 I'm going to sleep one night, but I must take a stare at the valve, something important. Good morning ladies and gentlemen again. It gives me great pleasure to introduce the second speaker of today, Brad Larkin. Now Brad is going to talk to us about the future of genetic genealogy. So we're not going back into the past, we're going into the future, which is kind of a strange experience for genealogists. Brad comes to us from Houston, Texas. He is the editor of the surname DNA Journal, the founder of genetichomeland.com and a DNA project administrator for the Larkin project as well as other projects as well. He's a member of ISOG and a member of the Southern California Genealogical Society. So ladies and gentlemen please give a big warm welcome to Brad Larkin. Thank you, Mars. At the Southern California Conference, Mars and I were having a cool beverage in the lounge and he was telling me about this exciting conference that you have here. And we were tossing around concepts of things that, you know, I might come and give a talk about. Somehow I think Off-Handed Lid said the future of genetic genealogy, not really thinking of it. And he's like, yes, that'd be great. So at least I guess it's a fairly unique presentation. We're not going to cover what percentage of your cousins are related to you in this one. But we are going to talk about how to more precisely find your common ancestors to them. So this is a very futuristic presentation, but I think a lot of it is going to come true in the future. And as I come into the RDS, I noticed how when I parked yesterday, too loud, I'm good. The horse stalls in this building. Here we are talking about great sites in the future, but of course they had horse stalls. And those people, when they put those in, I imagine that was probably the only method of transportation there was. People have been using horses right for millennia, but yet things change. And so that's the way some of this is going to be. There's going to be some big changes that are going to change some paradigms probably. This is kind of the overview that was on the website, what we're going to talk about today. Three big themes. One is that the sampling coverage is to approach 100% of the genetic lineages in Ireland and the British Isles sometime in the near future. So it won't be necessarily 100% of the people necessarily, but all those lineages. Continued expansion of the number of markers commercially feasible to test and the growth of reference databases with phased results integrated with pedigrees and geography. These elements in combination will one day allow an individual's raw test results to be linked to an established genetic lineage in its ancestral geography in a matter of moments. And so this discussion is intended for those who are interested in how science and technology of genetic genealogy will develop as a field. And so I know just from the question I took coming into the room, for many of you you thought that's probably where we were when you took your DNA test. How many folks here have taken a DNA test since you're here at this conference I've imagined most? Okay, thank you. So you probably would have imagined that it was all going to be sorted out when you took your test and you'd have the great results and the definitive. And what you've probably found is you've got some insight, but it's not been nearly as concrete as you might have hoped. So part of this presentation is going to talk about how it will get more concrete in the future. And you're all basically early days pioneers, whether you realized it or not. This is just a graph of some of the YDNA lineages associated with the British Isles. It's very busy. You know, there's a lot of people on it. When this was first produced in 2006, we thought, wow, this is a fantastic job of mapping these lineages. He's got Welsh and Scots and lots of Irish in here. The point is that even though there's a lot of lineages, there are a finite number of lineages that say have been here since the time of 1800, let's just say. Okay, so it's a finite number of lineages that we're trying to test linked to our genealogical pedigrees. It's not infinite. It's not going to go on forever. So someday the sampling coverage of those lineages is going to approach 100%. Especially here in Ireland, the British Isles where we have a lot of sampling done is they've talked about a TDNA's databases, a lot of Americans, but most Americans have ancestry from the British Isles or Germany, both. So in terms of DNA lineage variation, just have a little kind of history slide here. The population of Ireland, 8,000 BC, first humans arrive. Arrival of R1B males in about 2,000 BC. The Viking era about 1,000. Genomic genealogy as a commercial enterprise only started in the year 2000. It's only 14 years old. Okay, so what I wondered coming into this conference today was how close are we today to 100% sampling coverage like I'm envisioning? So it's kind of hard to measure because how do you measure what you haven't measured, you know? So I said, well, I thought about it a while and I thought, one idea of how you might measure it is for everyone that's in the FTDNA Y database that doesn't have any matches at 25 STR markers, meaning they basically have a unique lineage. If you think of that circle plot, that's a unique one where there's nobody matching them. Maybe for everyone that we know of like that, maybe there's one other out there that we just haven't tested. So still, how much would that be? How close are we? So I took some data out of the Y Irish mapping project that I run. Just trying to zoom in on the plot a little bit. So what you can see here, this is a plot by the year of the test kit when people have submitted their DNA sample. So FTDNA started in 2000, but we don't have any kits in there. It started a little bit in 2003, 2004, 2005, 2006 and 2007. So one of the things I noticed here that I didn't realize before was that most of the kits that have come in are really going to come in in the last three and a half years. So that's how new this field is. When you talk about how many matches you have, why don't I have matches, why don't my matches, why can't I tell which cousin or which genealogical