 Good morning everyone. We have a very busy agenda this morning so if I get everyone to please take their seats and we will launch right into this. The plan for the morning which will be very packed filled with summaries and hopefully valuable discussion is as follows. In a moment I will turn the podium over to Francis who will provide an overview of what has transpired especially with respect to some re-crafting of the plan and its structure. Following that we will have three summaries from representative breakout group leaders from yesterday. We will have them each speak one after the other without any discussion between them and then we will turn the entire rest of the first half of the morning session into a discussion on each of the key components of the plan as well as any other comments people would like to make. We will attempt to break at 10 o'clock and we do need to keep things fairly strict on timing. We will have a short coffee break and then we will reconvene at 10.15 when we will get as you will see up here some summaries. I wouldn't dare. You will see we will then have some some very brief summaries from the special topics groups from last evening and then Francis will give a summary and then Dr. Sahuni will be here to provide some closing remarks. So with that as the backdrop I will turn this over to Francis. Well thanks Eric and good morning to everybody. I'm glad to see that most of you did recognize that we were starting somewhat earlier today than we did yesterday but I'm sure there's some people who are still over there sleeping thinking oh I've got another 45 minutes before they get underway and they're going to be very surprised when they turn up and discover we've already decided everything but oh well the early bird etc. So initially as we set up this program I was going to not plan to say anything until after we had the presentations from the breakout groups and the special topic groups and then I was going to try to summarize all of this long about 11 o'clock or so but because things have changed fairly significantly in terms of the structure of the plan and I think in a very good way I thought it would be good to put that all in front of you now before people get up and talk about specific areas where the plan is going to focus because otherwise we'll get all caught up in this confusion about what is the structure of the plan anyway. Before doing that I thought I'd show you this lovely cartoon that Sharon Terry gave me yesterday which seemed to be rather apt for what we're all about here. A Dilbert cartoon is always a good way to illustrate almost any gathering of people so you can see here that the guy with the pointy hair is saying put together a team to decide who will be on the strategy council. Dilbert says you want me to form a committee to create a committee that will produce a document that will be ignored and the answer is no it's a team to create a council and of course the guy with the six hairs is saying can I be on the team that ignores the document. Well we're not doing that people. That may be somebody's world but it's not the world we're in the midst of here at early and I might say my goodness what a wonderful treat it has been that they have all of you so engaged in making sure we have a document that doesn't get ignored and I think that's one of the major lessons of this gathering is how invested our community is in making sure that the plan that gets generated here really does capture the excitement the enthusiasm the potential of this field and frankly I think many of you pointed out that the draft you were provided did not catch you in that same sense of capturing a lot of that enthusiasm and that's something we can now fix. In that regard going from the ridiculous to the sublime I will give you one other quote here that comes out of the book of Proverbs plans fail for lack of council but with many advisors they succeed so I guess we must be then on the right track we have many advisors and we have a plan that will not fail okay just to remind you when we were all starting here night before last I put this charge in front of you critique the draft plan in detail you did it you said what you liked you said what you didn't like you were quite free and open about that to the point where at least five people came up to me yesterday and said are you okay to which my response was well of course I'm okay this is the way that genome community does things this is our our method of operating for the last 12 years and it's a wonderful and the fact that everybody feels very comfortable weighing in on every aspect of this plan including the overall structure not just the details is I think a wonderful testimony to the openness of this particular kind of scientific dialogue which has been our tradition and I am very much okay in fact I'm quite excited about this you also pointed out what was missing you paid attention to the grand challenges and as I will come to in a minute we've totally recrafted what a grand challenge is it has a whole new definition because the concept as it was outlined in the original draft it was felt by many people not really to work and the challenges didn't necessarily fit and we asked you on Monday night to think about interactions and you pointed out that in fact the pillars was not a very good metaphor for trying to encourage such interactions and as you will see in fact you already have if you picked up this page on your seat we've now really changed the metaphor substantially in order to try to create the kind of structure that emphasizes those interactions as being critical you all were pretty good about not wordsmithing and not promoting your own favorite hobby horse the breakout chair said they occasionally had to gong people for that but not too often in fact it was interesting in some of the sessions I was in where people were specifically avoiding that to the point where those who were most in favor of particular areas were often those who didn't do them well again we had started this conversation with this document that phrased all of this in the context of these three pillars there were pillar problems throughout the day yesterday no one really liked the metaphor at least I haven't encountered somebody who said oh I love it this is so good it just captures exactly what I want to say the interactions are very hard to depict I mean try to draw a diagram of pillars where you have connections between them and you quickly have a diagram that makes people laugh instead of encouraging their enthusiasm computational biology is a specific example seemed to be underrepresented in the formulation that we put in front of you and there was even a suggestion it needed its whole new pillar and that didn't seem like that was quite the right answer but it was an example of one of the areas it's really cross cutting across everything we're trying to do but which in the formulation that you saw didn't really get captured very well there was a tendency I think a fairly broad one for people to take that pillar called genomics to society and say oh that's the LC pillar and that would be a very damaging way I think to view genomics to society in a damaging way to view the LC program certainly there's a lot of overlaps there but the LC program if it's going to be viable on into the future and it must be has to be interacting with all the levels of genomics and closely interdigitated with the scientific enterprises that relate to basic biology and the human health and to have it perceived as being relegated often this separate pillar that's not connected to the rest really was not the outcome we were hoping for and we tried in the document that you looked at before coming here to make the distinction that LC and genomics to society were not one in the same but I don't think we really succeeded at least in many people's minds that distinction did not come across and there was the question okay if these are pillars what are they holding up and there are answers to that but none of them were as satisfying as one might hope so I think we basically had to decide we're not going to go there so we were in search of by early in the process yesterday other metaphors other ways of structuring this that would better capture all of the things that we're trying to promote in that regard this is where this diagram came to life and for this I give a lot of credit to a whole bunch of people and I should stop right now by the way and say you know the breakout chairs yesterday did a phenomenal job these people who were called to do what they thought was going to be a simple matter of leading a couple of discussions were in fact drawn into intense discussions at lunch another a couple of hours of discussion at the end of the afternoon sessions and then we're kept in the Jefferson room until quite late last night trying to formulate all of this in a new fashion and it is to their credit that we got to the point that we are now where I think we have actually a much more compelling way to capture the excitement of genomics so let me quickly walk you through this particular diagram because as the other presentations come forward they're going to be referring to this so instead of tellers we basically are building a structure here notice that down at the bottom is the foundation of that structure which is the human genome project the success of the enterprise that has dominated the activities of NHGRI over the course of the last 12 years and which is a wonderful foundation upon which we can now build for the future which is what this plan is all about the first floor of this building is genomics to biology and you will recognize that that has some similarities to what we used to call a pillar but it's a lot more than just turning the whole thing 90 degrees there is there is a there's a substantive set of changes here that I hope I can point out to you excuse me genomics to health is the second story genomics to society is actually in this particular image and we may decide whether or not we think this works but it's the roof it is the top of the building it is the thing upon which it rests upon all the rest of it now notice what is really different here you know than what we had in the original configuration which I think is very compelling are the objects that are written in vertical type and which cross across these levels and by the way when we get the final version of this with better graphics assistance we will make it clear that these levels interact with each other I'm sorry Eric excuse me yes excuse me Eric Eric green did this lovely diagram I think we should all applaud that just the same there probably are few improvements that could be made to convey the connections between the levels which might be nicely done with double helical staircases or something like that when the time comes but I hope you get the concept anyway so notice here now LC in policy is not the same thing as the genomics to society a roof of the building it crosses all the levels and well it should and did oh there are other aspects of what we're trying to do here which in some way were called in one of the discussions requisites but they are part of the enterprise that allows one to actually reach the goal and the grand challenges under each of these components and you can see those include resources that includes information science which now gets much more attention than I think it did in the original formulation technology and methods education which crosses all of these and workforce issues and we could debate a bit about which of these belong at which level now you can see here that we have formulated in a fashion so that the genomics to society component has a somewhat different list then the rest of the other stories and that seemed to be where the group wanted to go last night but that maybe one of the things we could talk about so I really like this because I think it does point out by using these two axes the floors and then if you want to call these the supporting structures that are holding the building up these cross-cutting areas it points out the difference between those two in a fashion that the pillar model really did not so that's the structure upon which the presentations are going to now come forward I just want to make a couple of other remarks and then I'm going to turn that over to those group presentations but again it's just so that we don't get hung up on issues that I think we may have already gotten pretty close to reaching some agreement about some other major points that were made about the new plan the preamble is critical everybody pointed out the difficulty they had reading the document without quite getting the sense of why is this exciting what what is it about genomics now that deserves this kind of a planning process anyway what have we done so far where are we going and how does this fit in the context of what the rest of the world is doing in that regard the preamble probably is not something that should be written by a committee of 160 people it's probably something that needs to be done very carefully by one or two very articulate persons and that will happen hopefully with the help of Maynard in particular over the course of the next a couple of weeks or so but I would like to say I heard a number of very important points that ought to be in this preamble that I'll just try out on you and we can perhaps discuss it when we get to the discussion first of all the emergence of genomics as a central and cohesive discipline biomedical research not a bunch of technologies but a coherent and a compelling discipline upon which all of this now future structure is going to be built that didn't come across very well in the current version of the document the fact that this is a historic transition from the unqualified success of the very explicit goals of the human genome project is outlined 12 years ago to this new era of discovery which is going to be a real transition and one that we should celebrate and which involves now a unique opportunity to explore radically new approaches that will benefit human health and I think we want to be very clear in that preamble that we're very interested in those that this is in fact an opportunity to focus much more in the direction of health advances and I think that preamble must also point out that just as at the beginning of the genome project focused some of its attention on the societal consequences that we're at least as interested in fact much more so in making sure that that is a component of what we do and finally I think the preamble has to make it much more clear how NHGRI fits into the scheme of things that we aim to be a leader in the central core of genomics activities that will enumerate this morning but that is not the whole story here we also aim in as vigorous a way as possible to be a partner with many other public and private entities including all the other institutes of NIH other government agencies the private sector private foundations and so on so that is a bit of a comment on the preamble let me also say something about grand challenges you'll see on this document that the grand challenges now appear in each of the floors of the building in a rather different way than they did in the draft that you looked at before coming here so we're redefining grand challenges this morning to be basically those elements under each of the goals that involve the greatest sort of core activity things that the genome project and NHGRI can do to advance this field and many of them you will recognize have some relationship to what we called critical elements in the prior version the grand challenges in the prior version were largely the ones that seem most successful technological advances such as sequencing a genome for a thousand dollars or some such thing and I think we would actually like to go to that very same descriptor I'll come back to that other one a minute and perhaps add a new component to the plan in this new version which we could call for lack of a better term at the moment quantum leaps in technology making it clear that those really are technological goals that if achieved would dramatically alter the way we do research or practice medicine but which are in the technological zone and therefore will probably fit particularly well with the bottom floor of this house and maybe a little bit with the second floor as well but it'll keep us from having to force the grand challenge in the old way model in circumstances where frankly it didn't fit as well and so now again I hope I'm not confusing you the new definition of the grand challenges are really those core critical elements under each of the goals there was much concern about the way in which the document was written that confused people in terms of who would have the responsibility of pursuing particular parts of the enterprise and particularly so when it came to the elements that were marked with the pound sign or the squiggle because they were all sort of put together there on lists which made it seem as if they were of equal intensity importance in this document and while they are equal many of them equal in terms of the long-term hoped for future of the consequences of genomics I think there is a strong feeling that this document has to do a better job of shining a bright light on the areas of genomic research that NHGRI is particularly well suited to take the lead on and while commenting and encouraging these other areas of partnerships with other entities they should perhaps be placed in somewhat of a different format in the document and perhaps in a separate part of the document as part of a sort of a background and significant section as someone suggested and I think we can do that in a fashion that still captures the importance of those partnerships but makes it clear now which of the areas of the future research are really the central core and focus of this document which is what I think people are looking for which is what NHGRI aims to do in this next phase so those are the background items that I wanted to put in front of you before we got into the presentations and I think we'll hold discussion on all of that until we've heard from those three groups that have labored to come forward with their new version of this and again thanks to them for doing so on the fly in a circumstance where we had a fairly substantive revision going on all day yesterday and last night so without further ado I guess Eric Lander is next and please come forward and you've got the power point on there too first floor so I'm gonna report for the first floor previously the first pillar that Arvind and I each chaired two of the discussion groups let me briefly report the substance and I think this is true for Arvind's group as well there was uniform and I think almost unanimous enthusiasm for the science for the scientific opportunity for how exciting it was and there was equal uniform sense that the document didn't convey that but that we needed to draw out and sharpen the focus but in a very productive discussion both in the morning and the afternoon in these two groups the idea of getting a clear crisp goal clearly emerged the idea of defining these requisites more explicitly in the group really felt strongly about the role of the information