 really appreciate the opportunity to participate in this meeting. I think it's really remarkable actually to look around the room and see the diversity of groups that are represented. I've never seen this many specialties societies and accrediting bodies and the like all be in the same place at the same time to talk about this issue. So it's not hard to find a quote these days, hand wringing about how poorly prepared physicians and other health providers are to take account of all the opportunities afforded by genomics. I won't read this to you, but I'll say that you'll find, I hope that at least my perspective is one of an optimist. I actually believe, first of all, this is not the first time that medicine has had to deal with a major sea change in the way medicine is practiced. I think antibiotics could be an example of that. Imaging, I think, may be another example of that. I believe as this really does infiltrate practice, the community will rise to the occasion. But at the same time, I think there is a real power in having groups like this together to help lubricate and accelerate that process, which is, I guess, what I hope comes of this. I also think that the issue of education in genetics and genomics needs to be nuanced that it's not a one size fits all problem or solution. And I hope I'll convince you of that in the next few minutes. Let me begin following on Mira's point you learned by case examples with a story. This is a family that was seen in our genetics clinic recently. I have fictionalized the presentation a little bit. Actually not so much because I was concerned about the privacy of the family as much as I was concerned about the privacy of their doctors. So the story here is this young child presented actually to an endocrinologist with chronic diarrhea, or at least a long history of diarrhea that wouldn't have ordinarily been a reason to see an endocrinologist, I don't suppose. But for this very strong family history of multiple endocrine neoplasia type on a dominant trait, and you can read the different components of the syndrome. So the endocrinologist seeing this history ordered MEN1 testing, which was sent to a commercial laboratory and he got back a report. I've redrawn it, but it's pretty much an accurate depiction of what they said they found, namely a variant of uncertain significance. They somehow, I have no idea how actually, they placed this exactly midpoint between a benign variant and a pathogenic variant. And so there they were with a test now that really led them no farther along the path to a diagnosis than they started. And it's hard to know actually whether they appreciated that this variant of uncertain significance was a variant of uncertain significance or whether that was good enough to label this child as having MEN1. I'll tell you by reading the chart you would conclude that he has MEN1 and was treated from that point forward accordingly. By the way, his diarrhea turned out to be gastroenteritis. A little while later, the mother was referred to an adult endocrinologist with the question, did she have it? He wasn't so convinced. So he actually referred her to the genetics clinic to organize genetic testing. The testing was sent to a different, in this case, academic lab. And they found a known pathogenic mutation in the MEN1 gene, one that had been well annotated in other families as being pathogenic. The mother did not have that variant of uncertain significance, telling you it must have come from the father. I don't know that the father ever got directly tested, but it disproved that variant as being related to the disorder. Our counselor called the commercial lab that had done the original test. Were they sure that they would have seen the actual pathogenic variant if it had been there, and they swore that it was, but no, you didn't pay for us to do this. Therefore, we won't write an amended report telling that the variant hadn't been there. So the question could be asked, what competencies were required in order to manage this family? And I'm thinking from the perspective of the original endocrinologist who looked at this child, who by the way we now know does not have MEN1. Although I have a feeling it will be a long time before that's erased from the medical record, whereas CARES provided. So you would hope that they would determine that there is risk to the child based on the family history, and they did. That they would recognize that the child might benefit from the diagnosis. I think it's actually a little questionable at 18 months. But actually the screening begins at five years. So they're not so terribly far off, I'll give them a check mark for that. That they would order MEN1 genetic testing to elucidate the child's risk. Well, they did, never mind where they sent it. That they would appreciate the significance of a variant of unknown significance, and I think no in this case. That they would test an affected, known affected individual in the family first. So as to actually know if they could or couldn't detect a mutation, will they get a definite point off on that? And that they would formulate an appropriate care plan. I think that's an ambiguous point, especially and not sure what the appropriate care plan for an 18 month old is with this. But I guess the story here is not a rare one in terms of knowing more or less the right thing to do, but quickly getting in over your head in terms of the significance, and I think arguably doing harm in the sense of labeling a child with a diagnosis that he actually just simply doesn't have. So one could ask, what are the necessary skills, knowledge, and so forth required to analyze, interpret, and utilize genomic information? I would make the point first that we need to focus on competencies, not knowledge, I think one of the big mistakes many of us make in teaching is focus so much on the details of knowledge, you come out having no idea what you can actually do as a consequence. I think that point decision support tools will increasingly be used in making many of the decisions based on genomic information. That point's been made I think a few times already. I