 OK, ladies and gentlemen, it gives me great pleasure to introduce our next speaker. Barbara Ray-Venter comes to us from California. Barbara has a PhD in biology from the University of California in San Diego, and she's a member of ISOG, the Monterey County Genealogy Society, and she's a retired intellectual property attorney who specialised in patenting biotechnology inventions. Now, Barbara has been interested in genealogy for many years and in genetic genealogy for probably the last 10 years or so as well. But Barbara has been doing a lot of work with law enforcement agencies in the US and has been instrumental in identifying the Golden State Killer among other projects that you've worked on as well. So please give a big warm welcome for Barbara Ray-Venter. Well, thank you, Maurice, for that nice introduction. And it's very nice to be here and have a chance to present my work to all of you. So the topic of this is cold case solved, identifying offenders and victims of violent crimes. And this is, of course, using genetic genealogy. It should be on, yes. Try it again. There we go. So traditionally, genetic genealogy has been defined as being the application of genetics to traditional genealogy or family history research. But in fact, genetic genealogy is not just for family history research anymore. It's now being used for a number of different applications in addition to family history research. So direct consumer tests are being used for finding both relatives, for example, by adoptees, donor-conceived people, and living victims of violent crime. It's also being used by law enforcement via the development of forensic tests which allow the creation of the same kind of file that is generated by the director consumer companies for upload to jet match. This is a SNP file and you've just been hearing about SNPs from Maurice. And so in this case, you can now start identifying unidentified victims of violent crime and you can also identify perpetrators of violent crime. But there are some differences in how these different groups of people are treated. Even though, as I'm going to show you, the technique that is used in every case is identical. So if you have somebody who is a living adoptee or somebody who is donor-conceived, they're okay to test at any of the consumer companies and of course they can also upload their data to jet match. This is even though in many cases for adoptees, anonymity was promised at the time that the child was adopted. And also for those who were donor-conceived, anonymity was guaranteed to the people who donated eggs or sperm in order for the child to be conceived. Living crime victims, it's also okay to test them anywhere. I'm going to talk about one project in which I was involved, what I call the Lisa Project, where I identified a woman now in her 30s who was abducted when she was about six months old and had no idea either who she is or where she was from. It's also being used to identify, for example, Benjamin Kyle. In his case, he had been beaten up very badly, dumped unconscious behind a Burger King, and when he was revived in hospital, he could not remember who he was. But genetic genealogy was able to identify him. Nobody was concerned about him testing anywhere. But then if we start looking at people who are deceased crime victims, or we're looking at perpetrators of violent crime, suddenly the door slams shut, and those people, or at least those DNA samples, cannot be tested in any of the direct-to-consumer companies, and about the only place that they can actually be identified is using GEDMATCH about what you've been hearing a lot. So this is a quick reminder for those of you who haven't taken biology in a while, although Debbie did give you a quick refresher course. You have 22 sets of what are called the autosomes. You've inherited one copy of each pair of these chromosomes from your mother and one copy of each pair from your father. You've also inherited a pair of sex chromosomes. If you're a woman, you have inherited an X from your mother and an X from your father. If you're a guy, you've inherited an X from your mother or Y from your father. And as you've already seen, with this technology, you can then look at people who are related to you probably back as much as seven generations. If there's a lot of endogamy in your ancestry, in other words, you've got many cousins marrying cousins, then it's quite possible to actually get a little bit further back. What happens at each generation is that you're inheriting half of your father's DNA, half of your mother's DNA. So the DNA that he inherited from his parents, you're only getting 50%. So there's a dilution of 50% going here, 50% here, and that's happened at each generation. So let's take a look at who the players are here. In the U.S., there are four major testing companies. We've got Ancestry, New Kid on the Block My Heritage, which is actually an Israeli company, Family Tree DNA, 23andMe. And then I'm throwing in here JetMatch, although JetMatch doesn't actually do any testing. It's a repository for the DNA files that have been created at each of these companies and also actually several other companies. Their data files can also be uploaded to JetMatch. So depending on which company you test it in, of course, the source of the DNA if you're doing direct-to-consumer testing is either at Family Tree DNA or at My Heritage. It's going to be a cheek swab. If you've tested at 23andMe or Ancestry, then you need to create a spit sample. This then is analyzed using a chip, and on here