sciences about education about Elsie about resources the group spent a bunch of time pairing away some things that were felt to be inessential and spend some time with a very productive discussion on of course we all want to understand the function of components but function means so many things did so many different people and particularly afternoon there was a very good discussion about making crisp and precise what level of functional description was appropriately genomic and what level of functional description was writing in you know detailed cell cellular biological papers that was the I can't I won't go into all the specific many comments we recorded all of them there's full transcript all that quite transcripts but at least full extensive minutes of all of those discussions that will be very valuable in preparing a written report but what we did was try to abstract that discussion into three slides here are the three slides the first was a simple statement of a goal that you could tell anybody that floor one whatever point it's a floor the first floor the first floor is has his goal the comprehensive and comprehensible that that involves beta presentation understanding many other things comprehensive and comprehensible characterization of all components encoded in the human genome as well as other important organisms that's it what does that mean in terms of the grand challenges associated with that goal seven all of these are comprehensive do I need a mic okay but outside of the state I was I so rarely need a mic this is not usually have the opposite problem comprehensive identification of all the functional and structural components encoded in the genome there are many thoughts today about the best ways to do that amongst the best ways today for example would be the sequencing of dozens and dozens of related organisms and using evolution to pick those things out but of course the choice to sequence many dozens of organisms or hundreds of organisms is instrumental to identifying the components in genomes or some of these other things there are also other approaches to identifying all the functional and structural components encoded in genomes there's a comprehensive description of the key functional properties of these components that doesn't mean a total description of all biological function because we don't know how to do that but key functional properties include expression where they're expressed when they're expressed in health and disease interaction amongst these components protein interactions or interactions between proteins or other things and DNA elements and dependencies when you systematically knock some of these things out in in appropriate circumstances what things change in response to those but this is not the same thing as totally understanding biological function this is the genomic contribution to functional characterization comprehensive description of heritable variation in these components comprehensive description of biological variation epigenetic modification covalent modifications comprehensive description of evolutionary variation across the species who's changing rapidly slowly and why compliment why is of course always tough in evolution so we don't fuss too much about it but comprehensive ability to monitor these components a heavy technology component is going to be necessary to monitor those and a comprehensive ability to modulate these components such things as RNAIs and that should be chemical chemical genomics that's it this if we can accomplish that if we can make those comprehensive tools the feeling was that this will have then provided a tremendous foundation for the rest of biology including for the second floor in the third floor obviously as with the genome project making precise some of these descriptions has to be had the word comprehensive doesn't mean completely exhaustive it doesn't mean we would necessarily have every possible variation in the genome but such a comprehensive view that there would be relatively little return to getting every last one of them or something so anyway there are a bunch of bullet points under under under with the previous pillar one we felt many of them were just fine but they should be arranged under a thinking like this what are the requisites in order to do that the requisites have to do with technologies resources information science and policy the requisites some of them under technology have to do with describing some with modulating for describing this is where the need for extremely rapid inexpensive sequencing comes because we believe that sequencing is a powerful technology for being able to do some of those goals so we have to be able to do de novo sequencing resequencing genotyping which is of course a kind of sequence determination this is where the hopes for a thousand dollar genomes and quantum leaps would go hold the whole transcriptome already such things are possible but they need to be much better whole proteome such things are not currently possible and need to exist and for modulating to ensure that the development of such technologies are as RNA is in chemical chemical genomic libraries really advanced and proliferate resources to make sure that the reagents that emerge from this example cda collections and chemical libraries are created in a form that they can be broadly disseminated this is actually useful and disseminate a bowl information sciences something's off the bottom of that information sciences professional warehousing of large data sets needs a lot of attention this is a topic we never really discussed much but it's worth mentioning because they probably in the end may be the key professional architecture in order to allow the reuse and interoperability of software so everybody isn't writing their own smith waterman aligner but you can for example use a web services architecture and go off and pull down a smith waterman aligner as a module in the thing you're writing we don't do that we waste much of our time rewriting stuff and therefore also our stuff's not interoperable data analysis interpretation data dissemination of visualization and pathways system modeling come under that and then policy it's cut off at the bottom here's let me see if I can escape out of that and pull this but it's to ensure the broad access to no there's a couple more lines in here let's see if it comes up no maybe not that's a Mac it's a poor PC thing anyway ensure broad access to data resources and technology and the issue here was data availability was a crucial thing freedom to operate that is the IP issues around being able to for example use all the proteins in the genome or something like that work requires attention not just that the data should be available or the resources should be available but that the legal ability to operate those should be available and the same access to technology those are the things that have to be addressed under that policy bullet that's it education was also cut off the bottom education workshops were also cut off the bottom right here I'll just go below that well I think that the Eric's description actually demonstrates why the notion of floors makes a lot of sense because in fact the the floor of genomics to health and disease builds very strongly on on the floor that Eric just built first of all we do have fairly we have fairly broad goals here we want to identify all the genes and pathways that contribute to health and disease and then we want to develop interventions that are going to that are going to address that are going to address those to promote health and prevent disease that's a fairly audacious set of goals and rather than talk about grand challenges at least Rick I should by the way acknowledge that Rick Lifton was the other group leader and was incredibly instrumental in help formulating this particular presentation and document but I think both of us sort of think that what we really are talking about here is is developing pathways to achieve these goals and there are two main points here that I want to bring forth one is that all the groups thought that more attention needed to be devoted to single gene disorders that are currently reflected in the document and we think there are a number of reasons for this we think that single gene disorders may provide insights into other complex diseases but also that that it's an important obligation to deal with particularly rare single gene disorders and that we although other institutes also do this the NHGRI ought to be involved with this as well we also did endorse the notion that NHGRI ought to be engaged in the pursuit of trying to find the genomic contributions to common complex diseases but we felt very strongly that this ought to be in in a partnership mode this has already been alluded to but this is an area where this really comes comes to the fore very clearly first of all because NHGRI while it has enormous expertise in thinking in terms of genomics and the whole organism does have there are other areas of expertise with which NHGRI in all its wisdom doesn't have quite so well like epidemiology and some of the complex phenotyping and that there were enormous opportunities for partnering here at the same time what we thought was that NHGRI ought to be very strategic in terms of thinking about what the paradigms are for using genomic approaches to approaching complex diseases are and then be very careful in picking model projects in which to explore whether these paradigms actually work and if they do actually work to demonstrate their ability to give new insights into the etiology of disease so that was a reformulation of what we thought that the NHGRI could do now what do we need to do this and I think this is where we rely very heavily on the toolset that Eric just described because clearly there needs to be a lot more work that better faster cheaper technologies in a whole variety of ways and then but we would move to a couple of new levels in terms of thinking about information systems that I think are particularly important here one is that there needs to be more integration of genomic data sets and clinical information data sets and not only that but when people collect well characterized cohorts that there needs to be a mechanism that these can be shared with other investigators so that people don't keep collecting the same or similar cohorts over and over and over again now we recognize that there are a lot of barriers to having this happen some of them are legal like HIPAA and other issues some of them are frankly institutional and we heart back to Bob Tepper's early comments about the fact that they're serious institutional incentives that get in the way of this but this ought to be something that's very important for a genome to address the third thing that genome I think needs to address in a very clear way and this is going to come back again in the context of education is the development of tools that allow the investigate that integrate currently existing data sets genomic data sets clinical data sets in ways that are accessible to investigators and accessible to researchers in a way that can be very helpful to them in their research and in terms of doing their clinical care there is a lot more work of integrating across data sets that needs to be done that would be very important we also identified the need for better animal models because although we are the goal here is human health that clearly animal models have contributed and will continue to contribute a lot to our understanding of disease LC and policy we had a number of points to make here some of which Eric has also already alluded to but I'll make a few more one is that clearly research ethics is going to play an enormous role here not only in terms of not only in terms of the legal issues but just also in terms of the public acceptability of this research not only in the United States but abroad on the issue of helicopter genetics was raised and I think this is something that we have to take very seriously as we talk about collaboration around the world another issue that we talked about was the need to address issues of privacy I suspect this will come up in a while as presentation but but I just raise it again we also I think not surprisingly spent a fair amount of time talking about semantics and about what this document claimed and didn't claim and about the importance of wording and what we talked about and one of the most one of the most helpful comments that was made yesterday and one of my groups is by Allen Williamson where we talked about you know what our genome based interventions and what does that mean and what we realized is that what we are more concerned about here is genomically informed interventions rather than genome based interventions because that suggests that leaves open the possibility or the reality that environment broadly read also influences almost all of human health as well as whatever genomic variation is and then we also identified that there's the need to be careful about what we suggest that genomics can do and be careful not to overclaim but also not to underclaim that we need to be open to exploring the role that genetic variation plays in all disease and across all contexts. The final issue that we spent a fair amount of time talking about was education and one of the points that was made at length yesterday was that this is even more complex than people think and that you know we talk a lot about complexity but it's even more complex because typically when we're dealing with real patients they don't just have one health issue they have a lot of health issues and not only that they interact with each other and while this is something that is part and parcel of what real clinicians do nonetheless this is going to raise it to a new level of complexity and so we I think need to we need to take a leadership role in doing this and I think part of what this means and this goes back to my notion about information systems is that we really are going to need to take the leadership role in developing expert systems that help both clinicians and patients sort through the information that they're going to be getting to help them make better decisions for themselves about what interventions they're going to make that's going to improve their own health and I also want to say that we that we also recognize the incredible ongoing importance of educating the public there was a considerable debate about how much of a role in HDR I ought to take in that is at least as a part of its extra mural program but clearly it is a central issue for the long term. Let me just mention one or two LC items that did not make it onto my slide but that are enormously important and then I will hand over the mic to Wiley and the final point that I just want to make here is that we also acknowledge that the issue of access is extremely important. It's extremely important for any number of reasons it has to do with the acceptability of the research it has to do with justice it has to do with distribution across societies and around the world. These are issues that are going to be extremely difficult to solve. But there are ones that in a share I need to continue to devote a lot of attention to as it moves forward into the new era. So in presenting the genomics to society piece, I'm speaking for Sharon Terry Reiner Keagden, Joy Boyer and Jean McEwen all of whom labored late into the night trying to put a roof on this house. We're trying our best to reflect the discussion that occurred in all of the groups and what I think we found was that most of the issues that were identified in the critical elements did hold up as important issues and we've tried to reflect them in the framework that we're providing here. But that we really needed to avoid a stance of reactivity that is that issues of genomes to society need to be framed as important issues not merely as reactions to things going on as a result of genomic research and there were many comments in the groups that the way the document was written seemed to imply a stance of genetic determinism that that we knew really wasn't supported or intended. So we identified the goal as to identify and address the pressing societal implications of genomics and we've divided the challenges into two general categories of research, foundational research and policy related research. There's an important concept behind presenting this as a research agenda and that is the recognition that NHGRI is not a policy making body and yet many of the issues that arise in genomes to society relate to the development of policy. So the question becomes what is the NHGRI role and as you see here we're formulating that as a role in research that deals with basic questions those are the foundational issues and that research informs a more applied kind of research that we're calling policy related. So examples of foundational research involve analyzing the impact of genomics on concepts of race and ethnicity on concepts of normality. What do we mean by normal particularly in context of different kinds of conditions attributes or behaviors that might be influenced by genomics and implications for genomics for concepts of culture family and self and this kind of research could be approached by qualitative and quantitative data collection by normative and non normative analysis by other kinds of endeavors historical analysis for example. And then in policy related research what we're defining is very specific and that is that this research should define policy options. So NHGRI is not a policy making body but it is extremely appropriate to identify what the policy options are and subject different policy options to analysis that could again both be based on empiric research or other analytic methods because as we look at different policy options we want to look at their potential societal consequences. We've identified for that we think are the leading issues you'll see that there's redundancy here but I think that's expected redundancy with policy issues identified on the other floors. So use of genetic information in non health care settings clearly is an issue. Access to health care services it comes under genomes to health but it is really broader than just achieving health care outcomes it has major societal implications same with IP. And then another issue that received a lot of attention in the initial draft and we think is important to address is what are the policy options in terms of identifying and then potentially implementing limitations based on ethical considerations on genomic research. Now in addition to the challenge of addressing those issues we think there's a challenge in creating a sort of process challenge if you will of creating the right kind of structure to make this happen and we've outlined on this slide what we believe are the critical elements. So as I've said first of all there is the concept of accomplishing research both foundational and policy related and we see those research efforts as very interactive. But one of the points that we emphasize is that every research effort in this area should have a plan for dissemination and the plan for dissemination may be publication in the foundational research it may be only publication but certainly on