use the metaphor sometimes of the practice of medicine is evolving into what I imagine is the sort of activity of a pilot who is flying 747, which I'm told cannot be landed safely without computer assistance. I think medicine as time passes will not be possible to practice safely without computer assistance, without point of care decision tools, because nobody can possibly remember all the details that we need to know. But that doesn't excuse the health provider from having some understanding of the principles of practice. And if anybody thinks that we will be completely dependent on computers to make our decisions, let the stern visage of Abraham Flexner set you straight, who 100 years ago was I think largely responsible for the transition of medicine from a sort of trade education to a professional education. So let me briefly make a point about three sort of dimensions of medical practice as regards genomics, diagnosis, treatment, and prevention. Some parts of diagnostic from a genetic perspective are, I think, in fact, pretty straightforward. There is only one mutation that causes sickle cell anemia. And I don't think you need to be a medical geneticist to offer a test and interpret it when you're looking for a specific, very well-defined single nucleotide change. I think people have gotten pretty comfortable with more complicated situations, as typified, for example, by cystic fibrosis, where a handful of mutations comprise the majority, but doesn't take long after that to be in a sea of enormous numbers of mutations to interpret. And my favorite disorder, NF1, there are, I think, at last count 2,700 different pathogenic mutations that the lab has identified, many of which are subtle, some of which are not. And the skills necessary to interpret that can vary from it's quite straightforward to it's really very complicated. The ability to do cytogenomic testing, which I think has probably been one of the first really obvious successes of genomics and medicine in terms of really transforming the way the analysis of autism spectrum disorders and congenital anomalies are diagnosed, sometimes straightforward, but often far from it. And then the so-called diagnostic odyssey of families going from doctor to doctor, test to test, spinning around this wheel, spending thousands of dollars with each turn, not getting a diagnosis, but now the possibility of looking at their genome, or their exome at least, as a way of short circuiting this process, but again opening Pandora's box in terms of the numbers of kinds of variants that may be found. And then as you look, everybody is a carrier for something, 100% of us would have something to keep a genetic counselor busy if our genomes were sequenced. And in this one paper, this quote, they found an unexpectedly high proportion of literature annotated disease mutations that were incorrect, incomplete, or common polymorphisms. You cannot believe everything that you read and do need to be a very critical consumer of the results of genetic testing, because things that we thought were true often don't turn out to be true. You need to be alert to how to work through a family. This is a family seen in our clinic with a family history of breast and ovarian cancer, and actually lots of other cancer types. A BRCA mutation found not in our patient, but in her sister. Our patient turned out to be negative, and that's what made it possible to interpret this favorable result to her. But it's not the way most physicians are taught that in order to take care of your patient, you first have to learn something about another member of the family. So there really is a different mindset that often is required. Forgive me, but I think this is really pretty boring, looked at just in that way. And this is a bit deceptive, because if you are a practitioner now and you remember something about the genome that you may have thought, it might be that the genetic code is a three-letter code. And you know we often use this metaphor of life. So what is a book where three letters are the predominant length of words? And this is one that comes to mind. So it isn't impossible to come to the conclusion how hard could this be? And I would argue, and I think probably most of you would agree, that if you do want to use a literary metaphor for the genome, it's probably this, that you need to learn to read between the lines to understand obscure references because it is deceptive how simple and how concrete it can seem and how subtle, in fact, it can be in the background. And if you'll forgive one more literary metaphor, it does not take long in genomics to pass through the looking glass where the rules that you thought applied simply don't. And being alert to those nuances really is what separates somebody who is sophisticated user of genomic information from somebody who is a neophyte. So what are the competencies that one might need? I wouldn't argue this is a complete list, but they include recognition of why you would do testing, picking the right family member to test, which might not be your patient, being able to discuss risk benefits and the risks of not having the test be reimbursed, intelligently selecting a laboratory, if patent rules give you a choice, interpreting the report and recognizing the limitations of the test and of the interpretation itself, recognizing the potential for secondary findings if you're looking at the whole genome, you may find things that are medically important but are not directly relevant to the question originally asked, referring to a specialist as needed and of course being able to either discuss the results or at least to provide support to them if somebody else is discussing the detailed results. So it is not new to be working in an area where most clinicians are not intimately familiar with the day-to-day practice. Newborn screening has been practiced in the US for 50 odd years. In the last 10 or 15, the scope of newborn screening has increased dramatically. Largely thanks to tandem mass spectrometry and it has the impact of clinicians on Friday afternoon at 4.30, getting a call from the state lab with the