FALL MEETING IN NEW YORK CITY AND BIOETHICS FORUM The Science of Human Genome Research Dr. Jeffrey M. Friedman     Click here for Dr. Friedman's bio It’s really a great pleasure to be here. I want to thank Paul for giving me the chance to, at least, frame some of the questions that were articulated in the program here about the ways in which advances in genetics and genomics, as Hal defined them for you, have an impact on pretty much every aspect of healthcare and even more ramifications to almost every aspect potentially of the way we view ourselves in the world. I want to point out, first, if at any time I look disoriented, it’s because there’s a series of lights here going to green, to yellow, to red, then flashing red, so that’s the explanation if I forget where I am. So I am going to talk about genetics, genomics and the genome project, and how this is going to effect medicine and how it is raising a number of issues that we are going to have to think through in advance, lest we be deluged with a set of issues and problems that we are unprepared to deal with and this will relate to both management of medical conditions, how to prepare for that, how to deal with what will be great opportunities for advancing medical science but also at a great cost for that, as well as a series of ethical issues. I just want to mention I will put forth some challenges or problems that come along as we go through some of these opportunities and I want to emphasize, with respect to many of these, I’m not stating a position for the moment. We can discuss it later but I’m at least going to, as we go along, to try to raise some of these issues for maybe some discussion later. What I want to tell you about today is in the field of genetics. I think people have an intuitive feel for genetics. Here’s a slide of four sets of identical twins. I think people understand the fact that identical twins who share their entire complement of DNA or genetic material are pretty darn similar and that underscores the fact that a lot of who we are and what we are is encoded in our genes. Now if you’ll look at the person to your left and your right now, what you might be surprised to know is that each of you share with your neighbor, other than the sex chromosome potentially or perhaps, 99.9% of your genetic material. So we are all 99.9% similar to the person next to us but that .1% makes a pretty big difference. In a sense what underlies what I’m going to tell you about today are the ways in which understanding what that .1% does and how it does it is going to change a lot about how medicine is practiced. Starting with identical twins, we can actually go a little bit further and ask to what extent do genes, in fact, account for differences, in particular human traits? It’s because, as I mentioned, identical twins share genes and environment, where as fraternal twins share genes more or less in the order I just indicated as true for the rest of the population but also share environment. So in general, if identical twins are more alike for a particular trait than are fraternal twins, we can assess the genetic contributions to that trait. This has been done many, many times, often times with twins reared apart and what you come away with is the hereditability of most of the conditions you might be interested in is quite high, highly significant. What this means formally is that in the case of obesity, for example, 80 to 90% of the variance in weight which was the field that I studied, can be accounted for in differences in genes. There are significant genetic contributions contributing to schizophrenia, diabetes, hypertension, alcoholism, epilepsy, coronary artery disease, breast cancer and so on. The trick is going to be to try to figure out what those genes are to provide a framework for now thinking about what causes these diseases and how we can now treat them in a less empiric way perhaps, than the way in which they have been treated. Finally, to provide a framework on which to think about how environmental factors might act upon those genes. Remember of course, not everybody that smokes cigarettes gets lung cancer, so while cigarettes might be an environmental risk, it by itself interacts with genes leading some people to be affected and others not. The interest in trying to understand the genetic underpinnings of these and other disorders motivated scientists in the 1980s to undertake the genome project, a culmination of which was made public in the late 1990s, I guess, in the form of the published sequence of the genome, that our genetic endowment is now published, it is online, any of you could log on to our website and look at it. What I want to go through with you today, is what this seminal event meant and try to elaborate on, at least, some of the implications of it, looking forward. Now, of course, making predictions is risky, particularly about the future, as articulated by Dan Quayle, but I am going to try to make a prediction that the genome project in genomics, which is the field that surrounds this type of research will be at the center of 21st century medicine and I’ll go further and tell you a few of the points I would like to address. That includes the fact that the molecular basis of disease will become increasingly well understood with the ability to identify predispositions to disease even before the disease is manifest. That with this will come a new type of medical therapy where we are targeting specific causes of diseases leading to what some have referred to as personalized medicine. But these advances are going to be associated with many medical, social and ethical challenges, not the least of which is that the practice of medicine will become much more complex, leading perhaps to even worsening of disparities in healthcare. To go through this, I want to talk to you first about the information flow of life as it were and what you’ll note at the bottom of the slide here is the binary code of a computer. Up top is the quaternary code of life in the form of four different letters, bases as they are called, the