line they're in? You have to realize it's still early days. And just to have some notes here, one of the reasons I picked this data set was because it's fairly random self-selection. There's no real sampling bias in it. This group was screened for ancestry from the British Isles. So we're starting to reach critical mass now, though, where most people, like I know in my own Larkin project, used to be every time we tested somebody new, they were a whole new lineage. They didn't match anybody, certainly anybody named Larkin. Nowadays, I don't think I've had a new lineage in two years. They pretty much now start to fit into an established lineage. So it's happened for me in whatever it's been about 10 years. It's going to happen for all your surnames at some point in time, whether it's this year, next year, three years from now in the big picture of things, all these are going to get mapped. Going back to match count here, I'm using YDNA data. This is a histogram of at 25 markers, how many matches people have. Because as I was thinking about people that have no matches maybe represent untested people or proxy for it, how many matches do we have? And I found kind of an interesting distribution. It's kind of bimodal. This column is people that have 1,000 or more Y25 matches. This column 900, 800, 700. What you see is people with no matches that I was talking about using as a proxy, there's about 4%. People that have 100 or fewer from about 1 to 100, it's like 45%. And then 200, 300 some. And then it goes way down here, so it's definitely shaped like this. So these people are from very large lineages and they have many, many matches. But these people have relatively few. So some specific statistics. No 25 marker matches. 13% had 3 or fewer and 27% had more than 500. So the chart is based out of numbers I was able to get out of the Irish mapping DNA project. I also queried Bennett Greenspan, the president of FTDNA, about giving me some insights on his whole database. Everybody he's tested and he gave me these statistics. There's been 161,000 samples with Y25 tests performed that came out in haplogroups INR, which are the predominant haplogroups of the British Isles. And 4% have no match. So that's in the big picture. He's got access to data I don't have. But what that to me suggests that at least 95% of the lineages have been covered in only 14 years of commercial testing. And when you think back that really most of that's been in the last three and a half years, that tells you that I think we're doing a pretty good job of getting to the coverage of the lineages on the Y. So now right now we kind of look at our genetic genealogist as Y-oriented or autosomal-oriented. And it kind of works out that way. The thing about the Y is that it's more commercially established. It's an easier analysis because the Y chromosome, the parts we test, they don't mix with the X. They don't recombine. So it's just a lot easier to map. You know where you got that Y chromosome. With the autosomal DNA that Debbie and people were talking to you about, it's harder to tell because it mixes. But autosomal, we are going to get to the point where we can do this kind of mapping and analysis with autosomal. I know I talked just a little bit to Daniel Krauts last night of the People of the British Isles Project. And they're looking at autosomal SMPs. They're looking at these markers on all the chromosomes. So it will come from both your male and your female ancestors. So it's a big job because there's many times more DNA base pairs. We're in the very early stages. But the good news is there's twice as many samples because men and women have these. And we really need chromosome mapping and phasing to be most genealogically meaningful. And that's where we take individual pieces of your DNA and say, okay, we know you got that from your great-grandmother O'Leary from County Cork. So this piece on chromosome 6 came from your great-grandfather Flynn from County Tipper-Rary. That's what kind of the mapping is and we're going to talk about that. So someday, sampling coverage of autosomal DNA testing is going to approach 100% just as why DNA is starting to do now here in the British Isles. Bennett Greenspan has a phrase, moving from survey to census. Right now we're in survey. We're taking a subset of people and a subset of the DNA. And then we're trying to figure out what that tells us. And we're learning lots. But we're going to move on to being able to say we basically captured all the lineages across all the chromosomes. So I'm going to talk a little bit about phasing. So at the top I said sampling coverage is one of the key milestones for the future. And phasing is another. So as I kind of previewed, to identify the ancestral source of pieces of autosomal DNA, the way we do it is by testing multiple persons with the shared segment and who have an ancestor in common. And here's some good references on phasing with ATHIE. A good article phasing the chromosomes of a family group when one parent is missing. Dr. Tim Jansen, Excel-based phasing program. And there's an excellent ISOG wiki on phasing. So just to highlight how it works at a basic level. With the Y chromosome, we know you got that from your father, so we know what it came from. With phasing, the way it sort of works at a simple level is if you and a Flynn cousin both have the same marker, then you know that came from the Flynn's, right? And you compare that also with you and your Ryan cousin have a marker, a different marker. That came from the Ryan's, right? Phasing can get very complicated and very challenging in terms of writing software to do it. But at a simple, at a basic level, that's what it is. And people do it culturally with simple things like, oh, he's got his grandmother's nose or his father's eyes, right? That's a little form of phasing. So if you get your autosomal results done at Family Tree DNA, they have what they call the chromosome browser. Other companies have things that are kind of similar. We call this kind of a high-level impressionist phasing where we're getting very broad strokes of like this is chromosome one. This is an example with a mother, a son, and a maternal grandmother of the individual