the policy related side we have to look critically at what kind of dissemination are we looking for who are the right target audiences and how best could they be informed about this these kinds of policy options and dissemination is likely to take a form much beyond publication in scientific journals equally important we think is to create an evaluation process and this evaluation process will require new methodology. We think an early important effort will be to think about the kind of expertise that ought to be around the table in creating that kind of evaluation process because we think it's critical to the success of this kind of effort that evaluation should in an ongoing fashion look at the relevance of the topics that are being addressed in foundational and policy related research there ought to be an ongoing assessment are we looking at the pressing topics have we missed something it should look at the productivity of the effort what's coming out are we in fact identifying policy options in an appropriate and efficient manner and it needs to look and begin to think about the creation of perhaps a variety of mechanisms that look at the outcomes that might be relevant to measuring genomics to society can we for example create surveillance mechanisms that might look at utilization of genetic tests and tell us where the live issues are in terms of potential misuse of genetic tests or underuse of genetic tests can we identify effective means for tracking measures that might give us some indication of whether genetic discrimination is is an issue obviously having come to some definition of genetic discrimination that we can support equally important to evaluation is this critical issue and critical for this floor and all floors in this plan who does it part of the plan has to be a plan of inclusion that is an aggressive effort to bring into this process researchers from diverse backgrounds that includes researchers from minority and underserved backgrounds in that that has an implication in terms of training programs and recruitment efforts but also a very broad skill set in order to accomplish this kind of system well we're going to need to bring in a lot of research disciplines that aren't currently a prominent part of NHGRI and this really just summarizes what I've been telling you and just points out that we really see ourselves as having three segments if you look at the strips to the right on the floor one is ELSI policy that is identifying the foundational and policy related research questions under education we think there is this dissemination piece that I've just described and as I mentioned there's a workforce piece that is critical here as in other areas and that's it thanks okay thank you Wiley actually thank all the speakers for keeping things on yeah thanks all and already under budget so now we'll turn the floor over for discussion and we have precisely one hour for discussion maybe in the interest of keeping things somewhat organized I thought what we would do is take a few questions on on sort of the first pillar first floor then we'll take a few questions on the second few questions on the on the roof or the third and then we'll open it up for a sort of general cross cutting kind of questions comments actually it's also for comments the speakers are all in the front row I can have a microphone of them if you want a direct sign to them and please line up at the three microphones that we have one in each aisle okay thank you Chuck Langley UC Davis okay so sorry I'm I am looking for something in this document that builds on the history of the genome project which is a very definable goal I realize that this document reaches beyond five years but I still think all the more reason to have a very clear well-defined goal and I would like to propose that we reconsider the issue of resequencing a number of human genomes the reason for that is to completely rather than making a catalog of variation because there are two very important reasons to do that first of all we all know that the haplotype map is quite useful for picking low line fruit may not be adequate and certainly is not an end point like sequencing the human genome was and secondly in order to drive sequencing technology and to really get to the thousand dollar genome we need to have large well-focused projects based on sequencing that are out there driving the interest and the justification for developing that technology and I see that missing from this document I think it's just that you don't see it in the document but a comprehensive description of the characterization that means the resequencing of genomes because to characterize the variation it means knowing all of the variation there it's not haplotype maps it's not we all of that is out of there in fact if the best way to obtain that is complete resequencing then that is and I think probably that is that's contemplated we tried to say why are you doing it and you're doing it to try to get the complete characterization of all the variation meaning the different sequence well but no but I think the reason we didn't put in comma based on the complete sequencing is if there's another way to do it that turns out to be better fine but we don't think there is one right now but we're trying to distinguish between the scientific goal that we're trying to achieve and then below it in each of the paragraphs will be what is likely to be the very best way to get it which is indeed those bullets that were there and we agree okay that was at least least discussed please address the strategy issue though about having a definable goal that drives technology development right there are strategy issues underneath but the scientific goals got called out on top I don't think so I think there's scientific goals perhaps what's missing and that prompts Chuck to ask that question is a lack of quantitative goals I mean are they we just didn't see them yet or so these how many genomes and when and this this kind of specificity so how many genomes and when for a 10-year document the the sense of the group was not appropriate that in fact it was to define comprehensively the the functional and structural elements of genomes current thinking is that's going to take dozens of a million genomes etc etc but if the technology came along Richard the group felt that allowed you to find those structural and functional elements with sequencing no more organisms that's great too so sequencing dozens of more organisms was not regarded by the group as an end in itself it was regarded as almost surely the most likely and productive way to do it but again the focus was on the end so clearly there's a middle ground here and I think that the sentiment may be that the document is shifted over to being a little squishy because it's not it's not it's not being quantitative at all so I maybe didn't convey the very concrete nature of these discussions there's no backing off from below it's saying today it appears that we have to sequence 25 genomes etc like it already says in the document but the point is why do we have to sequence 20 25 genomes because we believe it is the way to accomplish this and the document was reached was restructured by both of the groups so that these activities that we all agree are the are very important activities are there in a context of why but there's no backing off on on many of these things there they're simply arranged under their motivations and there is a recognition that of five years from now there's a better way to accomplish it scientifically our allegiance is not really to the activity but at the moment these activities we agree on that was sorry if everybody can please introduce yourself since this is being webcast elsewhere and so forth okay i'm you and bernie from the ebi um i just to remind everybody there's not going to be a very nice clean goal like sequencing the human genome that we can point to i think that's a frustration from the two previous questions is that we can't just point at something scientifically that we you know it's this sort of fuzzy we want to know lots more about lots of things which i personally think is that i mean i i agree with all the statements i just i think we're going to come across this problem again and again that we can't find a very very well defined goal and we have to accept that and come up with goals that we can set ourselves that are what we want to get to so we just it's just part of this process in three slides we left out many things if if we don't have every last nucleotide in the genome there still was a sense that you can tell that you're approaching saturation when in fact it's not possible to define lots of new structural and functional elements as you add other organisms or try other techniques you have a sense that you have begun to produce a comprehensive catalog and when you're still adding at every single moment you're far away from it so if it's not quite as crisp as the last nucleotide it's not totally fuzzy either so the way i view this is that the data set should be both comprehensive and qualitatively change the way people do research so the comprehensive nature of data sets is a genomic feature but it's the qualitative change of research using that data which is the true contribution of genomic technology Janet Rowley University of Chicago the success of the human genome project was dependent in large measure on the development of centers that focused on on DNA sequencing and did it obviously extraordinarily well it seems to me that in the future we are going to be developing technologies that are going to be equally demanding equally expensive and i think that not only NHGRI but all of NIH should think about how they could develop such centers of very specialized equipment and expertise that could be used by the scientific community in a sense analogous to the advanced photon source which is presently at Argonne national labs and whether this is a place where any NIH and the department of energy should be considering joint projects that involve large and expensive technologies for the to help the research community I think is something that we should begin to consider okay we could certainly start entertaining any comments about the second floor as we line up but Kurt this is more general Kurt Fischbeck NINDS back to what the points that were made by the first two questions I think you know with any building it's good to have some kind of a timeline and or and I recognize that it's hard to have quantitative well-defined goals like we've had in the past but at least to have some kind of a framework is this going to be like an Amish barn raising or is it going to be more like the pyramids to know to know which components are going to be need to be in place first and then maybe as an appendix to the thing like tacked on to an R01 you know what's going to happen first and what do you want to have accomplished within five years and which are the longer term goals David I'm still on the first floor I haven't found the elevator yet David Bentley from the Sanger Institute just to pick on Janet's perhaps answer you and a question is one of the things from the morning discussion I think is yes and the initial paper focused on a lot of technology development being a strength and it's a recognized strength behind the success of the HGP and if we follow that route continue to follow that route what we're missing perhaps is the idea that this is being converted to a discipline which I think was is perhaps a central goal which NHGRI and nobody else can and should grasp to develop a discipline recognizing perhaps that's the true challenge it's an uncertain area we're moving into and the discipline perhaps is part of that is defining how to approach the uncertainty how to define rules guidelines within it to tackle it head on it's a much bigger challenge in a sense than a well-defined simple genome sequence and I would imagine that that sort of notion will probably go on the preamble and it was on and I think that would be certainly detailed there I have some advice and a question introduction Tom Pollard from Yale Eric three people have already said this so I'm going to say it again the description of the goals is not working they're not crisp enough they're not deliverable enough they're too vague and they were better yesterday afternoon they were better yesterday afternoon in the sense that the first goal was to complete the parts list second was to identify where each of the parts was expressed in the human body the third one was to identify the partners for all the parts you know it was no no no but the way you've described it the comprehensive you know it just yeah okay so that was my advice make it crisper and more deliverable like the genome institute has always done now my question has to do with what is chemical genomics how in the hell did that sneak in there because I don't know what that's all about this was strongly felt by many including in the broad discussion last night that having tools to modulate generic tools to modulate not developing the inhibitor to your favorite kinase but being able to put in the hands of investigators RNAIs that would be able to interfere with gene systematically and I think it's felt that in the timeframe being contemplated five to ten years having comprehensive RNAI libraries that have been validated is extremely important and similarly having the access to screening of diverse small molecule libraries to be able to screen phenotypically and be able to identify targets in a facile fashion was something that had to exist not nearly in the pharmaceutical industry for the development of pharmaceutical leads but as research reagents available to any cell biologists and that being able to create those technologies was crucial to ever figuring out what function would be so both of those are core technologies here what the big project no that's a good opinion but I think there's a certain sense by many others at least that was expressed that over the course of the next 10 years small molecules would end up highly complementary to RNAIs and that they would be as well distributed and facile it may prove false but that certainly was a goal and I don't think I would at this point give up on that but the goal is to modulate and if the right and that's why it was phrased as these scientific goals which I don't think are very different if it's to modulate and the best way to modulate at a genome-wide level proves to be just RNAI and chemistry turns out to not be fruitful then you stick to that as the way to modulate if chemistry turns out to be a fruitful way to do it then we should do it by chemistry so the focus is on the ability to perturb components in a genome-wide way and it will be up to the planners each year to figure out how best to invest to do that so I have a comment about the I'll introduce David Altschuler from Mass General and Whitehead Genome Center I have a comment about the overall architecture if you will of the building which is that I think that the first floor clearly has the potential to be very deliverable and quantifiable I think that many people have commented already that their level of enthusiasm would depend I guess on the specifics having the specifics which were there and I guess people want to see them and approve them but let's assume that I think that could be done the second and third floors which conveys at least dark again I'm partially dealing with the message that's sent seems like about two-thirds of the mission and I find it increasingly harder to see how you would reach that level of specificity for the second floor and the third floor and while I'm not suggesting that means the aspiration shouldn't be there I guess it seems to me that I think it would actually be pretty impressive if the first floor was accomplished in the next five years that there was a complete toolkit with all these different levels I understand I understand it's not a five-year plan but in the absence I'm just expressing an opinion in the absence of saying identify the genetic basis of common disease understand pathways that are disordered is something obviously that many of us are very interested and excited to do but it's just increasingly unclear how these goals or how you'd measure success and I think that's at least a level of you know at least the first part we can see how after additional debate those quantifications those specificities could be added to the first floor the greater difficulty doing that with the current goals and second third floor is a source of certainty as it should really be two-thirds of the whole at least visual picture and sense of what the aspiration of the group is David I have a I just wanted to say two things one is I think it's really important for all of Eric's goals that there be something written down there about what at least the current means are and while this is and I just wanted to be have a explicit statement my understanding was that under comprehensive blah blah blah there was going to be a list of things that are on the docket to do with some kind of timeline associated with each of them right that's what we should understand you just didn't make the slide okay am I correct we were told not to whatever whatever reason okay that's what's got to be under right okay that's important the second thing is I think that we have a danger of jargon here which I would like to again get a clarification from my experience in the discussion comprehensive really means genome-wide or could be scaled to genome-wide okay which is much clearer to some of us than comprehensive which could mean almost anything okay we suggested comprehensive for the preamble but I would like to lobby for the restoration of the operational definition that has served the genome extremely well for defining what is our business and what primarily and what is somebody else's namely if a proposal comes in whether it be from an R01 or some huge centroid okay the test is is it going to give you genome-wide information or can it be scaled to genome-wide information if the answer is yes we talk about it if the answer is no we send it somewhere else good so if we took out including chemistry yes if we took out all right so that distinguishes between the chemistry of a molecule and of a whole thing if we took out comprehensive and replaced it by genome-wide with a little asterisk that said or it could be scaled to genome-wide you'd be happy yes I would that was my suggestion that would be my suggestion comprehensive is there because it fits so nice and with comprehensive this is in the spirit of clarification because I believe that that was the consensus that was the goal of the committee and so there was no you know at least in the sessions that I sat everybody was comfortable with that maybe because they all come from the genome community but anyway okay yeah last point is do you think you could find a more elegant word than warehousing oh data warehousing yeah do you have an offer of a word just reformulated I just the idea that what we do when we disseminate data is run a warehouse just grits the wrong warehouse is the first part where you store it dissemination is there so you just want to explain about the word warehouse I yes I rely upon your your thesaurus