significance of which I will come to in a moment. But I want to cover in turn, first, what this information flow is and how it works; how a systematic approach to understanding this new genetic endowment is capsulized in the genome project; and the way it now is going to provide a type of systematic approach to human diseases and its treatment. Then I will go through, at least, what some of the challenges that ought to concern, at least in my estimation, responsible citizens going forward. First, what is DNA? What is this information flow? What is it that leads identical twins to be identical or quite similar and us to be different? Well the history of genetics started essentially with Czech monk named Gregor Mendel. And it was Mendel who was able to first show that discreet traits from generation to generation could be passed along in a predictable manner. It was found in the early 20th century that these traits were passed along together with chromosomes, which are shown here on an electron micrograph. So if you looked at a family, for example, the red box might mean a disease, you could trace the disease from generation to generation and correlate that with the inheritance of a particular chromosome. Somewhere on that chromosome it was inferred there is a disease gene that leaves some people in that family to get the red box and for others to get the green box. The question that drove biology in the middle part of the last century was what is this information? Where does the information that leads some people to get this disease—and others not—come from? How is it organized? It was actually at Rockefeller University where I work that it was appreciated that that information is carried in DNA ­ the structure that I’ll come to in a second. The next challenge was, of course, how does DNA provide that information content, which accounts for all of the traits that make us to a large extent what we are? Well, it was Watson and Crick, of course, who provided the framework in which the level of understanding was based in 1953 and the synthesis of their work followed by others is that DNA, the genetic material, provides the blue print for the production of proteins which do the business of our bodies through an intermediate known as RNA. So somehow DNA provides a blueprint, as it were, for the production of probably hundreds of thousands of different proteins in our body that make us who we are, that lead us to transform from a single fertilized egg into a living, breathing organism, capable perhaps of inadequately giving a presentation. How is this information content organized? If you look on the left side you’ll see the double helix posited by Watson and Crick, and if you look down that ribbon what you’ll see is a series of bases. If you look on the left-most one it goes CTA and then you look a little further there’s a TGC and so on. That is the genetic code that goes up and down each of the chromosomes, 3 billion in number. The way biology works, simply put, is that is then copied into an intermediate molecule which is read out into a series of amino acids which are the building blocks of proteins. There’s a machine, as it were, that reads out that four-letter code, three bases at a time, and inserting into a building polymer chain, one amino acid after another. That’s represented on the right part of the slide. Here’s another representation of that. So the DNA sequence through this RNA intermediate is read out three at a time in each specific triplet of these four-letter codes. GCA and T correspond to a particular amino acid which are the building blocks of a protein. A gene is that unit of DNA that encodes for a single protein. The gene is the blueprint for a single protein, I’ll tell you in a moment how many genes there actually are, there’s surprisingly few, but what was appreciated in the ‘60s is that there is a code that can read out each of the 64 possible triplets and insert each of the known amino acids into this building polymer. It’s these polymers of amino acids or proteins that do the business, largely at least, of life. What I have to emphasize to you now is that this code is not invariable; I already alluded to that fact by saying that 99% of our DNA is identical but .1% is not. Sometimes these differences have deleterious consequences. It’s as if the code is read out incorrectly and instead of reading “I am the gene,” it might read “I am the gine” and that would have health consequences. So the trick in the 1990s to a large extent was to take diseases that are associated with abnormal spelling of the code, errors in the code, as it were, and identify what they were, so you could understand which protein was abnormal in that individual and work from there. This is an example now of that pedigree showing now that particular scientific methods could be used to now infer that the red box or disease is associated with the C in a particular gene of the genetic code on not an A. That everyone who got the disease got the C in a particular position in a particular gene. It was these sorts of advances that lead to many diseases being identified and better understood. One of which is shown here, this is from our own research that lead to the realization that the child on the left was massively obese because of a defect in a single gene that lead this child to be in the 99.99 percentile of weight and that gene encoded a novel hormone that when provided back to this child made him thin. But the same conceptual advance has been made for any number of disorders that are so-called single gene disorders. These are cases where a single spelling defect leads to a predictable outcome, namely a disease, including muscular dystrophy, Huntington’s disease, cystic fibrosis, hereditary breast and colon cancer, familial hypertension and diabetes and many, many others. So what was clear in the 1990s, is that for these familial forms of disease, a strategy could be developed that could lead you to understand what the spelling error was, leading a particular protein to function abnormally and cause disease. I should tell