog that discusses AncestryDNA ThruLines in depth, and gives a great overview of how to use that effectively in your research. I'll drop the link to that in the comments as well, and you can check that out at legacytree.com/blog.
OK, next question, so this is a two part question and it comes from Gary.
So let's do one and then two if that works for you Paul.
So one it says “am I correct that the node of a triangulation on a shared DNA match's page means that I, the person the triangulation is listed for and the two other people involved all share a section of DNA in common and that the other two matches are definitely closely related to me?”
Paul: OK, so there's two parts of this first part too. The first part I want to address is that yes, if you are shown as being triangulated with two other matches on a particular segment, what that indicates is that you share a segment of DNA with the first match who shares that same segment of DNA with the second match. And that second match shares the exact same segment of DNA with you, it's a three way comparison and you're all sharing DNA on the exact same segment of DNA.
Now that being said, the other part of this question was, “if that's the case, does that mean that I'm definitely closely related to these individuals?”
And I suppose the answer to that question would be dependent on how you define closely. And it's going to depend also on how many segments that you triangulate on, how large those segments are, and it's going to depend on the random nature of genetic inheritance.
So, if you are all three triangulating on a particular segment of DNA that is quite large, I'd say probably 20 centimorgans or larger, that's a good indication that you share an ancestor at some point in a recent genealogical timeframe. That person could be your third great grandparents, it could be your second great grandparents, or it could be a little bit further back. Probably no more distant than about eighth great grandparents or ninth great grandparents.
Once we get pretty far back, it's much more unlikely to have some of those triangulated segments. That being said, there are lots of considerations for triangulated segments. You wanna make sure that it's not actually a region where lots of people from the population happen to share DNA. Those are known as pile up regions where lots of people are sharing the same DNA in the same region just because they come from the same population, that it's from a very, very distant ancestor.
You also want to make sure that you don't assign undue meaning to very small triangulated segments. If it's seven centimorgans, eight, nine centimorgans, you could be related within a recent time frame, or it could be a much more distant relationship.
In reality, with triangulation, I end up using triangulation as it appears but I don't actively seek it out most of the time. I tend to focus instead on the clustered relationships. I tend to focus on the total amounts of DNA that I'm sharing with individuals and evaluating that. Just because, as you move further back in time, and as you look at larger groups of matches, the likelihood of triangulation decreases drastically as you expand that pool and as you move further back in time.
Amber: Great OK. So this is part two of Gary's question. He says “if number one is true, then further, that our genetic relationship is fairly recent in time within, say, three to five generations.” Which you kind of already addressed.
Paul: Yeah. But again, just be careful. I have actually seen instances where you can have very large segments and you think I share a single segment that's 40 centimorgans, 50 centimorgans with somebody, we have to have a recent common ancestor. And maybe somebody else is triangulated, and we must share a recent common ancestor. And in exploring the trees, have discovered that in fact, they are fifth cousins, sixth cousins, seventh cousins even sometimes.
And so we can't always assume that just because it's a large segment, just because we're all triangulated, it may be quite far back. Most of the time it is going to be three to five generations or a little bit closer, but it's not always the case.
Amber: Great, thank you.
OK, this question comes from Sheila. She says “I would like to find out about my paternal
grandparents, where do I start?”
Paul: So there are a few approaches that you can take with DNA to find out a little bit more about your paternal grandparents. But, in reality genetic genealogy is just genealogy.
We use genetic evidence as part of research to make meaning and to better understand our interpretation of traditional document evidence. So…As you engage in genetic genealogy to address this question of learning about more of your paternal grandparents, I'd encourage you to simultaneously explore as much as you can regarding the document evidence, or the document trail that's already there. Talk to living relatives who knew your paternal grandparents. Find out information from them. Try and find records that maybe talk about or mention your paternal grandparents and try and build off of that.
If in pursuing that, you still come back empty-handed, or still come back seeking answers to specific questions, you can use DNA to learn a little bit more. You can perform DNA testing for yourself, for other descendants of your paternal grandparents, your first cousins, your aunts and uncles. And in that you want to prioritize testing people who were generationally closer to them. So if your parents are living, test them. If your aunts or uncles are living, test them. And you want to prioritize people who are more distinct from you. So if you have a limited budget, as most of us do, I might recommend testing yourself and a first cousin. That's an option. So try and get more than one person to test. You can use autosomal DNA testing to find out about their potential ethnic origins. You can use autosomal DNA testing to connect with potential DNA cousins through that branch of your family.
Another approach that you can also consider, that's specific to your paternal grandfather, is to search for someone who inherited his Y DNA. And Y DNA is passed from the father to a son to a son, in a direct paternal line of inheritance. So… If you have a brother who is a direct paternal descendant, say your brother or a paternal uncle, or a paternal uncle's son, somebody that has a direct paternal relationship to your paternal grandfather, then you can invite them to perform Y DNA testing. And that could provide some insight into the deep ancestral origins of that direct paternal line. It might provide some insight into the surnames that your family is associated with, if you have questions about where the surname came from or what other surnames it might be associated with. So there's also some opportunities for exploration there with Y DNA testing.
Amber: Perfect, thank you Paul. OK, so this question came from a user. She says “I just found out that I may have a half-sister. If her centimorgan count was 1551, do you think it was true? My father is deceased, so I wouldn't be able to get his DNA, any suggestions?”