tested. And the colors represent the pieces that are in common. So right here, this stretch has got from the maternal grandmother and it's been passed to all three generations. So we know that came from the maternal grandmother, grandmother Flynn say. Here you have an example where the maternal grandmother doesn't really have any. So that must have come from one of the other lines, right? And so you do these kind of comparisons overall and that's how you start to do chromosome browsing and phasing. And it gives me a good reason to put a picture of me and then go on the screen. So in real phasing, you look at individual base pairs of autosomal DNA. And for best results, you want to test all available relatives. Your parents, your siblings, your relatives and your cousins. And what you can do then is you can phase by inference of homozygosity as well as heterozygosity. Basically what it means is DNA bears pairs at any level. It's two markers, one from your mother, one from your father. So if you have say A and T and you test your mother and she's got A, then you must know by inference that the father gave you the T by process of elimination. But if they both have A, then you're not sure you have to look at others. So when Atheid did this excellent presentation and he talks about how to work this out and do it, I'm not going to go into all the details. We're trying to stay feature level, high level. But what happens is when you start doing that phasing and combine it with a good genealogical pedigree, you start to get some powerful examples. This is an example from Dr. Tim Jansen where he's tested a whole bunch of his cousins and so he's able to say, literally, like I said, he's got this person here is an ancestor of Gustav Jansen, Renatha, Elizabeth Lawrence, these folks I think are Mennonites from the United States. But anyway, look at what percentage of his chromosomes he can tell where they came from. He's getting pretty far down the road and that's just right now what he's been able to do today. And he's basically one man doing this on his own because he wants to know. So wanting to know is a key part of it. But in the future what we can see is something called universal human phasing. And that's where as the phasing coverage increases, we link it to genealogical pedigrees. It gets very powerful as we showed here. If every child tested, if every child born were tested for their sequence, we would have the phased results of everybody that's ever lived and had descendants. Okay, so that's a very powerful concept. So if we had the sequences of all the new children and we knew who their parents are, we could start to make a tree from the bottom up that by inference and by process of elimination one time after another time we would know the DNA of everybody that's lived and passed down descendants. Even if we don't know their actual names, we'd be able to have some identity for them. So it seems very far-fetched in the future, right? It seems like I'm talking about cars and everyone's trying to build a horse stall and go to the parade. So some people think it's controversy or not. How could that happen? Of course people that do genetic genealogy, we want to know. So we're typically advocates for that. But there's other reasons why universal human phasing could become a new play. And one of them is healthcare. Okay, one, we know that healthcare costs are a big issue. They're a very big issue in the United States. Imagine so here. You're also having a new generation of medicines and vaccines that are genetically focused, genetically tested to where we know that we don't have to try and do Kentucky Windage and guess if this medicine is going to help you. We can know by your DNA and how the medicine was made that this medicine will do great things for you. Or it won't do anything and you're better off not to take it, right? So we're very early days of that, but that's coming because I do work in the healthcare field and I know a lot of people in the United States are working on these kind of programs. Okay, we also have another trend which is integration with wearable real-time health informatics. For, I have a 24-year-old son and for his birthday, he wanted a watch that monitored his heart rate and his breathing and everything while he ran. That's what he wanted for his birthday. That's what we got him. That's real-time health informatics. So young people want that. They all think they're going to live to be 200 years old and that we're all funny duddies who only have one life expectancy. The young people clearly have the expected two or three, right? I mean, and they feel they're entitled to it. Who's to say? So anyway, that is going to be a big driver when you get the medicine, the health informatics and things like that. There are going to be drivers that are going to want, stimulate people to get, you know, if not all, then significant portions of their DNA sequenced. So it's going to happen in stages. First, we have to get people tested and that's where we're trying to do now. Then we have to start getting people phased as individuals. So Tim Jansen as an individual has gotten himself phased pretty well. And then we need to have phased databases where we integrate multiple people together and that's how we'll start to really tell the genealogical story, okay? I've kind of talked about this. Even if we do not know their names, we'll know the genetic sequences or at least the main S&P markers that identify them as populations. It gives us the power to connect new participants. So then when new participants come in or new people get interested in it, we can immediately know their lineage a lot better because we've tested so many other people. So it may be a case where we have universal testing, but most of the results are, say, turned off or blacked out, anonymized, right? But if an individual wants to start learning about their genetic history, they could say, yes, sign me up, unlock my genetic genealogy results. So you could still have universal testing but have some degree of personal, you know, confidentiality or, you know, choice in terms of what they know. The other thing is that as we do ancient DNA tests, as we're able to recover