you this is an amazing advance, in fact, if I think back to the time I was in medical school when it was unimaginable to think that the basis of muscular dystrophy, Huntington’s, so on would ever be known. Fortunately or unfortunately, however, most diseases that we would like to understand are not so simple genetically—meaning particular genetic variants interact with other variants in a complex way and environment to cause the disease. So what you see on this slide is a conceptual continuum whereby genes and environment might now interact to lead to a disease. On the far left are pure genetic disorders of the sort I just mentioned like hemophilia or cystic fibrosis, a purely environmental condition or to at least to some extent of an environmental motor or vehicle accident on the far right and then a continuum in between where somehow one or more genes interact with environment that lead to the disease. Now in a situation that is more complex of this sort a new set of tools is going to be necessary and so the tools of the 1990s are not easily adequate to solve these more complex types of genetic environmental interactions. Motivated by this, in part, came the conclusion that while genes play a fundamental role in the disease, so on and so forth, we can’t forget environmental effects but we need to remember of course that by understanding what the genes are we can now provide a framework for studying an environment. In order to facilitate this advanced level of understanding, the genome project was instituted actually in the 1980s, which in some ways can be viewed as a systematic approach to human disease and its treatment. And by that I mean to provide the tools that will make it possible to solve these more complex medical conditions. The genome project set out to build maps of all the chromosomes to basically cut up the genome into little bits and isolate them in bacteria so scientists could study and analyze it—to use this framework, to sequence all of the 3 billion letters that constitute our genome and then decode all the proteins that make us who we are. So from this inventory of all genes we can also advance to understand what are the DNA differences in those genes that make us all so different or in other cases, so much the same. This was the objective and what I’m going to try to tell you now is how this toolbox, which is the result of a worldwide effort, with major contributions from the U.S. of course, is changing the way we look at disease and ultimately practice medicine. So first, how many genes are there? Well, surprisingly few. It turns out that based on the analysis so far, there are probably only about 22,000 genes. That may sound like a lot or may not sound like a lot, but I need to remind you that it’s actually less than many plant species. You can draw your own conclusions. Of course, the question is how do these genes interact and what are the networks of gene interactions that lead us to develop in a way we do? But nonetheless, these genes can be accessed online. You can go to a website and pull up a chromosome and look at that chromosome and get a gene inventory on the bottom of every gene in the genome, each encoding a special protein that serves a particular function in the course of our lives. Some have analogized this advance—and I don’t think it’s misplaced—to the development of periodic tables for chemistry. That is, the periodic tables provide a list of elements that are the building blocks of chemistry. These proteins, these 22,000 genes and some greater number of proteins are in a sense the elemental building blocks that make us who we are to a very large extent. With this framework we can now make advances in a number of ways. I’ll just digress for one moment and tell you a way this gene inventory has changed, say cancer diagnosis and therapy. It turns out that you can assess in a highly parallel way using DNA chips as they’re called, the level of the activity of tens of thousands of different genes. What you see is, down the left, each little line is a different gene and across the top is a different type of tumor. What you can see clearly as you now color-code the activity of the gene from red to green is different activities and so you can group the tumors based on their level of similarity with respect to the green and the red. These are sort of patterns actually algorithms known as clustering programs. What you find out actually is that you can very efficiently group tumors that look quite similar in some cases or you can group them into disparate groups. And I’ll often times expand this to show my colleagues that one group actually responds to a chemotherapy much better than the other one. This sort of highly parallel information is being processed with great rapidity to try to advance our understanding of the particular characteristics of individual tumors and how we might treat them. There’s another reason to have a gene inventory besides being able to look at gene activity and correlate that with particular forms of cancer, and that is because it’s these genes and the proteins they code that include the variation that makes us all different. What you see here is the DNA helix, as it were, represented on the top of the bottom for two individuals who might differ for a particular spelling. So on the top in a particular position on the genome, everything is the same except there’s a C matched to its partner on the double helix the gene and in another case, a T matched to an A. This is a difference, a difference that could confer biological distinction upon pairs of individuals. What about this variation, how much is there and what does it mean? Well, as I mentioned the human genome includes about 3 billion of these letters. 