Night, atord, monday nine eastern, live on cspan. Next, using genetics and forensics in crimes and in society at large. The 10th circuit bench and bar hosted a conference in colorado springs. They met for about three hours. Let me try the experiment with the microphone. Ok. Hows this . Great. Well, good morning, everyone. It is a pleasure to be here with you today. I would like to thank the organizers for inviting me to come to this very beautiful place. What i would like to talk with you about his morning about this morning as hank mentioned, we will start with a very basic review of dna, what is dna, how it works, why you , andd be interested in it focus on an important intersection between dna and the law, and that is forensic genetics and forensic applications. I will illustrate some of the points with case studies in which the na has been used inference of context. In which dna has used has been used in forensic context. Cells,almost all of our in the nucleus of the cell, we can find dna. The dna is organized along chromosomes. We can observe these under a microscope. If we look very closely at these doubleomes, we see this helix structure, the classic structure of the dna molecule. Dna are these bases, a, t, c, and g. Have 21,000 genes and each gene encodes an important component of our body, an important protein or enzyme treated or enzyme. We can think of the dna sequence as the bodys instruction manual, it is the shock menu for the human body. I will show you a sequence here. This is about 3000 basis dna bases. Each of our cells has about 3 billion dna base pairs. What im showing you here is one millionth of the entire human genome. If i were to show you a picture like this one every second, it would take me 12 days to show you the entire human genome. That is the amount of information we are looking at when we try to sequence the whole human genome and is one of our major goals today, to provide whole sequences of humans and other animals, as well. That sequence allows us to understand more about our diseases. Tions to with 90 accuracy, if i had your dna sequence i could predict your eyecolor. If we had a saliva or blood samples from a crime scene, we can predict certain physical traits with some degree of accuracy from the dna sequence. For a long time, the barrier to obtaining the sequences was cost. This slide shows you the cost of the first human genome, sequenced in 2000 three. The cost of the human genome project, altogether, was about 3 billion. Thats a lot of money. If you think about what we spend on Health Care Every day in this country, that is about 10 hours of health care spending. That helps put it into perspective. A few years later, in 2007, a second genome was sequenced, still at the cost of 100 Million Dollars punitive when we started sequencing human genomes in 2009, the cost came down to 29 thousand dollars because of technological developments. As of the end of last year, we went down to three house and dollars to 4000 3000 to 4000. That represents a million full decline in price in less than a decade. I dont think there is any other technology that can boast that kind of decline in cost in so little time. For the cost of an mri, we can get your entire dna sequenced. Unlike an mri result, your sequence wont change. This gives us a tremendous amount of information about each and every human who is sequenced. Thousands, hundreds of thousands of people are being sequenced in various biomedical studies. Family toirst human be completely sequenced is a family from utah. That just happens to be my wife and her children. For me, we had an exercise in personal genome that turned out to be very fulfilling. Occasionally, in our dna sequence here is a sequence shown on this slide. A couple of helix, the bases, and an occasionally and occasionally a mutation occurs. This is what we learned from sequencing that one family. We were able to estimate the rate of mutation, how often do these occur . Each time we reproduce, we transmit about 30 new mutations to our offspring. Most of these are in the 99 of dna that is noncoding, it means it does not actually make proteins, structural components in the body. Of occasionally occasionally these mutations do affect the genes, and they can actually cause disease. This was a great quote from Lewis Thompson from lewis thomas, the capacity to blunder slightly is the real marvel of dna that is a lovely quote because it tells us the introduction of genetic variation, this is what allows us to adapt to a changing environment. This gives us the wonderful physical diversity we see in a room full of people like this. Although these mutations can cause disease, they also allow us to adapt. Natural question that comes up in this context is how much do we differ . If these mutations happen each time we reproduce, how much variation is there among humans . If we look at identical quote at identical twins, these are natures clones. We identify them at zero dna aces dna bases. We look at unrelated humans, do you have any guesses as to how much an unrelated pair of humans differs at the dna level . One in what . It is not very much, one in a thousand. At the dna level, this most fundamental unit of our biology, we are 99 point nine percent identical. There is an Important Message therapy if we compare ourselves to the nearest ilog correlatives , the champ, we are 99 identical to the chimp. This is for dna sequences we can actually line up and compare. If we compare ourselves to the mouse, we differ at about 1 6 to one third of our genes. If you compare this to broccoli, we are mostly different from broccoli at the dna level. [laughter] have 3think about it we billion dna bases. 