more remains, more usable DNA out of ancient graves, our insight into human migration by knowing more about ourselves becomes all the more greater. So we're going to change gears a little bit and talk about the technology now, just in terms of sequencing and testing DNA. Moore's Law, how many of you have heard of Moore's Law? Okay, not that many, about half. Moore's Law is a term that comes from computer science and information technology. It's based out of a 1965 article by a man named Gordon Moore who was a co-founder of Intel. It makes the computer chips the biggest one. And I love the title of the article, Cramming More Components onto Integrated Circuits. I mean, nothing could be more high-brow than that title. And kind of his key thing was the complexity for minimum component cost has increased at a rate of roughly a factor of 2 per year. Okay, that's what he said. Normally you hear that as saying computers are going to get, you know, they're going to double in speed every year. That's how you normally hear that. And so that happened really from about 1971 on into 2011. That's what the graph here is. This is the speed of computer processing generically here. That's what the graph is. So now let's turn from, you know, actual basic computers to sequencers, which sequencers have a lot of computer in them. That's for sure. So I decided to do a comparison of the AB370, which was kind of the first machine that was sort of commercially sold in bulk to do sequencing. And I've got some specifications on it here. It weighed 396 pounds. It cost about $2,400 per MB as million base pairs of what it could sequence. And it could do about 500,000 base pairs a day. So I know Spencer Wells is going to talk about Moore's Law today. Two, one thing I've done is I've looked at Moore's Law in terms of the number of base pairs per square inch per day. And these specifications produce about 12 base pairs per square inch of machine per day. The Illumina Highsec 2000 released in the year 2010 uses a different method, the synthesis method, where they sort of shotgun the analysis and they combine it with computer stuff to do it faster. It's 488 pounds. It's not a lot smaller. But the cost of actually running it is about seven cents per million base pairs. And the yield is about 750 million base pairs per day. And so that comes out to about 14,376 bases per square inch per day. So in the future, you know, chances are that's going to continue to grow. And we have a quote here that it's outpaced Moore's Law more than doubling each year. If you look at it a per square inch basis, this has gone up by something like a factor of over 50 per year. So we can expect that trend to continue and get faster in terms of what they can do. I also want to talk a little bit about databases because databases are important. Well, you know, you can seek in your individual results if you don't have matches. If your matches don't have pedigrees. If your matches don't have haven't been phased themselves, they're not as meaningful. That's kind of a state of where some of your ancestral results are now. These are some agencies that have big databases, the commercial testing companies like you're familiar with. The US government has a big model called DBSNP where they try to keep the map of the whole human chromosome. There's genealogical databases, medical DNA research, which is growing quite a lot. Atlas and geographical data as genealogists, if we want to know where do we come from, that implies a location and mapping, not just, you know, some set of sequence numbers and independent databases as well. Some of the dimensions of the database should have raw S&P markers, STR markers, shared segment markers, because for genealogical purposes these can all be helpful clues to tell us how we're related. In other words, you can't just pick one marker and go, OK, I'm just going to go with that. To have a really good approach, you probably need to combine all three. That's got to include the match data, the phasing analysis, the ancestral geography, and ideally it should work on a mobile phone, right, because nobody wants to carry big heavy laptops too much anymore. And that's what we're trying to do with genetic homeland is integrate these different types of data in a way that's genealogically meaningful. But where we see this all going is what I've dubbed the instant ancestry field meter. So it's a portable DNA field meter you can carry around your neck or stick in your pocket, input sample material, and it emails out a sample report in very short order with a list of hyperlinks to your ancestors, available photos and geographical locations. The way it could do this is, one, the continued miniaturization of sequencing, as I talked about, you know, how many base pair per square inch per day keeps growing. It links into these kind of integrated databases that contain the DNA, the matches, the pedigrees, the phasing, the geography, and then it kind of spits out the report. So I got a credit, Dr. Tyrone Bose is going to be here tomorrow to kind of, one of the ones that first started talking about this. So, and this device, it's not a DNA meter, it's an Abbott-Eistat wireless, but it's pretty far down that road. So when I got this presentation ready, I had this kind of notional concept of the instant ancestry field meter, and there wasn't any such thing actually in existence. But between when I started this presentation and when we're standing here today, there's actually been a news announcement from a group in New Zealand with a product called the Freedom 4, which is a mobile, it's the device is this red thing here. That is a field DNA sequencer that actually works, a little red brick. And so, you know, just in the time I've been trying to get ready for this, the future is already encroaching upon us. The FBI as well for crime fighting is trying to move to next generation identification system with new databases, and they're trying to get laws changed that would allow testing outside of an accredited lab. That's one of the challenges in law enforcement is that they can't do a test in the field because it wouldn't be an accredited lab. So the FBI is trying to get those laws changed. So what