is I will think on alternative storage storage is better than warehousing but I think the idea the idea of having databases and archives which are not the same thing that are accessible to various kinds of communities is a really important thing that isn't captured by warehouse all right and the other way anybody we can now move up to the roof if you want to come by the microphones it's after artists anywhere you want to go at Baylor Houston I want to approach the second floor and try to suggest maybe a format for some deliverables that one could do which would be a lot along the line of how the the sequencing itself has been handled one could say okay there's some number of genotypes per individual that would be useful to have in the idea world based something along the half map kind of data and say we will do that number of genotypes on some number of individuals on some number of disorders that we think our priority because we don't currently understand their pathogenesis and based on their abundance and we will take things like the NIMH collection of schizophrenia samples and we will dump all of the data all of the genotype data into a open database every night and so we'll say 100,000 genotypes on 100 individuals with these disorders and I think it would be painful and create a lot of antagonism difficulty at least but you make this list of disorders like we currently play off mouse versus Drosophila versus C. elegans I mean you want to play off asthma versus schizophrenia versus whatever but I think one could make a deliverable at the second floor kind of level the only thing I I think we might not know if those genotypes are really the information is going to solve the problem for instance I think we might want to add epigenotypes about the status of CPG methylation around the genome and so on but I think one could construct a deliverable that says this is the minimum we think we'll be able to go beyond our promise but we're quite sure we could do this many genotypes on this many individuals from this number of diseases maybe you don't have to specify the diseases now but say it'll be at least 10 or 20 or something like that but the goal of making this more quantitative yes well with all and also with the goal of these patients give their samples to be in the the NIMH autism or schizophrenia collection whatever to get progress I think if you took all those people all those samples and dumped all the genotypes in the computer every night it would change the architecture of how the research is done but it would vastly speed it up and you could have a few very high throughput centers dumping the data in and everybody can be mining it at their will any of the managers of the second floor want to make a comment well I certainly am not opposed to specificity I suppose that one of the reasons that we talked about strategic targeting and identifying paradigms and pilot projects was that we wanted to not constrain the leadership of NHGRI in terms of figuring out what would be the ideal the ideal projects to pursue but I certainly think that what you have in mind art is exactly what we have in mind that what we would really like to see happening here and maybe Rick can say something too about this is that what we would really like to see happening here is for a process by which it's possible to identify which would be optimal projects to go after and then to go after them specifically to do this kind of work I'm not at all opposed to what you're saying art I just didn't want to tell Francis which disease isn't exactly how to do it yeah I think that's an important point the other point is that it's not clear that what's available in any of the repositories today is the type of study and sample repository that you would actually want to use the hat map is being devised which for a project that really hasn't been formulated in terms of what will be necessary from patient recruitment perspective I mean I don't think there is any cohort out there that would be large enough to be used by hat map at this point if we said we needed 300,000 genotypes very few studies out there would survive a correction for power and but I think it critically underscores that we really need to be thinking about cohorts and we really need to be partnering across NIH I think it's much easier to see how we need to proceed from a societal and NIH perspective than it is from within the walls of NHGRI and I think that's really one of the challenges going forward is how to make sure that the fruits of all of this actually get broadly used across NIH why don't we try to stay on this topic because I see various people have jumped up specifically so Arden I just respond I think that the individual institutes in fact have invested in ordnance very large grants in making collections of appropriate types and yesterday morning when we were talking asked Alan Roses about how many samples would it take to find ApoE by this strategy the ApoE-Alzheimer relationship and he said well if you just look at 30 controls and 30 Alzheimer's you'd see the effect I mean I think for 30 and 30 if you can pick up maybe it's 100 and 100 and 500 and 500 but I think it is not a very large number Kurt would you want to talk on this? Yeah Okay I was just going to say in answer to arts requests for a quantifiable goal for the second floor I think you could have something for simple single gene disorders if as we heard only 10% of the OMIM diseases have identified genes or of linkage we could set a goal that in five years will double that I think there are quantifiable goals that are realistic and attainable with regard to single gene disorders that may be a little easier to define than for the complex trade Neal I'm assuming you want to comment on this right Neal Rich from Stanford I'm getting the feeling some of the problem here is that there was deliberate vagueness in this document perhaps because it wasn't complete consistency of what people were saying and so there's some that's why there was like maybe a lack of commitment to make very specific statements so this is I guess the devils in the details and one question I had for example was we're changing comprehensive to genome-wide now but still the comprehensive description of heritable variation in population there's a lot of nuance in that statement first of all the definition of population what is the definition of population how many populations which populations this is very important second of all is there going to be an equal effort coding and non-coding regions is it going to be equal depth in coding and non-coding regions because there's a lot of disagreement potentially in that area of how much effort and again if there's a fixed budget then there's a question of how much effort to allocate to different to different projects and so that's what I'm saying I think this is this is the issue I'm particularly interested in but I think there was vagueness perhaps I don't know you're going to have to say but because there was not consistency of opinion about exactly what roots to take right no I think you raise a really important point and I think there was a deliberate choice not to decide those issues in this document but instead to let them content and that if in fact this time goes on coding variation is the best thing to invest in because the return is high or we've saturated on coding and on the return from coding or we've saturated on the results of looking at 40 populations so I think this notion of genome-wide or comprehensive had a sense of having squeezed the valuable information on having saturated and we've just got to monitor that over time I think it would be a mistake for us to figure to think that we could predeside those questions so it is to get a comprehensive picture of variation the best way to do that you know that will be grant study sections continuing meetings of things I was going to say that genome has actually a long history of setting goals without specifying mechanisms of getting there I mean when I think back to the discussions about how we were going to do sequencing I think it would have been a dreadful mistake 10 years ago to say we're going to do it by this methodology as opposed to other methodologies and moreover genome has a long history of taking advantage of opportunities since they arise to make new insights from the LC side we have done a lot of work with CF carrier testing with predictive test predisposition testing for cancer and now for hemochromatosis those could not have been specified ex ante but when they came up and developed good models to explore cross cutting issues they were exploited I think it would be a dreadful mistake for us to say here that the right model system is asthma or deafness or whatever I think that that requires more work as we figure out what the methodologies are but it would I think it would be hubris and also narrowing our vision too much to say exactly what the methodology is and exactly what the right model project is because we don't know that yet Roger This is a question about the roof Yeah, introduce yourself I'm Roger Brent from the Molecular Sciences Institute This is about the roof and it may be obvious I buy the distinction I like it between foundational research and policy related research I'm at least a little bit concerned about the use of the word research in this context and the connotations that that has if Margaret Mead visits American Samoa and wants to investigate issues involving adolescent girlhood that might be universal to human cultures sure she does research she's taught by her professors how to run around with a notebook it's research if Alexis DiTope visits the young American Republic and runs around and observes things and says sees policy issues confronting the young Republic he's just running around and then he writes a book he thinks about things Research may connote too much in this context questionnaires and things that could be practiced in academic medical centers I'd suggest that you might consider broadening it to say research and scholarship and the thing about scholarship is compared to research is that it's really cheap you get a tremendous bang for the buck you can get a scholar going for a year and it costs nothing by NIH terms and if the person writes an influential book you won I'll just respond that I think we have a definition of elation here I would not limit research to a merit data collection I have no problem personally with using the terms research and scholarship but I would really make a strong point that research can include activities other than empiric data collection that would be my view I'd welcome others' comments Yeah, Aziz Sachadina from University of Virginia I'm going to raise a question about the roof again and I think I find a conspicuous absence of resources in LC and policy as if there are no resources or they are not important enough but I think our public success of the project depends upon listening to the different communities we are a multi-faith and multicultural community our society is rich in plurality and we need to somehow identify the resources that are necessary we are somehow surrogating the entire work to secular bioethicists or ethicists perhaps who are secular in their training thinking that that secular ethics will have a universal appeal and we'll be able to resolve the questions for the entire public in all its variations and differences quite to the contrary I think the acceptance of the project and the fear that is to their dominant in the public can be removed by these resources that we can identify in different communities in the country at the moment around the world because we need to communicate these ideas at the level whereby it is acceptable to the religious sensibilities of the people after all disease and health has a lot to do with religion has a lot to do with cultural understandings of these and if we do not identify the resources I think we'll not be able to reach out I think the roof metaphor is important to keep in mind that it really provides with what I call the cover for the entire project otherwise it is exposed to all kinds of criticism from the public I appreciate the comment very much we debated whether we should make an explicit point of inclusion of points of view of wide varieties of points of view and maybe we should have made that more explicit in our slides it was certainly implicit in our thinking we actually felt that was a completely cross cutting issue that is at each floor the inclusion of diverse communities in the efforts was extremely important the inclusion of diverse points of view in discussions in the small groups I think there was a lot of endorsement of the need to incorporate different world views and I would not say any intention to limit things to a narrow secular bioethics perspective but to be inclusive again other comments welcome Jeff Dewick South San Francisco so two comments one is really about the roof there's a tremendous framework there to think about how this new science will impact the interfaces between society and society as well as the impact on society itself but it fails to look inside to see that one of the biggest change the last 10 years is the fundamental change inside the scientific enterprise so I think we really need to consider how do we educate people for the future how do we enable them it's the intellectual capital question how do we enable them have a career in a scientific establishment that is increasingly dependent upon cross disciplinary research in interaction which looks very different from traditional academic departments and how do we and how do we set up the operational organizational modalities to do that furthermore given the size and scales of these operations we got to really think about our funding mechanisms both in the point of view how do we get people the flexibility to react to change uncoupling from the normal grant cycles in an era where technology platforms flip every three years you know technology is both indispensable and disposable how do we deal with that so there's no discussion of the inside there's a lot of discussion of the outside and I think fundamentally that's going to be one of the biggest issues if we want to maintain this so that's that's probably the bigger comment the other one has to do with the probably the first foundation in the chemical libraries so this is really a comment if you buy a car you better put gas in it and my point is going to be is it's too limiting to say chemical libraries I mean what we're really talking about is generating tool compounds so we don't want to limit it just to small molecules we want to think about aptomers and antibodies and things like that but it's pretty naive to think you're going to actually get something that's useful from simply screening a library that's like thinking you're going to make an antibody by just buying rapids in a few cages so I think if you're going to do the enterprise that not only you have to include access to this diversity whether it's small molecules or aptomers or large molecules but you need chemistry resources molecular pharmacology all these other little things so make sure that we don't basically get what we wish for and what we get we can't use Joe McNurney from Nitchpeg in Baltimore I want to pick up on a comment about the outside I was struck in the discussion of the roof that the education seemed to focus on what were defined as the foundational and policy issues I hope that does not imply that the the genome program will back off from a commitment to education about the science itself which is an extremely important piece of what we have to do to build the roof unless we educate the public and a broad variety of other individuals about the underlying science the discussions and formulation of policy is going to be scientifically bereft there was no intention to back away from that No, you know I think I think if there was any uncertainty was actually which floor to emphasize that point you could see that Ellen had that in her floor but I don't disagree and I actually think that's a cross-cutting issue that's an educational mission shared by all three floors So David Bentley again I want to pick up on a cease and and why is previous discussion on consulting with the wider issues of other communities and other beliefs and so on I think clearly we move towards an area of consultation in order to obtain a collaboration in order to develop resources or whatever community engagement for samples in far countries is a good example of that I feel we could take it further internationally as well this is not just an ensure I comment but it is potentially genomics is a good platform to do it to actually in advance to really engage communities and consult and learn more about the relevant needs of each other without actually demanding something in return straight away so not as needed but to develop it as part of the discipline and I think genomics and genetics and knowledge of self is actually one of the best platforms of which to discuss with remote communities with very different beliefs and perhaps NHGRI could consider that Tim Leshen with the NHGRI Well I just had a point of clarification you talked about research in the ELSI context and research in the policy context but the NHGRI has also done well we're not a policy making body we sort of helped to influence policy and I was wondering how that sort of realm would factor in your plan Yeah I'd actually like to defer to to people from NHGRI on that my understanding is that I'll just communicate my understanding that there may be a very important role for promoting certain policy positions that the government has taken and NHGRI may have a very important role in that but in terms of an ongoing plan for what NHGRI should in a perpetuating way be doing in terms of policy we saw the focus on policy options development of policy options I'd like to respond to that I think all of us who have done this work for the last five years have found the clinical environment increasingly difficult and NHGRI I think needs to be playing a more proactive role in the future to ensure that all of the tools that we've been developing for the last 12 years don't become useless because of unassailable obstacles to actually doing the studies that we created this foundation to do Here That's a statement about something that we're doing Well, we seem to have spent quite a lot of time on the foundation which is of course fundamental that's why it's called a foundation and then we've jumped up to the third floor and we've skipped over the second floor which was just my experience yesterday and I partly came here for the second floor and I think I just you know want to make this simple statement that the public out there which people talk about all the time, is deeply invested in this component of genomics. And I kept even hearing that this isn't genomics yesterday. I think that there's going to be a huge credibility problem if the Genome Institute doesn't work really hard on figuring out ways to take the enormous richness of genomics and translate it up to understanding disease. And that's really all I wanted to say. It doesn't even matter which diseases, single genes are probably a good place to begin. But diseases in