possible diagnosis of something they literally have never heard of, probably wasn't even covered in medical school or residency and if it was, it would have been a bad sign that they did remember it because there's something else they should have been paying attention to. How do you deal with that? And this is a project the American College of Medical Genetics and Genomics has spearheaded to develop so-called ACT sheets, which are actually designed to be embedded in electronic health records. I don't know that they are at this point so much, but the idea is this is what you need to know to do the right thing next, realizing it's probably inevitable you're gonna refer the child to a specialist, but you have to say something to the family when you get that call from the state lab and you have to know whether the emergency practices that need to be put in play right then and this isn't intended to be a tutorial on obscure rare metabolic disorders, but it is really just what is this and what do you do as the next step in order to make sure that the child is put on the right path. And I think the concept here of being able to provide these point of care decision support tools to help clinicians deal with things that there's no reason to expect that they would be familiar with is a powerful one and has opportunities to be disseminated much more broadly. In the realm of therapeutics, there are many examples where genetics and genomics has played a role in increasingly more that will and Matt Nibb I think is a poster case of a drug that was designed based on understanding of a genetic mechanism although I will never forget shooting myself in the foot with this. Trying to argue more than 10 years ago why a genetics sort of exposure should be embedded in a pediatric clerkship and when I was asked, name one thing that has really changed the way we think about treatment based on genetics. I used the Matt Nibb as an example where the Philadelphia chromosome showed there was a genetic rearrangement and then a drug was designed to overcome the effects of that to which I was asked, how much genetics do you need to know to prescribe a Matt Nibb? The answer was none actually. I mean it's good to know that without genetics there would be no such drug but it isn't necessary to know the details of that to know the risks and benefits and the dosage and so forth and I think it's an important thing to remember. However, genetic testing is increasingly being used to guide therapeutics. Cancer has probably led the way because cancer is fundamentally a genetic disease and the evolution of a cancer involved the acquisition of so many genetic changes for which drugs that are specific to particular genetic subsets really do need to be identified. Of course, this is getting exceedingly complicated very quickly as we move from one gene at a time to looking at the whole genome and how the pathology and oncology communities will respond to that complexity I think is going to be an area where huge educational opportunities exist. Pharmacogenetics talked about a lot, increasingly practiced, but where clinical utility can do. I believe is fundamentally a point of care decision support thing. I guess not being one that practices it maybe I could be educated but it would be my guess that if warfarin testing is used on a day-to-day basis, it will be embedded in the electronic health record when you prescribe, it will calculate what the dose ought to be based on a test that is in there. I don't know how many doctors are gonna think that they need it. I think if the institution decides that they need it, it will be there and the calculations will be done. Not optimistic that people will necessarily learn to do those calculations on a case-by-case basis. Perhaps others will have other perspectives but to me this is a real opportunity for electronic records and prescribing systems to embed all of this and it becomes more of a policy issue that it does a specific practice issue. What about the area of risk assessment? We've seen some sort of early adopters of risk assessment approaches in the context of direct-to-consumer testing. Of course, this world is evolving. There aren't so many left standing as were originally put forward. The model is a controversial one. I would argue that this may be an example of a so-called disruptive technology, the idea that you can send your sample to a company without the intervention of a health provider, get a website back that tells you about your various genetic polymorphisms and their at least putative significance. Disruptive technology is one that begins as basically a toy, which I think this is sort of recreational genomics, it's been called. And if you look at the top curve in red, the paradigm of one doctor or one counselor sitting in a room with one patient at a time, talking usually about one or two, gradually will get better, better, has over decades and presumably will continue to. But there is a limit to how much bandwidth there will be in that approach and will it be possible to incorporate a genome's worth of information into a traditional session? Whereas the consumer-driven testing is way at the bottom of the curve, it is really just a game arguably at this point if you're willing to spend the money to do it, but it is possible, I don't know for sure it will be, but it is possible that as the data sets get better and better, the paradigm will become stronger and stronger and may very well disrupt the traditional approach to medical care and although you only know a disruptive technology is disruptive after it disrupts and it's hard to tell that in advance, it is something to just realize that the paradigms may shift far more than perhaps many of us would be comfortable with. So with all that, what are the training needs? And this is actually a fairly similar non-nephrologic example of what I sometimes think of as the vector of education from pre-health professional and I think it starts at the community and the undergraduate level actually through medical education, residency, CME and the ability to use maintenance of certification as a lever to get