1 in 1,000 of them differs on average between any two people and there are about 10 differences in the average gene. In total, there are about a million differences between any pair of us because there are 3 billion bases, so even 99.9% corresponds to a lot of differences. The basic idea here is if we could identify all of these differences en mass and we had a large enough population of people, we could ask whether a particular difference is highly correlated with a particular disease in a very systematic way. I’ll show you more about this in a moment but first I need to tell you that these base pair differences, these spelling differences, or SNPs, single nucleo-type polymorphisms, as they are described, are being cataloged with an absolutely astonishing rapidity. Hardly any had been identified in 1998. In 2001, there were over 2 million such differences in the entire human population. There are now tens of millions of SNPs that have been identified in one population or another. Remember, there’s the aggregate number of spelling differences in the whole species and then there are individual differences between pairs of us and there are probably tens of millions or more aggregate differences worldwide. You likewise can go to a website and run down the spelling defect for, let’s say, chromosome 7 in a human, and on the website you could find sites of genetic difference. There might be a C in some individuals and a T in others. How do we want to use this information? It is now possible using new technologies including DNA chips to type hundreds of thousands of these genetic differences simultaneously on large numbers of people. How would that advance our understanding? Well, let’s imagine that there were two genetic variants—genetic variant A and genetic variant B—and we had done a lot of work to collect groups of individuals who were either affected or unaffected by a particular disease. So gene A, in the example on the left, is present in equal numbers in the affected and unaffected individuals and so we would probably conclude, based on this, that gene A probably doesn’t have very much to do with determining who is going to get that disease and who’s not. But now look at gene B, one of these other SNPs, or snips as they’re called, that were analyzed in parallel. The majority of the people who were affected have variant B, gene B, and the people who were unaffected don’t have it. From that kind of information with more sophisticated statistics, it becomes possible to now conclude that gene B is causing the disease and allows to distinguish to some extent between the affected and the unaffected with some, although incomplete rescission. And this is the basic tool that now underlies the genome project. Collect large numbers of individuals who either do or don’t have a disease. Catalog or analyze these genetic variants on a highly parallel fashion and from that over time more and more diseases, disease genes, risk factors will be identified. Francis Collins, in a recent presentation said, and I think correctly so, that with the next 5 to 10 years a number of genes contributing to diabetes, heart disease, cancer, Alzheimer’s, Parkinson’s and so on will be identified. So every week or so you see examples of this. Asthma genes being identified, Alzheimer’s genes being identified, either as single genes or as risk factors of the sort that I just described. What you can see here is, depending on the genetic alteration of the onset of Alzheimer’s disease, it can be quite early or frequent and advanced in the case of the gene that is one of the Alzheimer’s risk genes. These genetic differences can also account for differences in response to drugs. I’ll point out to you that adverse reactions to medicines kill more than a hundred thousand Americans yearly and have adverse health consequences for a number. Now it turns out these genetic variants account for how we respond to a drug. I’ll give you one example now that’s quite well known. You treat childhood leukemia with a series of drugs that are shown here. These chemotherapeutic agents are metabolized by enzymes. In this case, the enzyme is called TPMT. This TPMT comes in two versions. In one version it metabolizes the drug brilliantly. In another version it doesn’t metabolize the drug very well at all. So if you give a normal dose of the drug to someone who metabolizes the drug poorly based on the SNP in this enzyme, you’ve now given someone a lethal dose of that chemotherapy. And of course this enzyme now is tested for before this drug is given for patients with childhood leukemia, Crohn’s disease, rheumatoid arthritis, and others. And all you need to know is which variants does the patient carry. Because if they carry the poor metabolizer variant, you just give them a lower dose and have the same basic response. This is true for many drugs now. This is a drug called Pravistatin which is a cholesterol-lowering drug, and it turns out the genetic variants in the transfer protein called CETP particularly responds to the Pravistatin. And these sorts of observations in addition to understanding or elaborating on the basis of disease have led to a new field known as pharmacogenomics, which seeks to explore the way in which genetic variables influence the response to drugs. So, with this, tests are going to be available—tests that provide risk factors, tests that identify genes that can be the substrate for further biological analysis. Tests that will tell us what our response to the drugs is likely to be. And there are already a lot more tests out there than people realize, and there’s a subset here. You can be tested for breast cancer genes, colon cancer genes, cystic fibrosis, clotting factors that influence the risk for heart disease, or an enzyme known as P4 52 D6 which has a major impact on the metabolism of all the biggest selling drugs that are currently available. There are a lot of tests actually now that are not in common usage and there are many more on the rise. Many more. So one of the themes I want to emphasize today is that we need to be ready for the time when all of these tests become available because a lot of complexity will emerge both in terms of medical care, giving access to this information to the right people and not to the wrong people. So, what is the future here and what are some of the challenges? This is taken from the same talk by Francis Collins, the Head of the National