general have to be addressed over the next five years. And what I say over and over is it isn't going to be five years. This is, it's going to be 10 years, 20 years, even 50 years. Over that time period, it probably will happen and it absolutely must happen. Let me endorse your sentiment. I think the public supports this project in part because they really do believe they're going to be major deliverables here. And I approach this from the perspective of a general pediatrician who's going to be using these technologies. I do think that, and I do think that this plan does endorse the use of these technologies to really move forward in developing deliverables that are going to be accessible not only in the research setting but also in the clinical setting. I think it's going to be really tough and we were trying to talk about that. And I think that, that I think that this is an area where Genome needs to be very strategic about picking, picking problems that are really going to advance our understanding in these areas and really set up models that others can possibly use to explore other areas. But I would not even back away for a second from saying that at the end of the day, this is about primarily about making people, giving people the opportunity to be healthier. I mean, that's why we're doing this, I think. So certainly why I'm doing it. David Barker from Aluminum. I'd like to second that too. And also my comment is it seems to me there's a difference in specificity between the goals of the first and second floors and of the third floor. I'm just wondering if that was intentional. On the first floor, and I'm paraphrasing now. First floor, we have a comprehensive and comprehensible description of the parts list. And on the second floor, we have identify all genes and pathways that contribute to better, to health and disease, et cetera. And on the third floor, we have identify and address the pressing societal implications of genomics instead of the pressing and societal implications of genomics are what is the new impact on the concept of race and ethnicity, the concept of what is normal, the impact on culture, family and self, the change in education that must result from doing big biology, things like this. I'm just wondering if that's a difference in specificity that was intentional. I don't think, I mean, my impression is that all we're dealing with is cartoons that can be put up in a quick 10 minutes and spoken about, I think there's going to be a similar level of specificity in the ultimate document. Is that a fair, okay. Yes, I would like to come back to the metaphor that we have. Identify yourself. I'm John Miller from Northwestern University Medical School. It looks like to me that the new architecture is sort of a modest country cottage. And I'm afraid that the third floor is where the attic strikes me as the walls go in and it seems to have a certain characteristic to it. What I'd like to do is suggest that there are other kinds of architectural metaphors that may work very well also. I happen to currently live in a 34-story glass tower, which also is a house. And one of the things you could do is to see some of these as floors. And in fact, what we might do is be creative and see the public education side as glass walls because what we need to do, as I think Joe was saying earlier, is we need to have the public looking in all three floors. And we can be creative about doors and walls in a way to explain what we want to do that doesn't limit us to sticking social implications in the attic. And I'm a little afraid that the current metaphor puts social concerns away from the main part of the house. Would penthouse suite feel better for you? Well, a nice big hot tub, white bar, roof garden. I think when it comes to appropriations and fighting stem cell battles, you need to hold high ground. So somewhere up there, I would think that public understanding should be maybe seen as walls as opposed to penthouses. Hi, I'm Wendy Olman from the University of Michigan, and I actually also want to address the metaphor. I actually like the house idea because after all houses can be renovated and you can build additions. So I think this could actually service quite well in the long term. And I just wanted to convey the point that I actually think that education and workforce should actually be the electricity and the lighting in the house because if you don't have the workforce that's genetically educated to be able to provide the services or if you don't have education of the public and healthcare professionals, we're never going to see the furniture and we're never going to see the appliances and all the tools that are being created by the human genome project. So I do have to point out the three G's work very hard to get cross-cutting demographics and the attendees for this meeting. We forgot to invite an architect. We let you down. So I'd like to return to the concept of the proper level of specificity. Neil said, and I think he's right, the devil's in the details and making sure that we capture enough specificity. And when we said earlier, and this has come up throughout the two days, five-year plan, the answer is, oh, it's not a five-year plan, it's a vision statement or something like that. And I guess I would challenge at the risk of annoying the people at the front of the room, why that's a good decision? Because it seems to me that, no, I'm not done yet, Eric. That was rising to the challenge. Yes, sorry. No, because I think that the danger of a vision statement, it's not that we shouldn't have vision statements, the preamble or there might be places for vision, but I think that what I've heard over and over again in the last two days is the desire for tangible goals and you can't predict beyond the next few years. So clearly if you go, in other words, in a sense, if you take a vision beyond five years, well, and clearly you can't specify what you do, but if you were to specify the next five years, you can actually say things, I can example for, and I'll give something concrete for floor two. We might not be able to find, we're not gonna find the next five years all the genes for common disease, no, it suggests that. But there are a lot of pointed questions in the community about, and again, Neil raised one, it's a good question, how much we look in coding regions versus haplotype approaches. I could imagine if you wanna find the disease genes, what's the utility of founder populations, admixture mapping, drift mapping, cosmopolitan populations, family-based, you could have a program that would actually, and I'm sure that's included in what's considered, where you wouldn't be doing what every other institute's gonna do, because probably other institutes are not gonna have programs to try and explore the power, utility, et cetera. If you go beyond five years, you're probably gonna figure it all that out so you don't need it. I'm not sure why you wouldn't wanna specify those things. All right, Eric, you can talk now. I think it's a question of what this document is the proper analog to. I don't think this is the analog to the last five-year plan and the last five-year plan. I think this is the analog roughly to the Albert's report. The Albert's report was not a five-year plan. The Albert's report was a 15-year plan. It's not a vision. There's nothing vision about it. This has to have deliverable items. But the five-year time horizon, I think is not the appropriate time horizon for us to lay out here. Now once a 15-year time horizon was laid out, maybe Francis thinks this should be 10 or 12, but longer than five, once that is laid out, there is every reason to sit down and say, now what are we doing in the next five years? And post Albert's committee report, there was, we're gonna really focus on genetic maps. We're gonna try to figure out how to do physical maps and we're not gonna do a whole lot of sequencing right now because we don't know how to do it. So we'll develop on technologies. So it did reduce to exactly what you're saying, but I think we should seize the opportunity now to lay a longer than five-year plan out and then fill in specific. So I think that's fine, I'll just finish. I think that's fine and probably a need for it, but I just hear over and over and over again in every discussion I've been a part of, the desire for the tangibility, they can't be there if you go beyond five years because we obviously don't know what we need to do beyond those. So some way to bring that back in. Yeah, and a lot of the discussion last night was of course in reaction to a feeling that what was there was such tangibility without the clear sense of where we're going. I don't think there's any problem with melding those two. I think this is a framework into which now to put the tangibles for the next five years. Janet. Janet Raleigh. I'd like to pick up on Bob Lipton's, Richard Lipton's comment because many things that we talk about using large cohorts, et cetera, will be absolutely impossible with some of the rules that are presently in place for the use of material from human subjects. And if we don't understand that now and if we don't begin to work with other agencies, mainly government agencies, that are establishing these rules and continually making them more restrictive, we'll have great opportunities, we'll have the tools and the samples, we won't be able to use them. And I speak also from the standpoint of somebody who's interested in particularly leukemia but cancer, the inability often to connect the genomic changes in patient samples with the response of that patient to treatment and the outcome is a tragedy and it's enforced by these rulemaking organizations. And we have to deal with that upfront or you're having 300,000 samples, isn't gonna do you any good because you can't use a single one for the research you wanna do. I have a question about a wording on the third floor. Your first goal is to explore the impact of genomics on X, Y, and Z. Do you really mean genomics, which I view as a way of doing business as opposed to genotype or genes or something else? Yeah, I think we could phrase it as impact of genomic research or maybe genomic information but I wouldn't limit it, genotype might be a broad enough term but I think it's the impact of genomic research. Paul Miller, I'm a commissioner of the US Equal Employment Opportunity Commission. Just a comment, I'm back on the third floor. One community when we all talk about sort of the communities that we need to reach out to usually race and sort of generically disadvantaged folks. One community that I think is really critical that never gets mentioned but I think is always in the room and in fact I don't think is ever was mentioned in the original draft of the document are disabled people, people with disabilities and I think that you really have to tease that out and specifically deal with the disability community on LC issues because this revolution impacts disabled people greatly and if you don't bring the disability community along it's gonna be harder to move the revolution forward and there is tremendous concern and fear based upon a backdrop of eugenics and all sorts of identity issues and I think that there's a lot there in these LC issues that we need to think about so I would really encourage you that the community is gonna be looking for themselves in any document you do and if they are not explicitly mentioned they will feel particularly excluded and that will ultimately be a problem. Scott Ramsey from the Fred Hutchinson. A couple of suggestions just to throw out. The first relates to the second pillar the genomics to health or it's not a pillar now is if it's a floor, genomics to health. I think one of the things to make this real and concrete we need to have highly targeted translational projects that the genome project can identify and hold out preferably with a snappy acronym somewhat like the SPOR programs that NCI has where they're targeted for rapid translation from biology to the bedside. So I would think about formulating highly targeted programs like that to make the second pillar really concrete. The second suggestion sort of relates to the clarity issue and I think I underestimated the importance of this process before I came here and I also underestimated the breadth of the audience that I think is ultimately gonna be reading this document. So my slightly tongue in cheek plea is before this gets committed to print, call your mother. Now if your mother's a molecular biologist you have to call your sister. But try to explain this to someone who isn't in this room or someone who isn't have this level of sophistication. I think to the degree that they can understand that it will be or not understand that it will be telling and important as you think about shaping this document. Okay, see no one else queued up at the mic I think now is the time for several announcements and then we're gonna take a break. First announcement I've been asked to make is that in the back of the room on the tables are some copies of advertisements that are going out. The NHGRI program office is, extramarial office is doing a recruitment for program staff in several areas. LC research, genetic variation and translational research. And anybody grab those if you know of people who might be suitable to apply for that. Second announcement is that checkout is at 1130. So if you have not checked out you might wanna quickly go check out cause there is another group coming in and they need our rooms and all the usual drill. So 1130 you need to check out by. Third announcement is we will break now for coffee, et cetera. We will reconvene sharply at 1015. And we're doing fine on time but let's please reconvene for the next series of talks. Thank you. So as many of you know and participated in last evening we convened six special topics working groups to discuss an array of different topics. And so we're gonna now hear very brief summaries from each of these, from leaders of each of these groups. We're asking that each of these group leaders please spend no more than five minutes summarizing. We will have five minutes of discussion and we'll keep moving because we've got six of them to get through and we still have some additional talks. So the first one is Barbara Wald who will be talking about a special topic entitled NHGRI role relative to others. This isn't gonna clip. So this will indeed be. He can do it. This will indeed be brief. The topic that our breakout group addressed is something that we'd all been talking about all day. So in some sense this is a summary minus many of the things that were already talked about or addressed in reversioning the plan this morning. The first thing that we recognized in thinking about the relationship of NHGRI's commitments in genomics and its interest in genomics as stated in the plan relative to other institutes within NIH. And I think we all felt it's time to acknowledge that this is a good thing. There are fuzzy boundaries between the institutes in terms of their interest in genomics and that's part of the maturation of NHGRI from Genome Project to Genome Institute. Fuzzy boundaries are part of the NIH in general. And so here are the fuzzy boundaries and of course not all the institutes are here. If you happen to send a representative to our panel you're more likely to have your circle up here. And you can see NHGRI is placed in the way that we would all want it dead center. But you can also see that it's almost getting squished out and so obviously the tension here is for NHGRI to have those central things that we're listing as our main goals and our main mission that probably are things that other institutes would cheer for and like to see the output from but that they would not be so likely to pursue. On the other hand, particularly with time there's going to be a shift in the application of comprehensive and global kinds of technologies that may be generated first at NHGRI and come to be applied all over the NIH in the interest of various disease systems, body parts and other appropriate entities. It's obvious that for scientific reasons as well as budgetary reasons and cultural reasons we should simply say in this report that that is the case and that it's a good thing. Which may seem too much like what David Botstein calls Applehood, which is mom and apple pie and all of that. But there are places for Applehood and I think this might be one of them. There was the explicit suggestion that with respect to this plan as we write it that NHGRI staff which of course has infinite time should indeed do a little work on the ground between now and the time of the writing to really talk about some of the things that are off in the list of things that had I believe pound signs on them in our world of asterisks and tildes and other kinds of things. Those things that are good partnering things and good sharing things. And the suggestion comes from other institutes. She, why don't we chat a little bit about which ones to highlight on the list before the fact and that seems like a good idea and to explicitly acknowledge the interests of other institutes and missions that they've already identified that are genomic. Of course the problem is that the report's not going to be infinitely long. We're encouraging it to be shorter and explicit statement as in other areas risks failure to be utterly inclusive. So that's a hot rock that we're happy to toss to NHGRI staff and what you deal with. This is very short, just a few more points. It's clear that over time the activities that belong initially at NHGRI are going to move out and become shared or even no longer belong at NHGRI. We all can think of things that are evolving already in that direction, the sequencing of many pathogen genomes for example. And that's exactly how things should be. We do foresee the ongoing mission which we have discussed at length and in other ways for conceiving and launching global projects at NHGRI. And so it's just kind of an endless conveyor belt of new global projects that come out of NHGRI. Some as demonstration projects, some as the whole enchilada to deliver data that other people work from. Then we get particularly, we discuss the issues that were part of the old pillar three but that have to do with ELSI. And just to put a punctuation point if one could be more needed after Dr. Raleigh's passionate comments, there's the need, an urgent need for timely address of certain of these issues, especially as regards human samples and connection to clinical covariate information and outcomes. And this just has to be dealt with or our floor two is going nowhere. And that would be so tragic. And I think this document, and I'm reflecting what I believe the group, the general sense of the group, that this document has to be very bold in saying that. And it's also just a mission for NHGRI in all the other venues. I'm sure it's being lobbied for, but to try and get some real energy behind this. Now this may be a bit more controversial because obviously at some point we come down to trying to parse how do we distribute budgets that are not infinite. And in the way of our interactions with society in general and explicitly those things that are propelled by the genome ELSI program, the question is how to parse all of these good