you can see the results of the testing. So what's happened here is the DNA has been extracted from the cells that were in the sample that you sent in. Copies have made of that many, many thousands of copies of your DNA in a step called amplification. And then the DNA that has been amplified is then incubated with this chip, and then it's scanned by a computer to look at what SNPs you have available in your particular sample, and then your results are compared against everybody else in that company's database. And it was mentioned earlier, if you're somebody who is, for example, adopted or you have a particular line you're interested in, in order to maximize your chances of finding whoever it is you're looking for, you want to test it all of the companies, and also upload to JetMatch. Forensic samples take a little bit different path. You're going to have several different sources of potentially of autosomal DNA. It could be tissue, maybe some liver sample. It could be a blood sample. It could be bone. It could be a semen sample. And I've been working with a lab which is actually able to get autosomal DNA out of rootless hair. The analysis can be different, or it can be the same. There are a couple of companies that will do SNP arrays on forensic samples. One of them is DNA Solutions. The other one is Gene by Gene, which is actually the parent company of Family Tree DNA. They have just recently started offering forensic testing. The other possibility is to do whole genome sequencing, which Maurice mentioned. There's a company called Full Genomes that does that. And then from the whole genome sequence, they're then able to use an algorithm to extract from that enormous file the SNP data that is comparable to the SNP files that are created by the director, consumer companies. Of course, in this case, the matching is not within the company's database that's done the analysis. It's through JetMatch. So what is JetMatch? It doesn't do any testing. It's just a public database. There are actually two forms of the database right now. There's what I call Classic JetMatch, which is, if you've tested at Ancestry, at Family Tree DNA, or at MyHeritage, that's where you would upload your raw DNA data file. And there's a new version of JetMatch that's called Genesis. And this is for folks who have tested with the new chip at 23andMe, the version 4 chip. And then also living DNA. And if you want to help the cause of catching a criminal, please upload your raw DNA files to JetMatch. I use it mainly to run matching on my forensic samples. I also, there's a utility on there for estimating eye color and also admixture, which is the term that's being used for ethnicity. So this is some of the things that are available on the JetMatch site. There are quite a number of utilities, besides the ones that I usually use. So this is the Classic site. Here's the Predict Eye Color. And then here's the Admixture. But as you can see, there's all kinds of other things that you can look at. You can see if your parents are related. You can look for people who match maybe one or more kits or two or more kits, excuse me. Or you can even do a multiple-kit analysis. This is then the menu on the Genesis site. This is still a beta site. They are in the process of trying to migrate over the folks who are in this database over to Genesis because at some point all of the companies will be using GSA chips. You can see that the tools that are used are very similar to what are available on the Classic site. They also, on both sides, offer what they call Tier 1 testing, or these utilities. And so for a small donation, I think it's $10 a month, you can actually get some much more advanced tools from a much better one-to-many comparison. It has hyperlinks to the information that you might want to retrieve on here. And they also have a very interesting option that's called Lazarus. And so what you can use Lazarus for is to actually, if you have a deceased family member, if you have had enough relatives who test, you can actually recreate somebody. So my father is deceased. He was actually going to do DNA testing for me, and quite literally the kit arrived in California the day that he died in New Zealand. But I've had a lot of my paternal relatives test, plus I have two brothers. And so I was able to upload my brothers and my paternal cousins to this Lazarus utility and actually create a virtual version of my father's DNA. I don't have a lot of matches on Genesis. This is actually a really spiffy new tool that was developed by Kitty Cooper. And so what it allows you to do is take your top 20 matches that you have. And this will just, it will automatically then do a chromosome map for you as to how people are matching with you. If these were big matches, you could be really excited, but they're unfortunately very small matches. Because you can see I've got several little clusters of folks who are all matching in the same locations on several chromosomes. So unfortunately I just don't have a lot of good matches mainly because I'm an immigrant in the U.S. And even though this test was done by living DNA, most of the people taking it I think still are primarily from North America. But it's a really spiffy little tool. And you've already seen this. This is looking at the huge number of people that have tested. I'm going to be talking, as I said, about the LISA project. And the LISA project I did back in April which was March of 2015. You can look and see the number of people that were in the databases there. Not very many. And if you compare it to now, so back there, even 23 and me and Ancestry were less than a million. Family treaty DNA is not even, it's barely off the ground here. And GenMatch, I don't think it was actually, GenMatch maybe had, I don't know, 250,000-300,000 people at that point. You see with Ancestry, they're up to 10 million. It gets sold. This is as of August of 2018. This is in, it looks like, April of 2018. 