Center for Human Genomes Research, who predicted quite reasonably that by 2010 there will be predictable genetic tests for at least dozens of conditions, with, in many cases, interventions that will reduce the risk for several of these. And moreover, assessing the genetic factors that influence drug responses, so pharmacogenomics is going to become mainstream. Of course, there are going to be issues here. Who will get access to this? This isn’t going to be inexpensive necessarily, but can we make sure access is equitable? Will healthcare disparities persist or even perhaps worsen? And might there be circumstances or a risk that this genetic information could be misused to the detriment of people who have inherited one variant or another? By 2020 it’s quite likely that personalized medicine will become a reality. Francis predicts at least, that one would be able to assess the full complimented variation in each of us for $1000. So what does the world look like at that point of time? Where each of us walks around knowing perhaps what the complement or the genes that are variant in us are and what that might mean. Well, we have a clue to that already in some instances, where medical management has gotten incredibly complex. Perhaps the most complex setting that I can think of is HIV management. Okay, so these are the questions I just raised a minute ago about how to deal with this. How will healthcare providers deal with the complexity of this information communicated to patients so they can make the right decisions? How will we pay for this without exacerbating healthcare disparities? What about privacy issues? Who has access to this data and how can it be used? Okay, complexity. So, when a patient gets AIDS or is HIV infected, you have interaction between the individual and the organs. Now, the individual has DNA variants that influence their susceptibility to this disease and that also influence their response to the drugs that they’re going to get to treat the disease. In addition, there’s variation in the virus itself as well as the opportunistic infection that these patients might get that might also be relevant to know. So, how is a doctor going to integrate all of this information and make the right decisions? I sat down with David Ho years ago, who’s pretty much the leading clinical expert in HIV management, and put together an algorithm about how you might do this. And what we’re clearly driving towards is the need for electronically managing medical information in a safe and ethical way, and developing systems that can allow a doctor to act on that information without having to memorize potentially millions of interactions. And, at some point, this is where medicine is heading. When we worked up this algorithm, which was a modified version of an expert system, David Ho said that in his estimation if someone followed this algorithm, they could probably practice HIV at a level that is only currently practiced by half a dozen people worldwide. Now all of medicine is going to be this complex sooner than you may think. I’ll bring to your attention a recent headline in CNN and the New York Times today that says that the FDA has now Okayed implanting medical chips under the skin. Well, someday that chip is going to include the complete DNA sequence of that individual—the so-called aggregate number of SNPs. What will the implication be? Well it’s not at all hard to imagine that from this, at birth potentially, you could get some estimates about what your relevant risk of one condition or another is. This is just a made-up disease risk profile. You might also have a personalized formula or the particular doses that might influence which drugs will be good choices for you or bad choices, and which drugs perhaps you shouldn’t be on at all. Well, this is where we’re headed and as I said, I think it’s sooner that you might think. It’s probably decades away, but not much more than that. And of course this information is going to be incremental, so we’re already in a position where I think we need to start thinking about some of these issues. Now, one of the really important considerations here is that with respect to many of these risk factors, they’re not guarantees you will or you won’t get a disease. They’re risk factors that change the likelihood that you will. And so, there’s going to have to be a lot of education of patients and doctors to explain what it means when you say, “You have an increased risk of this, but not a guarantee that you will get it.” And conceptually that’s not actually the way most people are used to thinking about healthcare problems. Nonetheless, this will be implemented into medical practice. I also want to point out that where it becomes immediately evident with this is that there are no perfect human specimens. All of us carry a significant number of DNA glitches, so that at the end of this, it’s unlikely anyone is going to feel especially superior to anyone else. Okay, this is going to get implemented. For example, if physicians and patients are pragmatic and are given tools of this sort in time that have an impact, they will be assimilated. But it’s going to require much greater reliance on an electronic system for warehousing this data and expert systems for analyzing and integrating it. And we are not very far along with respect to the lab. And of course then a whole other set of questions that I’m not especially well equipped to address come up. How will people be enabled to make judgments based on this information? How are we going to communicate it? There is an absolute dearth of genetic counselors nationwide, an astonishingly low number, and physicians in their training are not trained to communicate or deal with this type of information. We need to start thinking about how we’re going to transmit this actionable information to doctors so that it can be transmitted to patients or else people are going to be terribly confused and the information won’t be used appropriately. And also who gets access? Just to sort of raise some of the sorts of questions that may face our