things that reach out very far and include how we relate to society in many ways having to do with anything genetic against a non-infinite budget. And so there is at least some sentiment though I think there may not be consensus that a real effort should be made to parse this so that at least these urgent things that are intrinsic to the pursuit of genome research as we're doing it in these five years be addressed and addressed well first if you have to make hard decisions between good projects. And that's not so much for the report but it certainly was part of a fairly animated discussion. It's sort of beware the dilution of the roof. And I think we've talked about dilution of the effort of NHGRI away from its very core missions with respect to the first floor and the second floor perhaps a bit more than the roof. And it's partly because we know we have to impact society or we hope we do in enormously diverse ways. And yet for focusing the research efforts something is going to have to be done. We discussed genome relationships with a private sector and one thing that I think in the new preamble and in sort of being positive in the document overall there has been the suggestion of pointing to the foundation and the springboard of past accomplishments and drawing on those successes in part for inspiration and for sort of the style of NHGRI. And that explicitly with respect to our relationships with a private sector there are some successes that you could wish to repeat and extend not necessarily to discuss specifically in the document but maybe to refer to one or two. The SNP consortium is certainly I think widely regarded as a success in things that subsequently flowed from it. Distribution of genomic reagents, having private companies supply libraries and act as a distribution agents works pretty well. Contracting for commodity services including in some cases sequencing has worked very nicely and is something that you'd want to do more of going into the future. Clearly the genome projects interactions with the private sector haven't been all love and kisses and gee that's a revelation. And then from these past difficulties we have probably learned a few things. Eric is saying no, it was all love and kisses. But to go back to our roots clearly one of the things that might be worth restating in a document like this just because it will be read very widely and because people will begin to wonder how the principles of public open dissemination of data and even instantaneous dissemination of data is going to play as we move forward from a project and a subject to being a discipline as we move forward from being an activity of sequencing model genomes and the human genome into being a full blown institute in the truest sense not just the official sense. So as we do that, this open access to data is an important part of our motivation. On the other hand it's probably not gonna be universal and addressing this issue and again the discussion went a little bit beyond what should be in this document of what people's expectations should be is a bit of a conundrum. And there are things the private sector does better. We should seek them out in practice and do them that probably doesn't belong in the document. Something else that came out of our discussion was in general, just for heaven's sakes, be more positive when you're talking about what is now our roof. There was a sense of defensiveness and so this had only to do with the fact that we had people back in the room and relatively little to do with our charge on crossover and in promise to keep it short, that's it. I think we may have talked this one to death. Okay, we'll move on then. Check right on roll. Jane Rogers will give a summary of a special topic maintaining high standards and institutional coordination in genomics. Put it on one. Okay, thank you. So the charge of our special topic group was to look at maintaining high standards and international coordination in genomics and to remind everyone of the benefits of international cooperation and coordination. We adjourned our meeting to the pub and held it there and discussion flowed freely. So despite a little noise from the karaoke people next door, I think we sort of actually managed to come up with some points of discussion here. So the Human Genome Project required extensive collaboration and coordination, not just internationally but it was unusual in requiring international coordination from the outset and agreement was needed on both standards and standards of the quality of the product that we produced and those standards changed throughout the course of the project, you may recall, from working at an early, very early agreement in 1996 that the quality of finished sequence should be no less than one error in 10,000 bases to defining a quality for the working draft and now going back to looking very hard at the quality standards for the finished genome. In thinking about how to go forward, we recognize that genome sequencing center capacity has grown and we no longer need the size of consortia that we have for the Human Genome Project. Projects can be undertaken by fewer centers but it is important to recognize that groups in other parts of the world are also interested in these projects and to work with them wherever is possible and certainly to everyone's advantage to do so. So we don't have to look at the lessons for future projects that do require collaboration. There could be projects that require participation by many, for example, the current efforts on annotation of the human genome, for example, or simply coordination over standards for a general data type generated by many different groups and this is particularly apt in the context of the plan as it's now being formulated to generate information about the whole working genome. And in many cases, coordination at some level is already occurring, this should be built on. It's important to ensure that it continues and is fostered as new large data sets emerge. So some particular lessons that came out of the HGP, standards are important, especially in coordination but also getting a product out to the community that's usable. Data release and ready access to that data, Barbara already mentioned, they're obvious imperatives. Data release as early as possible and the HGP did set the precedent of releasing data in the form of 1KB assemblies, Nikely, which set an excellent precedent and interestingly it's now something everybody seems to accept and I'm sure that there will be pressure for other data sets to be released in a timely fashion. By and large, groups working on a common problem figure out how they need to coordinate. They don't really need a top-down approach to tell them but local conditions, for example, people with different funding agencies, scientific culture and the size of the group do influence the outcomes. But it was felt that perhaps the most central coordinating factor would be the establishment of central databases for the data that's generated in different projects and they are a natural foci for coordination of policies and this can be reinforced by the journals encouraging or actually demanding deposition of data in central databases when the data are actually published. The other feeling was that there's been a new model for coordination about standards and data release was a success and I should perhaps explain that from the outset the Bermuda meetings were established to bring together groups throughout the world who were going to be involved or interested in getting funding to be involved in sequencing the human genome. There was discussion from the outset between those groups about how the project was going to be carried out, how they would operate in the different countries and how we could keep the coordination going and in general that was a success and we feel that there is a product that's going to be very useful coming out of this. So specifically now looking at the continuing rule specifically for NHGRI, for large scale sequencing continued coordination is needed among the data producers and the funding agencies in particular to avoid redundancy. As a first effort along these lines a meeting was held at Constantine Harbour this year coordinated by the Wellcome Trust and NHGRI to get together sequencers from different countries to in the first instance establish a database of projects that people were working on. This is still in the process of being finalised and I believe a website will be set up at NHGRI where a list of organisms that are now being sequenced or perhaps they will be sequenced will be posted so there is information for everyone about what's being done and how they can access the data. There is an urgent need to clarify policies about pre-publication sequence data release. There is an ongoing discussion which I believe recently have a contribution published in Science this week I believe about the first publication following the production of a dataset. One of the major aims in releasing data as quickly as possible that was agreed at the outset of the HDP were so that the data are available as widely as possible but there is a discussion about who should be reporting on the quality of the dataset and that will be taken up in a meeting that's going to be coordinated by the Wellcome Trust in January so I hope some very clear policy guidelines will emerge from that meeting. I should say that's another international meeting and that policy guidelines will then be taken back to individual institutes in their own countries. Finally, plans need to be made to anticipate the future types of data that will need repositories. With David Housler and you in my group there was obviously quite a lot of discussion about databases that got very free flowing but I think the most firm comment to come through this was that there obviously going to be a lot of data generated that we do need central databases that have good access and some account needs to be taken now and plans need to be made for the burgeoning increase in sequence genomes, haphazard type data, and all the other types of data that are going to emerge. I think that was what conclusion. Questions or comments? I'll raise one. I don't know if you're going to talk about it. One thing I have heard people comment on I mean actually non-genomics experts is that they often feel bewildered when they go to the database and they look for a sequence and they just don't understand and don't really... the data quality of that in part because their shotgun projects being done at different levels of data quality and everything you're talking about with Bermuda there was an organizing principle behind it that that was to sequence the human genome. What I'm concerned about is that with time and without the focus of human and with having sequencing being more available to lots of different groups there will be a lot of confusion between the biological biologists and counter sequence in genomic and not really knowing the quality of that sequence what's the engine underneath it in terms of data quality and so forth. I don't know if any of your group talked about that. I think we touched on it very briefly and the feeling was that this came back to the funding agencies when projects are funded at the outset they should be funded to achieve a sequence of a certain quality. We certainly need to do more in terms of educating communities but my point is I think you're going to see the funding base may get a little bit broader it won't just be NHGRI, NIH it won't just be NIH and so forth. Okay. Next Alan Williamson will summarize a discussion on public-private relationships in genomics. That was our title, that was our brief. You will see from Barbara's presentation that it was also subsumed in a sense within the first one which said interactions with all others and we were rather specific. Sitting here this morning listening to the building of the house I realized that with all the omics we were lacking a contribution to acronomics in this meeting and so my contribution is that this is the house that Jack built where Jack stands for joint action and collaborative know-how. Bill Gelbart told me always think of the acronym first and then see what you can fit to it. So I think that's appropriate actually because Jack in this case is a set of partners who are going to be involved in building this house and the specific brief that our workshop had to discuss was the partnerships between the public and the private enterprise systems. We took this in sort of three ways the opportunities for new partnerships we reviewed the successes and there are a number of good successes from public-private partnerships the failures in relationships were not partnerships as I would call them they were anti-partnerships rather than partnerships. I think wherever true partnerships have been attempted they've worked pretty well but they've got to be put together in the right way. We discussed the barriers that might face us in setting these up and I should forestall that and say that I'll come to that at the major one at the end but consumed a lot of time and that was IP issues and we discussed the ways in which they can be facilitated to put them together. Let's start so that we can put those other things in context but what did we come out with that might be possible proposed private partnerships works out differently. So these are our PPPPs proposed private partnerships. The first one was screening of targets arising from the study of Mendelian single gene diseases. Now someone said this morning only 10% of those Mendelian diseases have an assigned gene at the moment and their proposal was made from the floor that we should try to double that. Well, that's one possibility but I would also say that when we were discussing this last night I mean we came to the conclusion that few of any of these 10% have what you would call a good target gene for therapeutic intervention. We may know the causative gene but it may not be a target gene and I think that's going to be the case for most genetic diseases that the gene or genes where there are many of them in complex diseases will probably not turn out to be good drug targets. So maybe a partnership like this might attempt to come up with at least in one or two or three a few small number of single gene diseases a feasible target and the challenge then was would large pharma companies not necessarily do the screening because as we've pointed out rightly so most of them, not all but most of them have enough things to fill their screens several times over but would they contribute what they have is which is a unique resource and that is their small molecule collections for screening on these targets on the basis that anything that came out could be developed in some way for these orphan diseases it wouldn't be a commercial entity as is pointed out in one of the preambles in the health section but this might be a way to access now there are clearly issues here and I'm not saying there are any guarantees that anyone would do this but I think it's something that's worth exploring it's sort of in there in the report at the moment that maybe small molecule chemical collections should be available for this I agree with one or two people who said earlier this is a little naive the really valuable collections the really valuable assessments are not going to be made accessible because they're crown jewels but all you get out of them is a hit it isn't a lead compound you've got to turn that lead into a hit into a lead and that requires really good intensive medicinal chemistry so there are a lot of issues in doing a partnership like this but at least it's a concrete approach to some of the things that were reviewed there the second one was phenotyping it clearly is a need to bring phenotyping up to a more scientific level but genotyping is coming up too and it's a more complex thing and it relies upon the more traditional sciences physiology and biochemistry that tend to get swept away in the omics so I think we need to come back to these and we need to address how to do high throughput phenotyping and turn the art of medicine into the science of medicine and define things properly but to do that in the first instance I think we felt that there should be some attempt to phenotype a healthy cohort and this sort of fits in maybe with the healthy aged cohort but maybe it should be a broader cohort maybe it should be something like a framing study which everyone agreed it was an enormous success I mean several different major beneficial therapies came out of data that started in framing them so I think that phenotyping a healthy cohort and then also phenotyping human disease but basically some sort of partnership to develop phenotyping tools that are essential in the healthcare industry and obviously essential in healthcare delivery and they need to be worked together and there's room for some sort of cooperative venture there and fourthly the more success there is the more data there is the more complex data sets there are and everyone is struggling with ways of analyzing this data so basically development of computational methods to analyze large and complex data sets industry felt there's definitely room for some sort of partnership there these things should be explored now how do we facilitate this the point was very strongly made by more than one person that where partnerships have been successful it's been because they've been developed in a joint way from fairly early stages and where they would not work is if a central authority comes along and says N-H-E-R-I that's just specific comes along and says we've got this great idea we have some money for it I mean people do not want to be credited as just funding partners they want to be brought into the discussion help put together the idea and take it forward from the beginning and it's sort of the way Barbara was saying about do some research on the overlaps with the other institutes plan ahead as to how it's done now that's easier because you're all in the same place and you all talk to each other on a regular basis and you think you have to do this some work has to be done to build these bridges out to industry and to potential private partners but I think it's worthwhile because these bullets might not even be the right ones but along these lines try to see what can be done explore out the parameters of them and see what the mutual benefits costs know how that can be brought to us from each side and how to put it together so that's the major message on facilitating that's exploratory planning and include the potential partners all of them around the table as early as possible on the first one the point was very strongly made that one of the advantages of working with single gene diseases is there is frequently a very strong action group there may not be many patients in these diseases but many of them have very strong patient action groups associated with them and that's a group that should certainly be brought to the table because they facilitate access and have a lot to contribute actually and how you might go about this so that's critical and I think it's important that visions and goals should be shared from the beginning and people should understand the different goals the