people's attention, I think is something that's important, but really this is the entire spectrum of medical education and in fact, there was a statement to this effect a few years ago, the Association of American Medical Colleges and Howard Hughes Medical Institute had a committee that I participated in on the scientific foundations for future physicians. This is the competency that was identified as relevant to genetics and I have to say it predated any clinical application of genome sequencing so arguably it's already out of date. I won't read it to you but the idea here tried to encompass in one sentence the kind of paradigm of what we wanted medical students to have become competent in by the time they graduated. This by the way was an obviously successful document because it made every medical specialty equally unhappy. Everybody came with a phone book sized list of all the things they thought medical students should become competent in and nobody left with more than a paragraph or so of what was important in their discipline and it wasn't intended to be a curriculum, it was intended really to be a kind of statement of fundamental competencies with the hope then that the faculty would flesh this out. I'll digress for a second, this is a project we've been doing at UAB which we call the Personalized Healthcare Competencies Project, identified domains in genomics, pharmacogenetics, informatics and what we call culture and have tried to find competencies that apply to the undergraduate, free health professional student, the health professional, business line, engineering student, the idea is that all of these issues are not specifically for the health professional, they are for everybody. If you are an undergraduate who has no idea that you'll go into a health profession you still need to be an educated consumer of medical information and of course recognize career opportunities as they come along and if you're a business student there are issues of intellectual property and innovation that are relevant so this I think really is a much more than just medical school set of issues. American College of Medical Genetics and Genomics has been struggling with the question of how do we attract more medical students? Amir has pointed out the challenge, they're exposed to genetics usually in the first year but then more or less disappear from the exposure to any medical genetics as it relates to day to day practice. So the ACMG has put together this summer genetic scholars program for the past couple of years funding a set of students to spend their summers working in the context of medical genetics sometimes doing some research but a lot of it is observing medical genetics in practice from day to day. Probably now what about 30, 40 people have gone through this Mike, something in that range. 40, 50. 40, 50 and it is something I can tell you in our institution that there has been a huge amount of interest in and aside from getting a handful of students into this it's created a buzz about genetics and allowed us to set up support groups. By the way this graphic I guess is meant to show that there's kind of need for lots of hands although I worry sometimes that maybe at the very base of this there's quicksand that we're seeing. In 2004 and then again in 2006 we set up a meeting at the Banbury Center in Long Island looking at the question of how do we increase the number of trainees and how do we define what is a medical geneticist to begin with? It turned out to be a harder question than I think any of us had imagined and put together a variety of initiatives one of which was to define the competencies of a medical geneticist and we met now a couple of years ago to put this competencies document together. You won't be able to read this. There are two here that would not have been on our radar screen in the past. One was to actually participate in or at least assess a clinical or translational research study which we believe that genetics trainees should be exposed to and the other is to provide counseling based on the application of whole genome or whole exome sequencing. I don't think we appreciate it at the time but we did this in the nick of time because this was just as the Milestones project that Dr. Naska has alluded to was getting off the ground and then when our turn came, this was the blueprint that we had in stock that we were able to use and it has been translated into Milestones for medical genetics training that I think are in their final stages now of being vetted and hopefully will soon be approved so it will see the light of day. Meanwhile, we've been asking the question not only how do we train our residents better but what do we do with all the people in practice who learned to practice medical genetics before any of this was feasible and the college now is trying to launch a so-called genomics academy. It's kind of an immersion course. It will actually be aimed both at the clinicians and at the laboratory geneticists with some kind of common sessions that are intended to impart basic principles and then to have separate sessions focusing on counseling and interpretation issues versus laboratory issues. So this is in process now of being formulated. I alluded I think at one of the questions to the task force on genetics education which I chair for the college. This is one of the initiatives that has emerged from that and for which we're now trying to formulate the curriculum and also identify sources of funding to help support this but we believe it can be a way of providing a kind of basic education to a large number of people who long ago completed their training and now are being asked to do things that are really far different from what they were trained in. So finally, I think we really need to be willing to kind of bend the curve in terms of educational paradigms. There are lots of tools at our disposal that didn't exist just a few years ago. The massive online courses so called MOOCs that now are infiltrating college education probably most powerfully but provide an amplifier so that you don't need to replicate courses all over the country