23 and me was reporting around 5 million. I think they're now up around 6. You can see my heritage is out here at about 1.2. I think a 1.3 million. Here's GenMatch at 1.2 million. And then Family Treaty DNA down here is a little bit over a million. So just enormous numbers of people in these databases. There are a couple of places that you can go. You've already heard of one of them, a DNA painter for looking at when you do get matches, figuring out how somebody might be related to you. Are they a half-sibling? Are they a cousin? Or what are they? I'm a suit changer with a group called DNAadoption.org, and we actually have our own little site that we've been developing. It's a beta site, so it actually might disappear at some point, but you're welcome to use it. It's a little more precise than DNA painter, because in addition to putting in the amount of total matching DNA, you can put in information about the largest matching segment. And so it gives you a little bit better information about the matching. So this is the DNA adoption site. For those of you who would like to learn more about the methodology that I'm going to talk about today, we've actually been teaching online classes on this methodology since Ordisemble DNA became available. And that's around about 2009 or 2010. So this is not a new technology. It's an old technology. And so you can actually go onto the website and click on classes. So we offer two levels of Ordisemble DNA classes, and we also offer a class on why DNA. We also have on the site some free classes that are called First Look, which I'll show you how to use the various websites. So for example, there's one on each of the testing companies, sort of to show you how to navigate around the site. And there's also one on Jetmatch. So back in March of 2015, this webmail came in to our website. And one of the things that I do with DNA adoption is I answer the webmail. And so this came in from a deputy, Peter Hedley, in the San Bernardino, California, Sheriff's Department, Crimes Against Children Detail. And what he wanted to know is if the technique that we are teaching to adoptees to help them find their birth relatives could this same technique be used to identify somebody who didn't know either who she is or where she was from? And I said yes, and I volunteered to work on this project. And this is actually how I got myself the most in doing things with law enforcement. I retired, I'm supposed to be working in my own family history. And somehow I've ended up getting sidetracked into these other projects. So the first project then was to solve who Lisa Jensen is. This was the name that her doctor had given her. It was a time using the alias Gordon Jensen. And so my first project then was to identify who she really was. And of course what I didn't know when I volunteered was that this was not going to be a typical find somebody's birth parents kind of case. It's kind of had a life of its own. So I identified Lisa's mother as being Denise Baudin. And she was from Manchester, New Hampshire. She's missing potentially an additional vector. This is the person who abducted Lisa. And I'm going to tell you a little bit more about him. As you can see there are a number of aliases listed up here. So he went by Robert Evans, Bob Evans, Gordon Jensen, Gordon Curtis Jensen, Curtis Kimball, Jerry Markerman, Lauren Spanner. And actually there is another long list of names that he used. These are all stolen identities. He was an electrician and so he actually worked on a number of home improvement types of projects. And he would apparently steal identities while he was working on somebody's home. So at the time that he was caught up with nobody really knew who he was. And I'm going to talk a little bit about the Allentown Four and my attempts to identify who they are. Actually it turns out additional victims of Gordon Jensen. Their bodies were discovered not far from where it turns out that Gordon Jensen was living in New Hampshire. And then I'm also going to talk about the Golden State Killer and how he was identified. So as I said at the beginning, the technique that is used is really, you can think of all of these cases as being cases of unknown parentage. And so the technique that is used is identical in every case. It's a technique that we, at DNA Adoption, call the methodology. It's also called DNA segment triangulation. And it was actually originally developed to help adoptees find their birth relatives using multisomal DNA. So what is DNA segment triangulation? If two or more people match with a third person on the same segment of DNA and, and it is the end that is important, they all match with each other on that same segment of DNA. And they've each inherited that shared segment of DNA from a common ancestor. And if you remember when I put up a little chromosome map from Genesis, I circled the places where there were people who had overlapping segments. So if those people had matched each other on those segments, then those would be one common triangulation groups. So I'm going to show a rather complicated slide. It's just important to follow the logic