different sides have in this they may share the same vision but the specific goals may be different and those have got to be understood on the table at the outset so then come to barriers I mean the barriers sort of dovetail with the facilitation certainly a barrier is just to ask for money you've got to do it the right way you've got to come together and decide what everyone's going to contribute but the major barrier with IP and it's not necessarily a barrier IP is clearly necessary in many areas it's certainly necessary in the pharmaceutical industry but what we're talking about here is not ownership of IP but access to the elements that are served by that and a lot of analogies were drawn with the software industry where there's a lot of patent pooling where there's open access software there are all sorts of things and models of this sort have been talked about I mean clearly NIH took under Harold's leadership took a great initiative on research tools produced marvelous statements it hasn't had quite the effect I think you hoped it would have but it's helped and we've just got to go on trying to do that and try to make access possible and various bodies are trying to come to bear to rationalize the IP around the world to make access and you've got to realize that the US is in a slightly privileged position by Dole was set up to foster nationalism rather than internationalism and there was a somewhat US-centric view to it sometimes and I think I ought to raise that but it was agreed last night that revising by Dole in some way would not be a bad idea and looking at whether it did it work how was it meant to work what has it done and what has this effect been on international affairs so that really is the I didn't even put the last bullet in but IP is we need to be assured credit access to critical technology and that's one barrier that has to be overcome so take questions Comments? Questions? Speak and do Thanks Eric, Bill Gaul in HDRI I think that moving candidate molecules to lead molecules and also doing the phenotyping has as its rate limiting step not so much the molecules or even the tools but the personnel the workforce, the people who are experts in the disorders and I think that's something that our institute and other institutes can work on in other words training these folks and I would add at a relatively low cost as well compared to some of the other numbers that we're talking about which I think we need to develop mechanisms to do that better Good One thing I heard last night that really seemed like a good idea the success of the SNP consortium can be attributed in part to the fact that Arthur Holden was running it as a CEO who talks to talk walks to walk and is intelligible to other business CEOs not by a government official or a professor or anybody like that but a real honest to gosh CEO it seems that an opportunity exists now for a more ambitious effort led by a CEO type that particular thing is the development of software tools to better understand let's call it functional genomic data each pharma company has a group of informaticians who busily buy software products from outside and patch them together and write their own code reinventing the wheel in how to analyze say gene expression data Silicon Valley is if not awash is well supplied now with idealistic youngish people who would like to work talented computer programmers who would like to work on software that they thought would help people understand biology however they need a management structure they need a something run by a CEO with managers and deputy managers and all those levels that animate programmers there's an opportunity here for consortium to start generating some of this software it could be an international consortium it would have substantial buy-in from business it would need to be run by professional business people but I think it could attract a lot of support including a large number of the people who could actually do the programming and make the tools available to corporations and academic researchers in a sense that was meant to be subsumed within the last bullet and I think having been involved in the birth of this different consortium I would be the first to pay enormous tribute to Arthur and what he's done to run it and I think but I've also been involved with a lot of different bi-tech companies over the last few years and they're good CEOs and they're not so good CEOs so you've got to get a good CEO in there and in this case Arthur had just the right light touch dealing with prima donnas let's say heads of sequencing centers I think they're called and making things happen oh they don't have egos one more question thank you I'm Suzanne Paddy with the Cystic Fibrosis Foundation and I come to this meeting as a patient and as an advocate for the patient community but also I wanted to ask for a little more elaboration from you on what this group discussed last night about bringing in patient advocacy groups beyond just facilitating access one of my previous jobs I was with the biotechnology industry organization and I was responsible for doing outreach with patient advocacy groups to bring them specifically into the biotechnology industry and to encourage earlier collaboration than having the community consulted once a product was developed and access was needed to be facilitated and last week or two I was at a clinical research round table with the Institute of Medicine there was an excellent presentation about the ability of Genentech to work with the breast cancer community to provide them with scientific information on the development of Perceptin and to have the community provide direct scientific input to the development of that product and not just and to help with clinical trial recruitment significantly and not just to the development of access once the product is developed so I wanted to wonder I wondered what you guys got into more last night with this with this activity Let me just ask you has your experience been good or bad in this? I think both and I think I think the biotech community for the most part is realizing that it's important to step out there have been some good other examples like the development of philodomide for other types of uses than just with protections in place and the need to consult the community initially to get that product developed in a safe way so there definitely needs to be more but also I'm looking at around the room today and there isn't a broad representation from the disability for a patient advocacy community here today so I guess I'd like to see that broadened as well I gave access as an example but I did say that there was communities and it was touched upon not extensively but they have a vast amount of community I don't know whether Jeff Dyke is in the room with and it gave that a specific example where it'd been very good in terms of the planning and help and that was sort of a model that we drew upon and agreed that in any in any disease situation if there was a good patient representative group then they ought to be involved in the planning from the early stage okay I think in the interest of time we'll move on next Harold Varmas sorry for the delay so our group had an animated conversation about the role of genomics in the developing world inspired I think by three general ideas first that I think we all agreed that genomics has a lot to offer as a foundation science for improving both economic and health conditions in the developing world secondly the culture of the genomic enterprise the human genome project has been one that has been global and our character sharing has been a major component of it and it is inspired a certain sense that we are one world and we share a genome and that the concerns of parts of the world that have not traditionally played a major role in the advanced economies as they do science do deserve such a role and thirdly the excitement that surrounded the announcement of the human genome quasi-completion has been a way of generating interest in science and inspiring efforts to develop science in those parts of the world and we tried to focus our conversation around three broad questions a lot more could have been asked but first we wanted to talk about the ways the specific ways in which genomic sciences not all sciences or even the sciences that build in genomics but genomic sciences per se could help developing countries secondly what would be needed to facilitate the introduction of those sciences into the developing countries we take make the assumption that science will be done in advanced parts of the world but if you were to do this what's needed to make this science a way of building the long needed pockets of science throughout the countries that have been having trouble establishing it in the past and then of course what should our hosts role be in supporting genomics in those countries so we had an animated discussion about the kinds of things that could be done and I think that's the purview of the NIH a lot of attention given to nutrition and agriculture and the idea of sequencing useful genomes particularly cereals the foods that provide 70% of the nutrition in many parts of the world rice almost done corn sorghum others and the idea here is to improve plant breeding to facilitate genetic modification there are other organisms that might be subject to the study of genomes including organisms that infect plants and other agricultural organisms including animals organisms that infect human beings and cause disease in those parts of the world the vectors that transmit those agents then sequencing genomes of exotic living forms that might be a way of enriching the the the economies of these nations and then of course building on all that to encourage useful benefits one example of what this all means embodied in the example of malaria a good target for global genomics global in this in this in this talk David doesn't mean genome wide it means the whole world and and obviously this is a useful example to think about obvious health implications very strong economic implications that build upon the need to develop healthier economies many nations don't invest in certain parts other employees will get malaria if they go there the genomes that that make this disease possible have been nearly sequenced an all reported Plasmodium a couple of species have been done Anopheles the vector the target for human disease there are many strategies for use of that information and there is unlike the situation for most diseases of the developing world strong advocacy some of it emerging from the NIH and other funding the multilateral initiative on malaria but the other advocacy groups fostered by the World Health Organization and others now I think all of us in the room believe that the NIH and other funders should pay more attention to projects in the domain of genomic science that could help economies and health in the developing world those projects could obviously be done in the advanced countries and many of them are being done there but they are being played although our conversation didn't go in that direction for doing more of it but if it's going to be done in the developing world there's some concerns to address these are in a sense boilerplate concerns important valid and so forth but they emerge any time one talks about doing more science in the developing world training of scientists who are going to work in those places especially from local populations funding of the science naturally collaboration of the scientists in the developed world obviously important when you go to these places you learn that political support for these endeavors is crucial partnerships you've just heard about and the NGOs from Bill Gates to advocacy groups for specific diseases play a major role here industry has a role to play it's too cumbersome to try to explore in the five minutes I have been assigned one needs ways to attract scientists from those countries back to the countries and that means having laboratories and domestic situations education for their kids that allow them to return it's helpful to make the science cheaper I'll come back to that point in a moment the client on the cost of sequencing would help attention to many social and legal issues of course is crucial and some of these especially the intellectual property and attitudes toward genetics and genetically modified organisms is absolutely fundamental to trying to pursue genomic sciences in these countries a very important feature that was emphasized by many around our discussion table is the importance of being connected by email and by access to the scientific literature some organizations including the NIH are to be commended for efforts like PubMed and PubMed Central that help Biomed Central a private organization that makes global free access some journals have made some of their publications available to the poorest countries in the world that's to be commended as well but clearly we're very deficient in getting the full repertoire of what we do paid for by NIH and many other organizations two parts of the world that could do science if some of these conditions were met and what kinds of specific things need to be done to have developing science have genomic science and for that matter other kinds of science as well done in developing countries we've all recognized here in many times the importance of genome sequencing centers and if one is going to do genomic science and things related to it maybe not simply sequencing but genotyping of insects and other kinds of things you need centers especially in countries where the scientific expertise is slim and widely distributed you need to bring it together into a coherent place where people can talk to each other an example in Bamako Mali a center from malaria research and training surrounded by the one proponent of the NIH not the one we're here paying our bills today but another institute AID in collaboration with US AID and this is the kind of thing that's done the genomics there at this point is simply looking at chromosomes and doing karyotyping on some genetic marker studies on insects isolated in different parts of Mali a second issue is getting more scientists from advanced countries into these places not just as helicopter scientists but people going in and staying for protected periods doing some training and learning the culture an example of what that might mean is provided by the history of genetics the predecessor to genomics many of you probably don't know that JBS Haldane one of the founders of our field spent the last nine years of his life doing corn genetics and other forms of genetics and archeology and anthropology as well in India with his wife who was also a geneticist so this is a way for superannuated scientists who no longer can get their R01s to practice their their wares in a very beneficial and stimulating situation some of us we're doing the same experiment over and over again we need to change the pace we need to focus on a different kind of disease we might do better work there are other things like developing low cost technology I've mentioned and specific aims to do that could be incorporated into the plan that you're developing here making more international workshops some held in developing countries to stimulate work of the kind we're talking about and free access publishing and archiving needs to be much more widespread and there are efforts like the public library of science that many of you are familiar with that's an effort to not only advocate for but develop the free access journals so what should the role of the genome institute be well, two of these are sort of Applehood one is trying to increase activity in these domains that all institutes by persuasiveness working through Dr. Z and others and working with the Fugitive International Center which has this as an explicit goal focusing attention on the LC problems which are going to be fundamental to doing research in many of the countries we're talking about two specific things came up in our discussion I thought were particularly interesting one is to advertise grants for developing for example genotyping methods that would be particularly cheap and facile in countries where they could be used for instance we're doing genetic epidemiology and other kinds of things on for example strains of infectious organisms these kinds of things would have a big impact first of all because the development of the methods would be useful everywhere that they would be an important foundation for building science in these countries the final point is to use existing projects that are in the genome institute that have an international character like the HAPMAP project or new programs to build collaborations with scientists in developing countries and try to enhance the infrastructure of research locally I'm happy to take some questions Okay, questions, comments? I have sort of responsibility of the Anopheles genome in terms of information at the moment and it's been amazing actually how productive scientists can be in developing countries out of from the information base one thing to remember about the genome is that the export is predominantly information and if you can export the information you really empower the scientists I've been incredibly impressed by the sort of science that goes on as well in the developing countries I think there's a tendency to think that it's somehow second rate it's not second rate and I'd like to add that I really feel that the internet access is one of the is a surprisingly stupid bottleneck to the developing countries it's a stupid bottleneck to have because otherwise there'd be so much information to these scientists now so I'd really like to emphasize that point I couldn't agree more thank you very much Barbara? First I'd like to say that I think it's wonderful that a field struggling with its adolescence considers what to do with its superannuated practitioners There's room for adolescence too Barbara It makes my future feel good Anyway, but what I was going to ask about I don't think any of the people who I've heard suggest this are here and I don't know if your group considered it but I've heard the idea floated several times in the way of this impact on the rest of the world that as we draw in the private sector as well as in the public sector but especially in the private sector things that are eventually lead to treatments, drugs and profit from among other things the genomes of chimps, gorillas, etc that it might be appropriate to take some very small fraction of that profit and send it back to the struggling areas where habitat preservation is a huge issue and a compelling one and I have further heard that when people have floated this issue in downtown Washington where something might be done it's been laughed at as being, you know, totally there's just good idea no way it's going to happen we don't care now I don't know whether any moral suasion coming from this sector to even consider it seriously as part of our roof is appropriate but seems to me like it's worth a thought It's a good thought I'm reminded of an effort I made in front of the leaders of all the pharmaceutical industries to suggest that all of them that have bonanza drugs that make more than $500 million