but maybe could be taking advantage of the best expertise wherever it resides and creating systems that will be available to people in places that may not have expertise on site. I think the power of the Khan Academy concept and any of you who have explored that can't help but be amazed at how diverse the background of one person can be in teaching and it really I think points out the value of a master teacher but the ability to take advantage of the adult attention span of 10 minutes which I can tell you for sure is mine in terms of listening to a lecture but the idea of breaking complex things into bite-sized approachable pieces and why isn't there something on genomics in that style? Probably Khan himself will do it if we don't do it soon. The ability to do different kinds of ways of providing educational materials we launched an immersion course in genetics and genomics for clinical investigators at UAB this past summer and created an iBook to accompany it. So it exists actually for a few hours it was the top hit in genomics on the iBook store. I have to say we took it off because I realized that our faculty hadn't expected to be producing a free book that the whole world would see when they did this but it provided a way to embed both the lectures and lecture notes and problems and even a bioinformatics exercise in a module that was different and I think easier to work with than even going on to a web system. And the idea of putting things in places like YouTube has been alluded to, iTunes U and various other domains now exist as ways to bring the message well beyond the confines of any specific residency. So a few last points. I think we are a long way from having a fully annotated genome and it's deceptive because you can look at concrete reproducible findings and still have no idea what they actually mean in terms of clinical action and the constellation of particular findings that may be found in a given individual may really be I mean it's actually sort of the nightmare of personalized medicine which is it's so personalized that nobody's ever been seen with your constellation and are we going to be able to predict this? My personal view is that the prediction in genetics and genomics is going to be a lot more like predicting the weather than the kind of crystal wall predictions many people talk about. You'll have some idea of about what to expect but to expect it to be precise is probably just not in the biology. I think we're going to need to see more and more point of care decision support tools. I think we are here because of the opportunity for collaborative partnerships and hope that that really is the outcome of this meeting and I think I'll amplify the point made at one of the question sessions. We fully recognize in genetics that not everybody wants to go from soup to nuts in terms of medical genetics at the bedside of a newborn with congenital anomalies or with an inborn error of metabolism if what they care to do is cancer genetics or cardiovascular genetics or you name it and we are very open to creating fellowship tracks that a person who is trained in oncology can become trained in the genetics as it relates to oncology without necessarily going through every bit of what a board certified medical geneticist does and providing some appropriate recognition of the competencies that are achieved in that. And I think we're going to see more and more the need for new counseling paradigms because that one doc, one counselor, one patient sitting in a room I don't think is gonna be sufficient as more and more information floods to us. And I will stop at this point. Thank you. Thank you, Bruce. I think that I've been thinking about this as the metaphor of the navigation system that we have in our cars now that the knowledge versus competency we're not, we don't have to memorize all the street level knowledge but we have to learn how to read the map but we also have devices that can tell us how to go and I think that sort of continuum is useful in thinking about this and I think where we're at now is we don't quite know how to teach people how to read that map of genetics and genomics. Actually, so there's another piece of that metaphor which is another story which is I was, we were in Ireland one time and we had a GPS telling us where to go and I plugged it in this small town we were supposed to go to and it wasn't until about an hour going in the wrong direction that we realized that either the GPS had it wrong or there were two towns with that name and so it really led us down the wrong path and the point of this was that I was so blindly following the GPS that it didn't occur to me I was going in the wrong direction to where I, and so intelligent use of intelligent systems is really what we wanted. Well at least you didn't end up in Slovenia. That would have gotten pretty wet along the way. Any other, Jean has a question? You can take the metaphor a bit further. I'm told that the GPS system has built into a 38 microseconds of difference between the time and the satellite and the time on the ground because of relativistic effects, the gravity and the speed of the satellite so that simple map is much more complicated than you might expect and it's one of the few proving implications of relativity I think. Wait till we have to tell students, medical students they have to master relativity. Yeah really it's right, Heisenberg may have slept here. Heidi. So earlier we were talking about the notion of either credentialing physicians to be able to order tests or perhaps requiring a consultation. The interesting thing is as we move into the genomics era and we now have the ability to order whole exomes or genomes, you no longer have the gating factor of a physician needs to know what tests to order in order to order that test. Now all you need is some appreciation that there's a genetic component in the story to then order a test. And so I think we're now in an era where the real requirement is on the interpretive side. And in fact you need sort of three pieces of that interpretation. You need the laboratory genetics expertise in terms of understanding