of what I'm doing with this kind of analysis. You don't have to actually understand the mixing and matching that has gone on with the DNA. So we inherit 50% of each of our parent's DNA. And as was mentioned I think by Debbie, basically then you're getting 25% of your DNA from each of your grandparents. And so you can use this kind of information to then try and figure out which ancestors you might be related to in what way. So let's take a look then at this rather complicated slide. So over here, these big tall things are the autosomes. So for each of these people here, you're seeing you've got two copies of each chromosome. This could be any one of the 1 through 22 chromosomes. You've also got the Y chromosome. So the son down here, he's got his Y chromosome, which he's inherited down the paternal line, virtually unchanged from one generation to the next. We've also got the mitochondrial DNA represented by the circle here. And again, this has been inherited down his paternal line because the son inherits from his mother. This does not pass it on to the next generation. Again, this has passed on virtually unchanged from one generation to the next. The big difference with the autosomal DNA, and that's what all of these little colors are showing, is that this has recombined at each generation so that you end up with a different random mix. In fact, if there were more children here, they would all have different random mixes of the parental DNA. So if we're going to assume in this situation then that our son here has matched with two people in whichever company he is tested with, and he's matched with them as second cousins on this piece of yellow DNA. If we trace up here, we can see where he got that yellow DNA from. It came from great grandma up here. So assuming that there's nothing untoward about the family histories of the two people matching with, they are both going to have great grandma in their family trees. So how do we use that then to identify somebody of unknown parentage? Well, we basically just reverse the process. So let's say that son now is somebody who's an unknown perpetrator. He's got second cousin matches with two people, and we're trying to figure out who he is. Well, if we build out the family trees for the two matches, we're going to find that they have a common ancestor of great grandma. And so since son is sharing DNA with these two people, we know that he has to be a descendant of great grandma. So what we do is we then build out, having identified, and we call this the most recent common ancestor, and build out the family tree for this couple. And then we go down and we then identify a series of people who are descended from this person. We then look at that list of candidates for who our unknown person is, and we then apply some kind of a profile to those people. So it can be geography. It can be the color of their eyes. It can be whether they're bald. It can be any of a number of different things that you can then use to identify who amongst those candidates is the person you're trying to identify. So let's take a look at that in action. So you make a little profile of who you're trying to identify. Well, for Lisa Jensen, we actually didn't have much. So we're going to try and identify who her biological parents are. Information. We have no idea what her name is. We have no birth date. Based on her dental development when she was recovered in about 1986, it was believed she was probably born in about 1981. We had no idea where she was born, whether it was the US or Canada. The person who abducted her had quite literally been all over the US and up into Quebec, Canada around the time period when she had been abducted. We had no place of residence. Her last known address was in Scotts Valley in Santa Cruz County in California. So all that we really had for her was her DNA. But I was able to identify her mother, and this was a woman called Denise Baudin. She was from Manchester, New Hampshire. Denise and her infant daughter Dawn had been missing since November of 1981, and they had never been reported missing. The person who abducted them, it turns out, had told family members that they were going to be leaving town because they owed money to various people and so not to worry when they took off. So I just really want to quickly throw in an acknowledgement. There were great many people who worked on this project as I said it was done back when the databases were a lot smaller. With the whole group of us, it took 20,000 hours to actually identify who Denise's mother was. Part of the problem was that her ancestry on one side of the family was French-Canadian. Very large families with lots and lots of children. We actually printed out the descendants from our most recent common ancestor, and they printed out was seven pages landscape all taped together to print out all the descendants. So it took us a little while to sort out all those families down from our most recent common ancestor. So I had a lot of help. I had Janelle Davidson, who's a certified genealogist in Monterey, which is where I live. We also had a great many of Lisa's genetic cousins pitch in to help, and we also had members of our local genealogy society, and Dick, who also helped to work on this. I personally put in about 3,000 hours. So this was really, this is when the story started sort of taking some twists and turns. So I've now identified Lisa's mother as Denise Baudin. Denise is missing, but it turns out that Denise's father is still alive. And we were able to do autosomal