3 million dna bases. Differencesmillion on average between each pair of humans. Provides the basis for forensic identification. Of us is genetically unique, unless we are an identical twin. To 4 are roughly 3 million million differences between individuals, which means there are two to the 3 millionth power possible combinations of these differences. This ensures that all of us are genetically unique. The implication, then, is if isugh genetic variation tested, each of us can be uniquely identified. And we dont need to look at every difference. We just need to look at a subset of those differences. In, as you know, is found nearly all cells of the body. Letter, semen, hair, and even quite often in fingerprints. Blood, semen, hair, and even quite often in fingerprints. Dna from an evidentiary sample can be matched and compared with dna from a suspect to implicate or exonerate that person. That is what we will be talking about for the next several minutes. We have some celebrated examples of dni of dna identification. One of the questions was, was this really saddam or was this a decoy . The na had been extracted from his two sons earlier that year. Froma had been extracted his two sons earlier that year. From thee was created suspected saddam and compare with his two sons. There was a match between his two sons and Saddam Hussein for a number of locations, helping to prove that this was in fact Saddam Hussein. Last month this article was published, where the last victim of the Boston Strangler dna was identified on that victim mancould be traced to the who had accused who had been accused of being the Boston Strangler, but who was killed in prison before he was tried on those charges. We have this famous example, the worlds most famous Navy Blue Dress. The one on which a dna sample turned a he said she said case into a president ial impeachment. The na evidence has served very important roles in a number of high profile cases and in everyday crime as well. What are we actually looking at when we try to identify people from their dna . These something called short tandem repeat. I am illustrating that here. Here is part of a sequence. Ist of this sequence repeated over and over again. That is why it is called a tandem repeat. It is a short one, it is only four bases. The important thing is these repeats tend to bury in their number from individual to individual. Chromosome five they may have one repeat from the mother and to repeats from the father. Dna, we cant hanks see he has six copies on his paternal chromosome and four copies in his maternal chromosome. You would be able to tell his dna apart based on the number of these repeats at a specific location in the genome. Here are some specific locations. Coded these4 are the core short tandem repeats. They are typically used in friends at applications. They are on different chromosomes that ensure they are independent of each other, that is an important property, we will talk about that in a minute. Suppose we want to look at a couple of these. We can label them with a blue and green label here. And then what we have to do is make lots of copies of just that little piece of the chromosome that has that repeat in it. Of the use a technique called pcr to do that. So this is sort of our xerox machine for dna. This is a pcr machine in my lab. Individual,from an we load it into that pcr machine , and what comes out our dna just the ones with interest, just the ones that contain a repeat we want to examine. Importantly, from individual to individual the sizes of those fragments are going to be different. A are going to be different in length because of this variation they are going to be different in length because of this variation of the repeat tandem number. We use a process called electrophoresis for dna products. In this case it is pcr products from five individuals. We load them into a gel i can this. We into a gel like this. We apply a electrical current. Ones with the shorter fragrance, with the fewer repeats, can we go through this gel very rapidly. If they are larger it takes them longer. We can start to see patterns. These are five different individuals, five different patterns. St ours. Two different you can see there is a lot of variation from individual to individual and the pattern of these str lengths. Lets take an example. Here we have several dna samples that we are going to load into the gel. We have the victims he na, an evidentiary sample, and three suspects. Dna, an evidentiary sample, and three suspects. Find that suspect 1, 2, or three left dna at the crime scene. None of them left for treatable dna at the crime scene. May have leftcts dna or the data could be inconclusive. From thehe dna sample victim and evidentiary sample and from the three suspects. We load them into our gel. Those are the pcr products appear. Run and look at the patterns. Betweenone see a match the evidentiary sample and any of the suspects . Shout out if you seize which suspect matches the evidentiary sample. One, thats exactly right. He look for a pattern match between the evidentiary sample and any one of the suspects. If you were looking at a crime lab report, you would get something that looks like this. This is a essentially what i showed you turned on its side. Each peak represents the position of one of those fragrance fragments that i mentioned. We are looking at 10 different rs all at once. What we look at are the positions of these peaks and for d16le, the16 here here, would you say that is a match . No, its inconclusive. Let me take you to how we use this as a real practice. What he years ago, Michael Scott was accused of a homicide case in salt lake city. There was a note the na evidence. He was identified ultimately through a fingerprint on a piece of duct tape. He was also accused of the rape of 214yearold girls in salt lake city. The question was does the dna sample match the suspect . At that time the crime lab was using what our card what are called rf lps. The same principles i showed you apply here. What we have loaded into this gel, he have the victims dna, andave the defendants dna, then we have free samples from the evidentiary sample loaded at different concentrations. Pattern in the defendant, heres the pattern in the evidentiary sample. Is this a match or not . What do you think . I see some heads nodding. This was established as a match at the time. Is this sufficient to put him in jail . To convict him . We have tot because ask the question how common would this profile be in the general population . We do some statistics, that i will not talk about here, and we estimate the frequency of this profile in the population of in thencestry, population of european ancestry is one in 50. It is not convincing enough to establish the identity of that evidentiary sample. We look at another repeat system at another chromosomal location. Evidentiary sample and are defendant, is that a match . Yeah, it is a match. Again, we asked how frequent is this profile in the general population, in our reference population. It is one in 70. We have two pieces of information, one with a frequency of one and 50 and a nether with a frequency of one and 70. 