children or our grandchildren and this could be the basis for further discussion, I kind of want to list a set of questions I’m not going to answer, but that you can all answer for yourselves. What would you do if someone told you that your unborn child would inherit a treatable disease, such as phenylketonuria, which is a disease that is screened for? What would you do if someone told you that your unborn child would inherit an untreatable disease, like Tay Sachs disease? What would you do if you knew that your child or your newly born child had a very high likelihood of being athletically gifted? Would you do anything with that information or not? Should you? What would you do if you knew your child would become obese? This is something I can tell you the answer to, at least as far as what the public would say? What would you do if you knew that the child had an increased likelihood of being obese? What I’ll tell you is I was once asked this on television, I think the statistic that I heard was that 70% of individuals said that they would abort that pregnancy if they knew for a fact their child would be obese. What would you do if you were told you were highly likely to develop Huntington’s disease? What would you do if you were told you had an increased risk of developing Alzheimer’s disease prematurely or heart disease? Now what I tried to do here is illustrate a few examples of diseases for which there are treatments and other for which there aren’t. And that might be the basis of discussion. And lastly, you can begin to ask, which of these things would you really want to know? It’s not a simple question and I expect each of you would probably give a different answer to different questions. Well, there’s a growing consensus that whatever you would do with this information that it should remain private and numerous bills are in one phase or another, trying to ensure patient privacy for their genetic information, so that at the point at which you could walk around with a chip of this sort, it belongs to you and only you have access to it, or those you give permission to. But I just want to point out that ensurers already make judgments based on tests that are to some extent genetically determined. And I won’t go through a long list here, but I’ll just say, high blood pressure. High blood pressure in many cases is completely determined genetically and in other cases their genetic contributors. So, there are already healthcare judgments made over things people don’t have any control over. We probably need to think about these broader issues, I think beyond just simply just thinking about what the genetic variants are. And I want to close with one case where, and I have no idea what the answer is, but where patient privacy could conflict with the public good. And this is meant purely to be provocative. I’ve never talked about this publicly before so I’m a little nervous mentioning this in front of esteemed ethicists, but here we go. There’s a syndrome known as Long QT syndrome. This is a hereditary arrhythmia. It’s rare. It affects the electrical rhythm of the heart in otherwise healthy people. And the consequence of this is that these people are highly susceptible at random to an abnormally rapid heart rate that could lead people to faint or even die suddenly, in a completely unpredictable manner. There are seven different genes that lead to this condition. All of the ones discovered so far affect ion channels, or pores in the membrane, that allow ions to transit, and this in turn regulates the rhythm. And moreover, these seven diseases can easily be identified by genetic testing. So, I just want to ask the question, and I’m curious to hear what the answer is, does an employer or the public have the right to know if someone had Long QT syndrome—if that person were a stock broker (well, for this organization it’s probably yes), or a CEO? What if he was a bus driver and driving your kids around? Would you have a right to know whether this person was at risk for sudden death? Or what if this person was airline pilot? Do you have a right? I don’t know. Do we have a right to know that information or not? And moreover, let’s take it a step further. How would we handle this if we found out that a sibling of an airline pilot had this condition—knowing that now the airline pilot had a 50% risk of having this condition? Well, these are the kinds of questions that are going to come up in a very broad context that I think are the kinds of things responsible citizens need to be thinking about and making collective judgments about because I can say for at least myself I’m not better able, I don’t think, to contribute materially to these kinds of questions compared to anyone else. Look, the objective of researchers is this, and it was capsulized by William Ostler in the late 1800s, to rest from nature the secrets which have perplexed philosophers in ages to track their sources to the cause of disease, to correlate the vast stores of knowledge that may be quickly available for the prevention and cure of diseases. These are our visions and I think we’re about to move forward in a way that’s going to realize what Ostler was hoping for, much more quickly and completely than he might have imagined 30 or 40 years ago. We’re going to need to be ready for that and I think it’s going to require that responsible citizens understand what this is about and help lead the way in trying to make sure we do this correctly. Because genomics is going to lead to better healthcare and it’s going to allow us define genetic environmental risk factors for disease, develop strategies for treating people individually, developing gene-based therapeutics for single-gene and complex disorders. But, if we’re going to do this correctly we need to figure out better ways to educate healthcare professionals who provide the necessary tools, consider in advance what will inevitably be potential healthcare disparities, and deal with privacy issues in an ethical and responsible manner. So, with that, I have a flashing red light so I will stop. Thank you.