DNA testing, which showed that in fact he is Lisa's grandfather. So I had worked, I had built some trees to try and identify who Lisa's father was, and so we asked the grandfather, you know, so who was her father, and he said Bob Evans. And I said, no Bob Evans in the tree that I've built, that's not the right name. And so law enforcement then showed him a picture of Gordon Jensen, and he then identified him as in fact being Bob Evans. So the person who abducted Lisa had been Lisa's mother's living boyfriend. And it turns out that Bob Evans worked for the owner of some land next to a place called Bearbrook Park. This is in New Hampshire. He worked as an electrician, and after doing some work on sorting out who this character was, they determined that he in fact was the killer of the Allentown Forest. So this is a facial reconstruction of the Allentown Forest. So we have an adult female, we have two little girls who are mitochondrally related to the adult female, and then we have a third little girl. So right now I actually hope to have the results by today. We have sequencing running on these folks, so I'm hoping that we'll have an identification within the next couple of weeks. So of course one of the thoughts that occurred to everybody was that possibly the adult female was Denise Baudin. So CODIS testing was done on the Allentown Forest and it showed that in fact the adult female was not Denise Baudin. But a huge surprise, the youngest child, or at least the middle child, the one who's not mitochondrally related is Bob Evans' daughter. This has been so difficult to identify the Allentown Forest. Well unfortunately their bodies were out exposed to the New Hampshire winters for a very long time. They were in two barrels in this place called Bearbrook Park. The first barrel was discovered in 1985 and that contained the dismembered remains of the adult female and the eldest child. They were wrapped in some kind of plastic tied with electrical tape. The second barrel was not discovered until 2000 and in that one was the two youngest children and one of them being the younger child who's not mitochondrally related and Evans' daughter. So because they'd sat out there too long, obviously the DNA is horribly degraded. What we've done is we first of all started using bone and that was very heavily contaminated. It was only 2% human DNA. The rest was bacterial. It was not practicable to really amplify further the human. It could be done but it would be extremely expensive. We do have autosomal DNA that we're sequencing from the liver of the eldest child and then what we're using because we did several rounds of trying to get useful DNA out of bone and failing we tried something different. I saw a newspaper article in the San Francisco paper about a professor at University of California at Santa Cruz who had identified a little girl who had been found in a glass coffin in the basement of a house undergoing renovation and he had used ruthless hair to identify who she was and so my eyes popped open when I saw that and so I gave him a call and so he's actually has done the work on pulling out the autosomal DNA from the hair shaft of then the adult female, Evan's daughter and the youngest child since we don't have any other tissue that we can use for them and we're now in the second round of processing on that DNA. The big problem with the autosomal DNA from the hair shaft is it's in very small fragments. They're mostly less than 150 base pairs so the only technique that you can use with this is doing whole genome sequencing. So then one of the questions was then who was Bob Evans? He had multiple aliases that he'd been using since about 1981. I was able to get a sample of his DNA. He died in prison in 2010. California autopsies, all of its deceased prisoners and so I was able to get a blood card with his DNA from which we then extracted the autosomal DNA and obtained a snipper ray which we then ran in a jet match. So it turns out that who he was was actually a guy called Terry Rasmussen. This is what the data from a jet match looks like. We didn't have much in the way of matching. The biggest match that we had, it's very hard to read on here was 115 centimorgans which is maybe a second cousin once removed probably about a third cousin. So lots of trees to build in order to figure out who this guy was but we were able to do it. In the process of building the trees we discovered that he had three children and so we were able, once we'd identified who we thought he was the detectives went out and got DNA from his son and confirmed that he was in fact Terry Rasmussen from Colorado. Of course I am way ahead of my slides. What I had done is to identify him I had actually built a profile and this was actually kind of interesting because when I identified Denise Baudin as being Lisa's mother and called the detective on the case he called me back a couple of hours later and he said she doesn't exist. Well I knew she existed because she was in her grandmother's obituary and also her brother's obituary and she was in my tree. So it turns out that the law enforcement databases at least in California they are compiled from driver's license records from voting records and so on and so since Denise presumably has not been driving her voting since 1981 she was no longer in those databases. So I used that piece of information for Bob Evans I asked the question since he's been using aliases since 1981 is he also going to have disappeared and they said probably. So what I did as a first screen was to