now? So we've talked about all this kind of future stuff. What does it mean now? What's going to take time to get to the instant ancestry field meter? Instant coffee was not instantly discovered. The practical solution right now for you all in the room is to test as many people as possible because it's a lot cheaper to test folks while they're alive. It may someday be possible to test those who aren't, but right now it's a lot cheaper that they're alive because that's going to help you with your phasing. That's going to help you with your ancestors. All available family and identify your ancestral geography. Your testing now also documents the human tree and helps us map that. Alice Fairhurst at the FTDNA conference talked about just on the Y chromosome, the number of markers in 2013, these are genetically significant markers, 4,115. This year, 14,000. So it's growing like crazy. They're barely able to keep up with the amount of new discoveries. So right now this moment it's like a crazy, it's like trying to hold a sled down the hill, it's just running away from them. But the thing of it is, again, is that the number of lineage is finite and over time the number of markers that are genealogically significant, it's going to be finite too. So eventually we're going to catch up. So the other thing that will mean in terms of instant field monitoring is that the number of markers that we actually have to sequence on that instant meter isn't actually going to have to be all your base pairs. It's going to be the ones that we've mapped and identified our genealogically significant. So that again means the cost and how far in the future it will exist is maybe not that far. So in summary, the future of genetic genealogy, more DNA sequencing performed faster on smaller machines, sampling coverages to approach 100%, and bigger databases but more integrated with more types of data. So with that, I'll take any questions. Thanks very much for that presentation. It gives us a real insight into what the future will hold in a couple of years' time. And maybe even before we leave the room today. So who would like to ask any questions? Brad, any questions from the audience? We have one question down here. My father was adopted. And protesting my full brother, the white DNA we know, we are descended from non-goers and deafers. So it's like, you know where we came from and not having that here. So I'm trying to find the actual term of grandfather. I have a half-brother. I've got three half-brothers and a test of the one family binder. My brother has all three. He's got live family binder and MD-DNA. So there are three half-brothers. I should have them all tested. I mean, I'm doing a lot of coordinating. Like whenever we have matches in common, I'm trying to narrow it down to what particular lineage of Jambroerson's line. Yeah, if you're trying to identify a specific lineage of descent, then you should do the... With you're doing family finder, yeah, you should definitely do more cousins because as they've talked about with the family finder test, because they don't inherit the DNA uniformly. There's variance and randomness. So the more that you can test, the more likely you are to get the signature of your ancestor. And in what 8thies presentation, he talks about... Let's see if I get it right. I think if you're missing a parent, but if you have three siblings test, you should be able to do a complete phasing of your missing ancestor. Okay, oh, it's working. So yeah, we discovered through normal sleuthing methods the birth mother. So we got that tent care, but it's trying to find the birth father. And my half brothers, there's my brother, a full brother and me, we're the only two. And then from this man, and then these three half brothers. So obviously they had a different mother, so we can start to rule out what on the chromosome belongs from the mother. Because we had two different mothers, so... Okay, okay. So I mean, I was interested about phasing because all the parents are deaf. So we can't, you know, until they start allowing us to dig them up and grind down the tooth or something, you know. And you've tested all the... Have you tested all the descendants? What... Can you talk about, like, my brother's children? Well, I'm not sure. I don't follow your whole family treatment. First of all, I would say with a very complicated case, I'd recommend that you work with a professional genetic genealogist and get, even if you're understanding it pretty well, to get some help on solving your particular problem. Definitely get some help. In general, if you're looking for a lineage that you're not sure about, you'd want to test all the people you can. What you need to bear in mind is... I'm not sure about the way the exact family tree here in your case, but C.C. Moore did some good work showing that sometimes that mitochondrial full sequence test can be very informative to help you know which branch or how to find relatives that you wouldn't find otherwise. So if you're having a tough case like that, also do full genomic sequence test, the mitochondrial of the people involved, because they can help you find a match that you didn't get in the autosomal. Okay, yeah. In this case, it's on the father's side. It's not on the mother's side. Let me introduce you to Rob Warden here at the back. Rob is the IT genius behind DNA adoption and DNA get-com. So I do have a chat with Rob. Rob is going to be talking tomorrow around about 3.30 in the afternoon. Yep, on the story of DNA adoption and how the whole thing arose as he went looking for the birth family of his wife. And then I'll be talking about solving adoption mysteries in your family tree at 5.30 this evening. So there's several talks today that will actually cover that topic. Yeah, another question here from Paddy. You talked about phasing as we do it today. Will science or technology ever be able to read the paternal chromosome in full and the maternal chromosome in full so that we know which is which? Well, in full is something that people that do DNA testing have arguments about. I think he's trying to lure me into a trap there because the white chromosome in particular has some regions on the ends that do