50 and another with a frequency of one in 70. Now we have more information. We looked at a third system. Again, a match. The estimated frequency was one in late 90. 90. Urth system one in a fourth system, in this case the suspect got the same length of tandem repeats from both his parents. The frequency of that is about one in 10. All four of those frequencies, we multiply them together and we say how how often would someone in the general population have the first profile, the second one, the third one, and the fourth one . That turned out to be one in 3 million. Time the crime lab was using four of the systems. This gave us a probability of someone else in the general population having this profile of one in 3 million. That was sufficient to convict him of the rapes of the two teenage girls. He was convicted of the homicide and now he is serving a sentence of life in prison without the possibility of parole. This just shows several cases that i have been involved in over the years. A year after the decorso case this was a multiple rape place. Rape case. Sere were five of these rflp ystems and tandem repeats. The point, the root probability that someone else had that same profile went to one in 400 million. By 1999, we were using the pcr rs. Ed st ar with nine of the systems putting all that information together, we had a random match in 215lity of one billion. Of one of the questions a prosecutor asked was how many people are there on the face of the earth. Gives the jury the idea that it is pretty unlikely that anyone else would have that same profile. The important thing here is as we incorporate more information, these probabilities tend to get smaller and smaller. This case was exceptional because in this case the endant was a member of some conservative assumptions had to be used in estimating the probability. As you can see, we get down to some probabilities here, in the case of the Navy Blue Dress the random match probability was one in 8 trillion. Used toence can be exonerate the innocent. This is a case i worked on a few years ago where a man had been the murder of a former girlfriend. He spent 19 years in prison. His ychromosome dna was examined and it was an exclusion. As of this month, 300 convicted him americans convicted americans have been exonerated read they have spent 13 and a half years in prison. Not only did dna evidence excluded them, it identified the actual perpetrator. Are issues, of course, that come up in dna analysis. Dna is not magic. Becan be altered, it can inaccurate. We haves of questions to address is the chain of custody, was there the potential of the innate contamination, or was the dale analyst appropriately and interpreted accurately . The reference population, it is sufficiently matched for ancestral background. I think this is another important question. Was the expert really an expert in the area under consideration . Is useda evidence routinely as tens of thousands of criminal cases over the year. Contains dnase profiles of more than 12 million americans. Therell there were also arrestees in that eight obese and that database. From dakotas database, something that is controversial, when you have an have an and evidentiary have an evidentiary sample there are more than 200,000 criminal investigations. The use of this kind of evidence is becoming increasingly common, even in everyday burglaries. Dna testing is sometimes used at a cost of 1000 to 2000 per case. Number ofa interesting ongoing developments in this field. It is possible to estimate approximately the ancestry of the contributor of a dna sample. This is another controversial area but has been used in a number of investigations now. Take the dna sample and try to figure out the ethnic background of the person who contributed to it. As i mentioned earlier, predict certain physical features. There is a specific gene] three look where i can specific gene where i can look at three variants and identify people who have red hair am a sometimes referred to as the red hair gene. We can predict eye color accurately in a 90 of cases. We can even estimate age. An estimate of the age of the contributor of the sample. The book are working on rapid ena typing at the crime scene. People are working on rapid dna typing at the crime scene. Helping to decide what is appropriate, how can we best use these elements . One thing i always enjoy about talking to groups of judges, this group has the collective wisdom to help figure out how we can use this information, use this information to its best possible best purposes and avoid any potentially harmful outcomes. I would be happy to address any questions. [applause] we have time for two or three questions. Please come up to the microphone in front and remember you will be immortalized on cspan. Likes [indiscernible] [indiscernible] how accurate are those predictions and is it worth the money . The question was direct to consumer testing, were you send a saliva sample into a company, they type your dna using a dna chip, and you get bad get back a wad of information about your possible ancestry and risk of disease conditions. Is fought that it is important that companies refer recall as they dont they dont claim that these results have a sufficient accuracy for biomedical applications or for diagnostic purposes. Did send my saliva off to one of the companies. Theas interesting to get results back. The charge was 400 at that time. If you take it with a large grain of salt where it gets problematic is the diseases we are most interested in, things like diabetes, heart disease, common cancers, because they will give you some risk information for those conditions. In theblem is we genetics community has not identified most of the genetic causes of those diseases. They have gotten different results depending on which methods those companies happen to use. I would take it with a grain of salt. Its ultimately recreational. Sometimes, in criminal cases, there are accusations in which you would collect samples from a suspect. Was is a sample that somehow taken and claims to be the evidence sample. Is there a difference between the evidence sample and the collective example . If there really was too much similarity, would that be a suggestion that there actually has been some contamination of the collected sample with the suppose it evidence sample . The question was if y