look for people who were disappeared so maybe in Ancestry you could find both records or maybe even a marriage record but then there would be nothing else there would be no voting records there would be no listings in public directories and so on and I used that to try and find men who might be Bob Evans wise. I also knew that he lived in Texas and California and Hawaii and I knew he was an electrician and so after having applied our little screen from somebody not in the contemporary databases it did come up with a candidate. It happened to have gotten divorced in San Mateo County which is just a little bit north of Monterey where I live were able to get a copy of his divorce records and this particular person we had identified was an electrician so we contacted the deputy and then the part that I already gave you about checking the son we did at that point to confirm the identity. The other thing which was kind of interesting is that we had found high school yearbook pictures from when he was in high school in Arizona and the folks who were working on this and law enforcement they were real excited about the pictures they did age progression on them and said yep that's him it matches the match shots and I said wait a minute I want DNA so that was then when they ran out and came with me and got some DNA from the son to run in CODIS. So what this showed was that Terry Paytter was a serial killer. He'd go into prison for killing his common law wife in San June dismembering her and burying her in the basement of her house under 200 pounds of kitty litter. He has obviously he's now being determined to be the killer of the Allen's Town Four. He's presumably also killed the mother of his own child that's presumed then he's also presumably killed Denise Baudin. This work was also done in collaboration with Janelle Davidson and she did a great deal of the work relating to determining that in fact she obtained the marriage records and the divorce records and so on so she was just an enormous help in sorting out who he was. So let's now talk about the Golden State Killer. He makes muslim look like a choir boy. So he's a serial killer, he's a rapist and he's a burglar. Attributed to him are 13 motors, more than 50 rapes and over 100 burglaries in California from 1974 to 1986. So in terms of building out a profile for the Golden State Killer we knew that the attacks had started in 1976 of course we don't know if those are the first attacks but those are the first ones that we're aware of. So we're presumably looking for someone who was born in 1956 or earlier. Given that these crimes occurred in Northern California primarily in Contra Costa County and then moved south along the coast in 1979 we knew that he probably lived in California possibly even Northern California. We knew from witness descriptions from his rape victims that he was fairly tall and had blonde hair. From his eye color, from deadmatch, I got pale blue gray. I also actually ran his DNA through Prometheus and they also said his eyes were blue. So these are 13 of the known motor victims for the Golden State Killer. This person here is the most recently identified and so he actually confirmed that not only were the East Area Rapist and the Golden State Killer the same person but also the Visalia Ransacker because he was killed by the person at the time who was identified as being the Visalia Ransacker. So one of the things that I, the first things that I do actually is I look at at mixture because it's going to give me some clues as to when I'm building out trees am I looking at the right people in terms of identifying the person that I'm going after. So they happen to have actually done testing with Parabon before I was working on this case and Parabon is a company that does forensic testing for law enforcement and they primarily focus on doing a lot of phenotypic type stuff so what they do is they have a product called Snapshot and it actually produces a stylized kind of picture of what the person looks like, who's DNA it is and sometimes it's actually fairly accurate. There's actually a case out of Texas earlier this year where detectives had used Parabon for their Snapshot and when they received the copy of the Snapshot from Parabon they sent a copy to the family and the son and the family, whose sister had been murdered he looked at the Snapshot and he said that's the guy who sits behind us in church so the guy was then obviously approached and he then actually did get up in church and confessed to the motor. So using the fact that we inherit certain percentages of DNA from our ancestors you can actually then do a little guessing as to who the grandparents and great-grandparents might look like in terms of their admixture. So whoops. So from this admixture we can propose that we've got two grandparents then that are Western European because we've got approximately 45% Western Europe. We've potentially got one grandparent who is Scandinavian, we've got 28% Western Europe and we've also got a grandparent potentially who's Southern European and then maybe another great-grandparent who's Middle Eastern. The biggest match that we had on Gen Match so that's probably about a fourth cousin. So what happened is we started building out trees and we did learn something else in the process of doing this because very quickly the trees that we were trying to build on what it turns out in hindsight were the paternal lines they were heading back to Italy. It turns out that the family when we identified who the Golden State killer