recombine and then it also has some regions in the middle that are hard to sequence, have a lot of repeated patterns that don't seem to be well traceable or that change a lot. So as we're doing Y and S and P a lot of times we're trying to avoid those regions and use let's just call it 3 quarters of the chromosome. So I think we will be able to definitely use those 3 quarters of the chromosome that don't have dead ends, let's just say. The full thing remains to be seen because we will need another sequencing technique to be able to do it is my understanding some of those middle regions. Do you wear those regions, Patty? No, I'm talking about the autosomal rather than the Y. Oh, yes, we will be able to use autosomal S and P the same as we used the Y and that's like Daniel Crouch's project that he's going to talk about is that way. We definitely will be able to do that. I'm sorry I misunderstood your question. On the Y chromosome there are sections that they really can't reliably do right now. That's just the way it's basically you can think of as parts of it they're like a ball of knot, a knot in yarn that just don't do well. But that's very technical, Manusha. Question from Jared? Yes. Thank you for that great presentation. By the way, I agree with most of it. The isoc tree as you mentioned went from 4,000 slips to 14,000. Yeah, 14,000. From 4,000. It went from 3,000 in something. Those were Alice's figures as of... That one there, yes. Yes, 4,000 to 14,000. Yes, right. So in Brexit, an exponential increase and you mentioned about the square inch which is the cost is going down. At the moment, Illumina can do a full genome they say for a thousand and this is coming down very, very fast. And I got my own full Y results and that's generating from 1 to 12 gigabytes and the whole genome generates one terabyte of data. So the sequencing is becoming commoditized it's obvious, right? The cost is going and it's generating huge amounts of data. So the big challenge will not necessarily be in the sequencing, it will be in the analysis of this data. Okay. That's one thing. Second thing, you mentioned that red brick could still look fairly cumbersome. Yeah, the Freedom 4. Otago University, New Zealand. Nanopore is issuing to researchers the little thing, the size of a memory stick which will do long reads and... It's not very, very accurate but it's, you know, they are sequencing full genomes today, right? Yes. And they are issuing it to researchers at $1,000 per device. So the future is here. Yes, there's that. And there's also a thing where this current, that generation of Illumina sequencers I think uses, it does 100 reads at a time and they're going to go to ones that do 250 in a batch. So it's growing all over, it's getting faster in many ways, many types. Now, I suppose if one of the things that will come in the future is full genome sequencing. And at the moment we're just doing 750,000 markers on the autosomal DNA, for example. And there are certain, like Patty said, there are certain areas of some of the chromosomes that we cannot access. The Y chromosome is one where we only have access to about 48% of the actual chromosome itself. 52% is inaccessible. And there's a couple of other chromosomes I think around about 19, 20, 21 where it's actually very difficult to get in to the chromosome to do the analysis. But we will be moving from 750,000 to 6 billion in the very near future. So the data issues that we're facing now are going to be magnified hugely by moving to full genome sequencing. And a lot of the work that we're doing now in autosomal DNA is probably going to go through a complete revolution in the near future, maybe the next five years or so. So I think it's going to be very, very challenging. It's interesting to see that in the space of time that you started doing the presentation, the future was already here. And your image of the sled running down the hill in the snow and we're not able to hold on to it is very, very accurate as well. So any other questions or points? Yeah, we have one here. Is there not a real concern about home genome sequencing and the healthcare implications and that data being in a database accessible by healthcare providers, health insurance providers? Is that not a real concern? Well, I guess I'd say what is the actual concern? I think there's sort of an amorphous, like we just hear the term and we think we should be scared of it, we should be scared of it. What is it we're actually scared of? I mean, your address is published in the phone book, typically, or maybe you have an unlisted one. You're surrounded by people that may have a sickness or something, so there's risk out there in all their phases of life. I tend to think that these are just kind of new technology and because it's new, we sort of are just scared of something we're not sure of. But you call genome sequencing, you can predict cancers, you can predict everything from that. Sequencing, let's say, a baby, are you not predicting you're 20, 30 years down the road, this is what you're going to get? Well, probably yes. But what's wrong with that? What's wrong with having healthier children? I mean, what's inherently wrong with it? Values are your choice. If you feel that there's something wrong with it, you don't really feel that way. But when you go to the doctor and they take your medical history and they ask you, well, did you have any relative that did this? Did you have any relative that died of cancer? Coal and so on and so forth. They're doing the same thing. They're doing it with much less precision, but they're doing it anyway. So, yeah, in terms of controlling access to that sensitive information, sure that you've got to have that. You have that now with all your other medical data. So I don't see that it really changes with DNA. It becomes larger, but I don't see that it actually is any different. That's a really interesting question that you raise. And of course, 23andMe, which is one of the other companies that provide DNA testing, they provided a medical risk and medical health assessment as part of their work of your DNA because they didn't just look at ancestry-informative markers on the DNA, they looked at health-informative