variants of uncertain significance whether they have an effect on gene function or protein function. You need that sort of medical genetics expertise appreciation of family histories and genetics context. And then you need the subspecialty expertise which really gets into detailed phenotyping within a particular medical specialty. So it's not just this individual has hearing loss but in fact they have temporal bone abnormalities, otocoustic emissions and very detailed aspects that only a specialist in that area would understand. And so in my mind in order to effectively interpret that genome for a given patient, you need to touch on all three of these levels of expertise and have some sort of collaborative approach to return a result that's useful to that patient. And so one of the concepts that we've been working at proposing a leadership at Partners Healthcare is the notion of a genetics enabled expert that sort of spans these domains and can facilitate communication and interpretation between the lab and the physician. And then that could be a remote interpretive process much like a radiologist or someone might interpret an EKG remotely between the lab and the actual physician. So I wonder if you can comment on your thoughts of who that type is. We've thought about certain laboratory geneticists that have particular clinical expertise or molecular genetic pathologist that might fill certain domains of expertise or in fact a specialist like a cardiologist who has developed genetics expertise. But have you thought about some of those sorts of concepts? So I guess the main point is that I think it's true in all of medicine and it's certainly true here that medicine is a team activity at this point. The era of the doctor being a master of everything and just being the one individual that a family would anchor to. I think is just not feasible anymore. No one person or even type of person encompasses all the expertise that is required. So you could parse this into the quality of the lab reports that are generated needs to be at a high level so that there is enough information embedded in it and taking account of the clinical indications. Although as you say, the clinical indications may be very vague actually increasingly. We think maybe there's something genetic here and that's about as much as you may see from some clinicians but I think the lab reports do have to be written in a way that walks people through what is being found and that's gonna be increasingly challenging to do and requires training laboratory geneticists to be able to manage that. On the other end of the spectrum, somebody needs to be able to sit down with the family to walk them through what has been found and I think it'll vary. Sometimes it'll be pretty straightforward and overwhelming. Maybe they do have a genome's worth of information but there's one thing that was found that is so important to them that it will dominate the discussion but other times there's sooner or later going to be a need to look at things in far more detail and people, counselors probably especially will be working I think at the bedside to a substantial degree. In between that, particular specialists as it relates to a domain that has been the focus of a particular test. An otolaryngologist could well be that we'll need to appreciate the nuances of if you found this, here are all the things it implies and I think Mira's model of the embedded geneticist is a very powerful one and having somebody who is at the sort of shoulders of the specialist who is sophisticated about the interpretation of genomic information available to help guide them is so we're talking a four part process for many of these and I don't think that's an unreasonable guess as to where we may be headed. Yeah, we have time for one brief question and then we'll take our break, Mark. Bruce, I just wanna echo your conclusion that decision support systems are critical particularly in an area where we're using it pharmacogenomics and in fact we wrote a paper about our GPS a number of years ago, genomic prescribing system. So we have taken this metaphor into actual practice and are actually using it in a decision support system working with a dozen physicians at our medical center which include primary care physicians, non-genetic experts for sure and it's absolutely fascinating because each user of the system is using it differently and this is, there's an interesting learning curve here. We sort of handpicked 12 people that we thought were receptive and the thought of actually bringing this out to the great masses is gonna be an overwhelming challenge even with a decision support system. I think that one of the things that many people worry about as they talk about decision support systems it causes the clinician to be looking at the computer screen not at their patient and I think there is a risk of that in the current 1992 iteration of electronic health records which is largely what we're working with at this point but in fact I think the potential of really powerful information systems to my mind that's another disruptive technology waiting to happen is enormous and I actually think it has the potential of personalizing medicine rather than depersonalizing it because if some of these straightforward and principled or complex but in some sense mundane calculations are done for you it ought to free you to actually listen to your patient. Eric was talking about the undiagnosed diseases program which I've been able to kind of see at close quarters. To my mind all the technology they use the thing that was most impressive is that they talk to their patients something which clinicians do all too rarely I think these days. Thank you Bruce. We're going to first of all before we take our break I'd like to thank all of our morning speakers that was excellent talk. We have a 10 minute break listed but that's a little shorter than we had intended so we're actually going to reconvene at about 10 20 so 15 minutes we will be ruthlessly on time you will get a stentorian two minute warning from me if you're not in your seats or approaching them thank you.