was they in fact had been recent immigrants from Italy to New York and one branch of the family had moved to California. So in any event we were able however to develop a most recent common ancestor using the folks who had UK ancestry and had been in the U.S. for some years for some generations and so we were able to actually identify him using a non-Italian line. So it turns out that only a few descendants of the most recent common ancestor had actually migrated to California and so we actually got a list of nine candidates who lived in California and who were of the correct relationship to the matches. So what we did at that point is we decided to do target testing because we had started out so distant from a fourth cousin match and the amount of matching DNA when you're getting out there is very variable. So we decided probably what we should do just to make sure that we were actually on the right track here was to do what we call target testing. So two people were approached and asked if they would do DNA testing for us. We did their testing and we looked at how they matched with the Golden State killers DNA and very fortunately one of them, a woman, she actually was a second cousin match and she had an X chromosome match. So that meant since a man only has one X chromosome and he gets that from his mother, we knew that we were looking at a match through one of his maternal lines. So based on that we were now down to six candidates. So we applied the rest of our profile and one of those was the blue eyes that I had determined from both GEDMATCH and from Prometheus. The FBI ran the DMV records, the driver's license records in California for our six candidates. Only one of them had blue eyes and that was Joseph James D'Angelo. So this is then looking at his tree. So as I said it turned out that the Italian was the Potono line and these lines just kept going straight back to Italy. So here's our most recent common ancestor here in what turned out to be the Potono line that we used to actually identify him. And I actually had the same thing happen with Rasmussen. His Potono line, there were recent immigrants from Denmark and those lines kept heading back to Denmark and we also in his case identified him using only the Potono lines. So the final step when you're doing these forensic cases is that the law enforcement then go out and they run surveillance on whoever it is you've identified and they collect what they call discarded DNA or trash DNA or surreptitious DNA and they test that using CODIS and so they then compared the trash DNA with D'Angelo's DNA sample and got a match of 99.99% probability that they were the same person. So we now identified then Joseph D'Angelo as the Golden State Killer. So again this is teamwork and this was a very interesting team to be working with. There was Paul Holes, for those of you who've seen any of this on any of the social media pages, he has a hashtag on Twitter of hot for holes. He's a very good looking guy and I guess the ladies have appreciated that. I also worked with Steven Kramer out of the FBI in Los Angeles and also Melissa Parasau. It was very much fun building trees with the FBI because it was really pretty easy for them. And then there were two people out of the Sacramento County DA's office who also helped on this, Monica Kozowski and Kurt Campbell. And I'd like to add a congratulations to Monica and Kurt after working on this case, they then went on to solve who the NorCal Rapist was. And I'm now going to attempt to do a link and this is an adoption case I worked on. It's a really kind of a cute little case. This woman, her granddaughter, contacted DNA adoption in the same way actually that Deputy Hedley had done asking for help in identifying her 98-year-old grandmother's both parents. And this turned out to be one of the small world things. The granddaughter's father was actually the original detective who worked on the Lisa case. So it was kind of interesting that I was now hoping his mother-in-law figure out who her ancestors were. So hopefully I can make this work. Oh no! Who does that? Okay, well, you're not going to get to see... That's a pity. Oh, wow. I didn't know they did that. They bought stuff like that. He has probably one of those geophysical barriers to that. Oh, that's too bad. Oh, well. Okay, well, I'm sorry I'm going to miss that. She was really pretty cute. I promised to say something about privacy. So people who were adopted, I think most people would say that they have an absolute right to learn who they are. In other words, to identify their both relatives. Using matches to genetic genealogy, or these genetic cousins in the various databases. In her case, she had only tested at 23 and me, but partly I think because she was so old, she had some very close matches and it took me only about five hours to figure out who her birth parents were. There was also a lot of information on Facebook and other social media sites. There's information in newspapers. And in fact, for the Golden State Killer, because I'm a civilian, I was not privy to most of the information that law enforcement had. And so when I was building my profile, I was very much using things like newspapers to figure out information about the sky. Of course, there's also information on ancestry and elsewhere in the form of census records, both death and marriage records and so on. I don't think too much is said about the use of genetic genealogy by people who are adopted, people who are donor conceived, and so on as being an invasion of