markers. Of course, the FDA now has stepped in and said, you actually need to have this regulated. They probably would have got away with it if they didn't interpret the results. If they just gave you the results without an interpretation, then it probably would have been okay, but they actually interpreted the results and then the FDA said, you're using the DNA as a medical device legally. Therefore, it comes under our jurisdiction and we have to regulate that and you will suspend that part of your operation until you have validated every single one of those health tests that you're offering to the public. The FDA and the FDA are now going through the process of getting each of those individual tests validated by the FDA and the first one they're doing is on Blum's disease, which is a relatively rare genetic disease, but they'll go through each one of them. So, we have a question down here. Can I speak to that just a little more? You know, I'm reminded that you're entitled to your view and that's fine, I understand that. And you should have the choice of how your records go versus maybe I have a different choice. I'm reminded of that movie, Shakespeare and Love, where they're trying to put on a play in the theater and the magistrate comes in and shuts down the theater in the interest of public health. So, there'll be no plays here today. So, when people start talking about you're not entitled to know your own DNA, because that's where the FDA has essentially done is you're not allowed to know it because you're not smart enough or it's not safe to let you know about yourself. I reject that as a value. I personally do. I think you're entitled to know all that you are able to find out and afford about yourself and good riddance to those who say different. But that's a political viewpoint and you're entitled to your choice too. We have a question here from Dan. Yes. Go along with what Brad said in response to the question and Slant Murray said it's good that the legislation is in place to deal with these kind of things. What I would add is the interesting thing about DNA is it doesn't just belong to you. It belongs to your relatives as well which makes it a slightly more for ethical question. But otherwise, I would go along with it. All right. I'll consider that a good company then. Well, actually that's one of the things that I've run into with doing testing with my relatives between the states and here in Ireland and everyone keeps saying what about the privacy? So people here in Ireland are being even a little bit more concerned but they're being very open about it. Getting back to what she was saying about a database insurance companies look at those too and your insurance rates could go up if you're at risk. But my original question was in tests, samples in FTDNA and we've increased the number of markers as they've become available and we now have our base comparison people up to 111 and we're getting more people in but my concern is at what point is our sample used up? I mean do I have to go get some more sample just to make certain that we can keep going? Okay. A good question and that came up a lot at the Family Treat DNA conference last weekend in Houston. So I'd like to speak to a couple of points. One is in terms of your sample they say they can store it a long time but I've seen cases where it runs out. Also the more times you order different tests typically the more the sample they take. So let's say you do a new order and you're able to do five different new upgrades. Just imagine that that existed. If you could do five different upgrades with one time generally that would be say one X of sample. If you do them five different times that's kind of five different draws out of the vial that's degrading how much you've got. Number two what we've also found out is that the next generation sequencing takes I think it takes more DNA and more high quality DNA and so it tends to use up that sample more. So like the big Y test that uses a lot more DNA than some of the other tests that we had before. So that can lead to the vial being empty before sooner rather than later. So if you're concerned I think what they said was that you can't put in you can't send in additional stuff. Go ahead Emily's got more to say on this. So I was just going to say that I do want to switch gears just a little bit and answer a question that I often hear with the STR markers and there's an analogy that Tim Duncan of the Donicky project taught me which is that the SNP is a good way to think about them in terms of genealogies that's going from the trunk of the tree up through the branches and down to the limbs and so as we're getting more markers we're coming down this way. The STRs are sort of like looking at the end of the leaves of the tree and looking backwards for similarities to see what leaves you're close to. As we get more these are going to converge and the STRs will match up with the SNPs but so sometimes when you're saying you know I've got 111 STRs I've gone way up the little leaves and stems that's good but the SNPs are helpful for coming the other way from the root down and that's why the two complement each other and I think there's going to probably always be a need to do both kinds of analyses when you're interested in genealogical era history. Just to add to that I think if you want to give extra samples to Family Tree DNA they will allow you to do that so that's what Emily's going to tell you. In part maybe a little more. As far as your samples are concerned when you test with Family Tree DNA you get two vials. If you have an older ancestor make sure you get them if you have anybody who is ill enough you may lose soon get them however you can email Family Tree DNA and ask for an extra vial. Also you can email and say how much of a vial do I have left? And if there's a full vial then you may be okay. When they run out and you order a test and hopefully your person is living they will send you a new kit and then you sample again but to keep it in the future ask for an extra vial. Tell them the person testing is aged and you need one and they'll do it. Okay. We have more questions? One here for Jim. It's partly a comment then what I did the 