Paul: OK. So there are a few things that you'll want to keep in mind with a match this close. So 1551 centimorgans is a 100% probability according to DNA Painter which provides a tool to help you evaluate that amount of shared DNA. According to DNA Painter, there's 100% probability of a half sibling and, or uncle or grandparent relationship. So someone who is a close family member within one of those relational levels. Something you'll want to consider here is are they approximately the same age as you? And are their known parents approximately the same age as your parents? If that's the case, it's more likely that they are a half sibling. If they are about 20 to 35 years older or younger than you, then you might consider the possibility of an aunt, to a nephew or niece. So in your case it might be an aunt to a niece relationship. If they are more than 50 years older than you or younger than you, you might consider, a grandchild, grandparent level of relationship.
If they do come back, and based on their age, and based on their family context, if they appear to be more likely a half sibling, then you're going to move into the next question of, are they a maternal half sibling, or are they a paternal half sibling? And one way to explore that would be to test known maternal or paternal relatives. And in doing that, you may find that by testing a known paternal relative, that they may not show up as a genetic match to you.
And that could indicate that you have an instance of misattributed paternity. If you do test a paternal relative and they show up as a match to you, and you test a known maternal relative and they show up as a match to you, that could indicate that this other individual may have a case of unknown or misattributed parentage, either paternity or maternity. And based on which of those paternal relatives that you've tested or maternal relatives that you've tested, based on who they are matching from your known family tree, you can determine if they are likely a paternal relative, or a maternal relative.
If you don't have the option to test known paternal or maternal relatives, perhaps they aren't agreeing to testing, they're not interested in doing that for you, there is still a way that you might be able to determine how you might be related to this individual.
Let's see, you could invite them to transfer to a third party tool, a third party website like Family Tree DNA or gedmatch.com where you can see if they share an entire X chromosome with you. If they do, then that would indicate a high probability that they could be a paternal relative. Ideally you would also consider maybe doing mitochondrial DNA testing to see if you share the same mitochondrial DNA which comes from your mother. If you both share mitochondrial DNA, you'd be maternal half siblings. If you don't share mitochondrial DNA, you'd be paternal half siblings.
So there are several ways to kind of get around the problem if you can't find relatives who are willing to test for you but that would be the first approach that I would take would be to test known paternal and known maternal relatives, make sure that both of those are matching you. And to figure out which side of the family this potential half sibling is matching you on.
Amber: Perfect, thank you. OK, we're going to have to make
this our very last question. We're running out of time.
This one came from Jill. She says “I am a black woman in Bermuda. I'm trying to figure out how I received one and a half percent Japanese DNA. I have researched Japanese DNA to get more insight as to how this could've showed up in my DNA, but I found that most Japanese have Korean DNA, and that indigenous Japanese only come from Northern Japan. This is such a mystery given that I live in Bermuda and my family tree here goes back to at least the 1700s. I also have Basque, Nigerian, Kenyan and German DNA, which I can account for. But Japanese DNA is a complete mystery.”
Paul: And this is similar to a lot of questions that we get regarding ethnicity, regarding where is this small percentage of this unique DNA coming from?
Something that you may want to consider is testing at several DNA testing companies. While each of the companies are fairly good at figuring out the differences between particular broad ethnic regions, determining exactly what is British DNA versus Scandinavian DNA or particular ethnicities within a larger geographic region can be much more difficult. So I would encourage you to test at several DNA testing companies to see if that Japanese DNA is in fact persistently shown or estimated across several companies.
If it's only showing up at one company, and it's not showing up at the others, then it might be a mistake. It might be some background noise. The ethnicity estimates are going to vary depending on the people that they have in their reference populations, depending on the proprietary algorithms from each testing company on how they analyze your DNA and estimate your ethnicity. It's going to depend on several factors.
As the analysis of this DNA improves, the ethnicity estimates are going to continue to get more representative of reality, of what we would expect, given our biological family tree. If it is a fact that there is Japanese DNA in your family tree, we would have to look at the cultural history of Japanese in the Caribbean. I'm not familiar with that. I'm not going to claim to be an expert on Japanese history in the Caribbean.
Amber: Well, you can't be an expert in everything!
Paul: I can't be an expert in everything but that might be something that we could explore to see if there is something that could help explain why this is occurring in your test results. But before jumping immediately to that conclusion, I would test at several different companies and see if they're consistently coming back with this Japanese ethnicity estimate. It may just be a mistake, it may be a region of the DNA that is highly conserved across many populations but just happens to be most commonly found among Japanese but is also found in Europeans, is also found in Native Americans. It could be a sequence of DNA that is commonly found in many areas but is just found in a majority in Japan. So there's lots of ways to consider that.
Something else to consider in that case, would be to look at the ranges of probability. Often with the ethnicity estimates at each company, they're going to give you a range of probability and it may range from, for 1.5%, it may be zero to 2% probability. And in your case, if it doesn't make sense, you can always err on the side of it being closer to 0%.
Amber: Great, perfect, thank you so much. So we are out of time for this evening. But before you go make sure you like the Legacy Tree Genealogists Facebook Page to receive notifications
of future live DNA Q&As. And also if you want to access this video again, you can share it to your timeline by hitting the share button at the bottom of the video.
Thank you for joining us, and we look forward to seeing you again in the future.
Paul: Bye.
For more information on genetic genealogy visit http://www.legacytree.com/genetic-genealogy-dna,
and be sure to check out our blog articles on genetic genealogy.