anyone's privacy. I tried to go back and look at when the sites were first being used by adoptees and I really didn't see too much with people complaining about that. So one of the questions is why then happens to be law enforcement who's doing exactly the same thing? Why is this suddenly a privacy issue? I don't have the answer to that. Obviously, I was actually taken aback by the amount and some of the vehemence of some of the folks on various sites about how this was unethical and an invasion of people's privacy. Because I've been doing adoptee cases for so long, it didn't ever occur to me that anybody would think that this was either unethical or an invasion of privacy. There is a lot of information out there about that DNA is falsely identifying people. From the articles that I've seen in all of the cases that I've read about where that statement is being made, there's a confusion by the folks who are writing these articles between YDNA testing, such as it's used by CODIS or just straight YDNA testing using YSOCH. As you saw, YDNA is passed down from father to son virtually unchanged, and so it's a familial test. It's not specifically identifying an individual. The autosomal DNA, on the other hand, is identifying immediate family members, and you can tell whether or not you've got the right person, based on how closely they are matching with someone, and you can then also run out immediately and do a quick test to see if you've identified the correct person. One of the issues I think is probably different here in Europe versus what it is in California, for example, where certainly collecting Sarptitious DNA is perfectly legal, and the use of CODIS to confirm the identification is also legal in California, and in California you can actually do familial searching in CODIS. Concern that medical information will be re-available, obviously this is a real concern. There are lots of laws in place, at least in the U.S., regarding how genetic information can be used. However, when you actually are doing the matching in GEDMAT or any of these other sites, you're not getting the DNA information. What you're getting is a list of people who share DNA with you. There was an article within the last few days talking about somebody who had been a de-identified person in, I think it was the 1000 Genomes Project, showing that they were able to take her DNA profile and upload it to GEDMAT and surprise they were able to identify who she is. Well, of course they were able to do that. They had her DNA. It was the only way that they could possibly have identified her. They could not identify her without having access to that DNA file. The other thing that I looked at really quickly was this is from the ethical standards that were put together by a group of genetic genealogists in the U.S., looking at some of the things that potentially would be good for guidelines for people who are doing genetic genealogy. One of them is that genealogists can only obtain DNA for testing after receiving consent. And so one of the other issues that has been raised is, well, Joseph D'Angelo didn't give his consent for his DNA to be tested. Well, excuse me. He abandoned his DNA all over Northern California. He has no privacy rights in that DNA. In fact, the DNA belongs to the law enforcement agencies, not to Joseph D'Angelo anymore. And there's actually are numbers of cases on abandoned tissue, tissue that's been removed to surgery, and so on that go along with the same concept that once the DNA is abandoned, then the person does not any longer have privacy rights or ownership rights in any DNA that's extracted from that tissue. And the last point that I wanted to make is that unknown victims, in fact, unknown people in the ancestry databases, and so on with their GEDMATCH database, they don't have any privacy rights. And as I already mentioned, another DNA from the perpetrator is abandoned DNA. I haven't seen the results of your little quiz yet, but one of the things that is really gratifying is there being a number of polls that have been done. I know we're all waiting with bated breath for yours. And they overwhelmingly show that for, if we look at this, if I can do this right, so we're looking at violent is the blue, nonviolent is the orange, children is the gray here, and then missing persons. For violent crimes and for missing persons and for children, people felt that the databases should be used. Creating fake profiles. I'm not quite sure what this is about because I'm not quite sure when I'm uploading Joseph D'Angelo's DNA to GEDMATCH, I don't know who he is. What do they want me to put? Serial killer. This is then looking at companies revealing information with a search warrant that's already in place that they do need to do that. Searching cell phone records, of course, have been released in recent cases on this. And still, there's a very high percentage of people that feel for the same three groups of people that should be okay. And then searching social media accounts, surprisingly, a very high percentage also in favor over here. Thank you. Great. Thank you very much. We're going to leave questions for the expert panel, which is going to follow on now in about two minutes or so. So, fantastic presentation. Great work that you've been doing. We're going to talk about it again in the expert panel. I'll show you the results of the survey, and we will take up this conversation in about two minutes' time. We're just going to arrange the tables up here. So thanks again very much to Barbara. Thank you.