23andMe test and in addition to just the DNA test portion there's a significant questionnaire and that's that combined probably provides the greatest validity to their medical assessment rather than just basic it on the DNA alone even though that may be significant. The other thing is that I live in California. I did not buy earthquake insurance because I had enough information to decide whether the benefits were worth the bother. And that's I think is another contribution that this DNA testing will make in terms of your own health care as to whether you are a potential potentially going to have cancer in your life you want to take care of that early rather than late to extend your life and so there's a lot more benefits than I think are the negative sides. It's also an implicit political judgment to say we should all pay the same low rates for insurance even though we have different health risks. So that's very much even though we kind of take that for granted that's actually a political choice which you're entitled to make but you can also say that maybe people should pay or maybe not. But you're making a political choice embedded in your question there about access of insurance companies. I took the 23 and me tests way back at the beginning and they came up with a list of about 200 possibilities for genetic disease and disease associated with genes anyway and 8 of them were hyper-pensity. Well they were too late I already had 7 of them. Right, but it shows the accuracy of the 8 I'm watching now. Bravo! I was interested in the comments about privacy in DNA. I don't think we need to worry about what specific DNA companies are going to do with our DNA we need to worry about what our respective governments are going to do with our DNA. Yeah, he's a point there. And there's various levels of anonymity that I'm not really touching on but obviously need to be in place and I do absolutely agree with that that you should have the choice of what it's released to and why and the people that hold DNA and run DNA databases definitely have an obligation to be good stewards of that. That's absolutely true. There's no change in that. They have a the Gena, the Genetic Information Undiscrimination Act and that says an insurance company has done that to discriminate based on your genetic data. Now how effective is that? We need a lot of watchdogs to make sure that that's applied properly. Right, and yes it's huge issue privacy. We don't have a similar law here in Ireland. So anybody who's done a 23andMe test in Ireland and has their medical risk assessment, are they at risk of being discriminated against by Irish insurance companies? What about their children? What about their children? Community raising? I've done those tests as well and I've got to say the things they conclude on there I find to be very early days and very preliminatively speculative in point of fact, but we shall see. Yes sir. I think I have more distrust of the insurance industry than the DNA industry. I've heard of Gina, but I can't see how that can work because if the DNA tells you that you are at lower risk than the average member of the population and you're charged the average surely you're being discriminated again by the good news, not being taken into account in your insurance premium and anyway the insurance company would probably want to continue to charge the average premium even for those who are lower risk. We'll have to go to Sheila's wheels for our health insurance then. He's a man who understands Lloyd's very well. I'm sorry to intercede again, but I just want to clear out something. 23andMe no longer sells the health test. Family tree DNA and the ancestry DNA have removed what we now know that are health markers. We don't know what the future is going to hold and we do would like to know all this information. However, since November of 2013 23andMe has been denied selling any of the health tests according to the US FDA so if you tested before that, you may know your health issues. However like whoever was saying up here it gives you the propensity there are only a few health markers that we know that if you have it, you got it and at our age, most of us know what our problems are. We have to wait until all of that changes in the future and that we end up doing medical tests or whatever. I do know Family Tree DNA has a medical arm and they're doing testing but I just want to clarify that if you tested with 23andMe now and since November of 2013 you will not receive health information. Unless you go through Canada because they have actually started selling it in Canada now for $199. If you want to get your health assessment then you can go through Canada. Last question from Hogan. Hi, I just wanted to ask you mentioned during the presentation about it's much easier to get samples, DNA samples from people who are living at the moment but that may change in the future if I understood it correctly and I just wanted to know what you had in mind with that. I don't think it's going to change I think it's going to always be less expensive less legality to test life people I didn't mean that was changing. I just meant that your individual window of opportunity as genetic genealogist your window is right now. My question related to for example if we have say for example a postage stamp was leaked by somebody 50 years ago or 70 years ago is it possible to retrieve DNA from that? Well I don't think it's done commercially it's done right now in university research labs and this question came up last weekend and some FTDNA folks acted like they were going to approach it the cost for doing that would easily be in order of $1,000 and I would say more like $10,000 and could go to $50,000 real easily so is it possible yes but it's not really commercially feasible for the average person and I would say it probably may be one day but it's going to be a ways off look at what they do with ancient DNA there's a lot of craft and work and many hours of lab work just to get the sample into the tube and the process varies because because of the nature of the material. Well we have to call it an end there but can I just ask you to give a warm a round of applause applause it's beautiful excellent