in a single year put the money above a certain level maybe 10% above a certain level into a bank on a kind of tithing system and that money would be used to found a kind of inter an inter-industrial company that would work on diseases of disadvantaged countries if they made money some of that money would go back to the companies but it was a way of trying to think about the same thing to take excess profits and pop them back in but we did discuss some of these things the educational problems the ethical sensitivities to some of the things you're mentioning actually we consider the first barrier that getting the money back is I think possible but difficult One more Just a point that came up last night clearly the problems of nutrition infectious disease are important in the third world but the leading health problems in the third world are still heart disease and stroke and application of genomics to those problems and then taking the fruits of that research to the third world is something that would be worthwhile If you look at the World Health Organization's commission on macroeconomics and health which was just published last year on the leadership of Jeff Sachs you'll see an analysis of these issues and what the commission tended to focus on were those diseases that we thought had low cost remediation in the short run and I think that's an important issue I think the kinds of things that we're doing in this country of cancer and heart disease and stroke and mental illness are going to be important in translating whatever solutions we come to to these parts of the world will be a very important endeavor so I appreciate your comment Okay, I think we will move on Next speaker is Paul Miller discussing public policy priorities for the NHGRI Oh yeah, sure First, let me explain as maybe the only one of the few lawyers in the room I am hoping that by the time of Arleigh three meeting Congress will fund my little small agency so we can afford PowerPoint and present some very fancy slides It seems very as exciting to me as the genome project itself Otherwise, we have a handout I also want to acknowledge Barbara Fuller from the NHGRI who was our scribe and who typed faster than I've seen anybody type and who was critical to pulling this together The purpose and goal of our discussion last night was to look both backwards and forwards over the landscape of public policy leadership that NHGRI has been engaged in and in a sense to decide where we go from here in a sense maybe a little bit of what that roof might look like We spent a bit of time in the beginning talking about what is policy in the LC context and maybe it was because of the participants in the room or maybe just the nature of where the LC program has been We were most comfortable in talking about policy in the context of federal legislative and executive branch issues That was primarily our focus and yet we acknowledged that policy LC policy in this context needs to be much broader than simply worrying about Congress and the executive branch agencies and so we need to in a sense we made an effort to break out and I think that the program needs to make an effort as important as federal legislative policy is to break out of that box to include such issues as sort of thinking about state regulatory issues professional standards and policy education and international issues the lens in trying to think about LC policy the lens through which the public policy agenda needs to be developed is that public policy should facilitate the genetic science we thought that that was a pretty important theme and focusing the LC program policy has to be grounded in NHGRI's science research even though science can inform policy it's a loop and so that policy also informs science so it's a feedback loop but that's one way of grounding the program to talk through some of the historical ways in which NHGRI does policy primarily through funding specific research and specific issues consensus panel approaches and so on and so forth however we acknowledge and we talked about that there's been a limited ability to have LC research inform policy development particularly legislative policy that the research somehow has a difficult way of translating itself into sort of more generic policy and the challenge that NHGRI faces is how to make those research papers that it funds relevant to the policy process and so maybe just because a bunch of ethicists want to carry badges we came up with this notion of an LC SWAT team which in a sense is a quick response team of experts who have a mastery of the library of information of knowledge of issues who are available to inform the legislative process in real time when Congress decides to pick up an issue or when the state decides to pick up an issue there's no time in a sense for going back and funding a research program we need to have real time people who are available to policy makers to make this all make sense and to be involved and we have to think of a way of translating the LC program so it's really relevant and along those lines at the legislative level the opportunity to capture the attention of policy making bodies is very very limited thus the leadership and strategic thinking is critical in these issues and the notion that NHGRI is uniquely situated to take the broad view of the policy impact of the genomic revolution within the federal government and beyond and is an organizing body by which it can step up to the place not representing any particular position but really bring people together within the federal government and so on and the same is reaching out to stakeholders prime directive and we spent a lot of time talking about how NHGRI must be cognizant and respectful of its stakeholders the audience for policy and that sometimes that should not get lost both in terms of what stakeholders are looking for and how NHGRI reaches out and talks to stakeholders thinking about the end users of policy and in that sense really creating an outreach model rather than just an outreach strategic model rather than simply a reactive model to whoever walks into our door with a proposal for a grant or with an idea or a response to a grant but really reaching out to stakeholders in a proactive and strategic way along those lines not to let big P politics dictate NHGRI's policy agenda it is certainly irrelevant as you move along but not to short circuit the brainstorming because something might be politically difficult or nuclear or what have you prime directive for developing policy NHGRI's stature and influence decreases exponentially the further it gets away from genome science and that's a very important theme to consider and to ask this one thing whatever the policy initiative is does it get in the way of people taking advantage of this remarkable science again that's the in a sense an organizing principle for thinking about what might be appropriate policy and so on talked about education we talked about the need for empirical data particularly within the legislative process and lastly again this sort of goes towards the leadership role of NHGRI and that is many professional organizations have developed their own policies their own thoughts on lots of these issues that are LC issues either with NHGRI funding or without and there is no way of really nobody has yet really collected analyzed and disseminated these policy statements making them more accessible to stakeholders thinking through the different ways that different organizations have thought about these issues and that is a valuable a resource and leadership role that we thought that we thought that NHGRI could provide at the end of our little handout we went sort of a catch-all list of just some ideas issues, policy issues that might be relevant to put on a policy agenda and in the interest of time you can sort of read through that I thought that the process and structure and thematic perspective of our conversation was more important to relate to you Questions and comments? Great, thank you Okay, in the interest of time we will move on and our last Summary leader is Maynard Olson discussing the study and the genetic basis of the healthy phenotype For the background to this special topic really had to do with one of the main themes of this meeting which had to do with really what is the NHGRI going to do on the second floor Jack's house There are a number of questions there and they've been vigorously discussed for a couple of days but a lot of them really hinge around the relative role of the NHGRI and other entities particularly other NIH institutes and private industry So in a general way I think the motivation for the second floor and for indeed a major emphasis of the NHGRI in the future in this area is the clear one that there's a responsibility to deliver on the promises that have been associated with the Human Genome Project for better health But one of the motives here was to try to explore with a little greater specificity examples of how that might be done There has, of course, been a strongly articulated view that the role of the NHGRI as well as providing data sets and new technologies that can be used widely in exploring avenues to better health for example by other institutes But I think that it's important in just looking at the context of this special topic to appreciate a couple of difficulties with that generalization One of them is that I think it's clear that as data sets and technologies move somewhat away from their very generic roots the reference sequence of the human genome efficient sequencing, efficient genotyping it is simply going to be essential that these data sets and technologies be developed in a way that's driven by real applications That's how you get innovative technology You don't get it by abstracting it far away from the intended application And I think there's also a sense that there may be some structural barriers in just the way that biomedical research is done to the vigorous exploration of paths to new therapies that cut broadly across disease categories and actually don't have a natural vigorous home in any one categorical institute So that was the context and really I saw at least this discussion as an effort to take a first look at something I would present as an existence proof that there are areas of this type that the NHGRI might profitably explore on the second floor So Lisa Brooks did an outstanding job late last night after I went to the pub preparing these PowerPoint slides and they involve the potential benefits of increasing our investment in studies of the genetic basis of what we decided to call disease-resistant phenotypes I'll come back at the end very briefly to the broader issue of healthy phenotypes So the discussion was guided based on earlier input from the Genomes to Health workshop along the lines of a fairly specific proposal whose main motivation are some quite clear examples in the infectious disease area about how there has been a very profitable focus on the genetics of resistance as opposed to the genetics of disease susceptibility which has clearly been the dominant activity not just at NHGRI related research but broadly in the biomedical research community Two, just dramatic examples are actually a long list from the infectious disease area are the well-known finding of a relatively frequent allele in some populations of the CCR5 gene and all allele which when homozygous leads to essentially absolute resistance to HIV infection and indeed there are drugs in clinical trials which are really based directly on this observation and another example which to my knowledge has not been pursued pharmaceutically but certainly could is the long-standing observation that a homozygosity for a regulatory mutation which eliminates expression of the so-called dark gene the dupianogen receptor for chemokines on erythrocytes leads to essentially absolute resistance to debox malaria and indeed this allele is nearly fixed in sub-Saharan Africa So really the focus of our workshop is their room to try to extend this model more broadly beyond infectious diseases There are some interesting points about disease variants that have properties quite different from disease susceptibility alleles Our basic definition of them is that in some sense they're just the flip side of disease variants There was actually a lot of discussion about what that means, how you measure these disease-resistant phenotypes But the basic notion is that they provide resistance disease such that people with these genotypes do well under circumstances, environmental or genetic background where others do poorly and that there may be unique opportunities to exploit the study of this class of genetic variants for developing new therapies And really the core of this idea and I have only two more minutes so I have to oversimplify but the core of this idea is in the first proposition here that most disease variants clearly there are exceptions and most current drugs again there are exceptions to the production or function of a gene product and it's clear that this loss of function is indeed hard to correct and that in some sense as is the core of the dilemma which Alan Williamson referred to of taking molecular information about disease susceptibility and even building any rational path toward therapy at least any rational path other than gene therapy itself So the idea is that understanding protective mechanisms might lead more directly to the identification of doing drug targets because the idea now is to make small-balling drug which in a person with a normal genotype actually mimics the effect of the disease-resistant mutation and that indeed has been exactly the strategy in the CCR5 case and there's also the point that the genetics themselves actually provide one class of information that's extremely valuable to the pharmaceutical industry which is that it gives a human model if you like for the effects of eliminating that gene product and in the CCR5 case it was of particular interest to the pharmaceutical industry that CCR5 Delta-32 homozygotes as nearly has so far been ascertained after quite a bit of study completely normal phenotypes whatever defect is associated with missing this gene product is quite subtle under most environmental conditions So that's the core of the idea There are numerous challenges we're supposed to call these challenges rather than problems associated with implementing this type of idea and much of our discussion focused on them really at the very core of it is that people with resistant variants cannot show up at clinics The whole history of biomedical research is really a history of having provided health care services which attract people with diseases and we gradually build up this whole wonderful system of institutes and medical schools around these clinics and we draw blood and so forth as these people come through for medical treatment and even very rare diseases then present by this route for study people that aren't sick and doing well don't show up at these clinics and indeed we lack mechanisms by and large to ascertain them in society So there are numerous approaches to this challenge There was discussion of course that genesis have been doing this type of thing in the context of modifiers and lack of penetrance So that's an interesting strategy as essentially looking at relatives of affected individuals and people that as we have an increasing ability to genotype everyone and extended families to take more seriously the challenge of understanding lack of penetrance in severe diseases But I also really would just like to emphasize that we undoubtedly could come up with clever new ideas and there were actually some very interesting ones presented last night but this is not the time to discuss them I think really this is a problem that our best geneticist and this genomicist and indeed broadly the biomedical research community has actually not thought about very hard and one thing that NHGRI could do would be to stimulate some very bright and experienced people thinking hard about this problem There were a few ideas about how to approach this and I've listed them here The point being that I think that this list of examples could increase rapidly if there were more focus on the potential of this area We're in the process just for example to take studying both ends of the phenotypic distributions we're in the process of turning most diseases or trying hard to turn most diseases into quantitative traits pre-clinical symptoms if you like pre-clinical phenotypes are under a very intensive study and we're getting better and better ones for many disease processes and as you turn disease processes into quantitative traits then setting the disease resistant end of the spectrum going to be increasingly effective Clearly other institutes should be involved and we've already had quite a bit of discussion of the point that NHGRI should not be going it alone in studying phenotypes that are already under intensive investigation elsewhere at the NIH but I think there was a clear sentiment in this group that doesn't mean that the NHGRI should just keep its hands off the studies that involve phenotyping and recruiting human subjects and attempting to do genomic studies that are on this more direct path to therapy development there will for example as in many other relatively new areas of research be unique LC issues that need attention so that's it Questions? Wayne Frankel from the Jackson lab so not supposed to be hobby horse but one issue, one area that could help is use of animal models for looking at resistance Yes, no it's a good point and I really should have mentioned that there was considerable discussion of the potential of animal models especially mice where there is of course people have been quite struck by the genetic background dependence of many knockout phenotypes and that's a clear indication that model organisms could be used more intensively than they are now to focus on genetic background effects Richard The chemokine receptor mutations weren't found by genomic approach they were found by a functional traditional biological approach does that change your view at all? It really doesn't there were a lot of human mutations that had been identified and associated with phenotypes before there was any genomics or any ability to dehuman linkage it's just that it's a highly knowledge driven activity it was detailed knowledge of how HIV infects human cells in the PCR5 and candidate gene in those studies the point of genomics is that if you look at the fraction of interesting human phenotypes that yielded to just guesswork it was low and the genomics is brought to the table what Eric Lander once called an ignorant space approach to looking for what genes are responsible for what phenotypes and my own view of this subject is that there is a whole human genetics out there to be discovered that is essentially the mirror image of the human genetics that's described in OMIM for example I'd like to actually thank the six discussant leaders and actually for great presentations and I will now turn the rest of the events over to Francis well I would certainly echo your applause there for the special topics presenters I thought that was a remarkably interesting set of topics and well presented clearly those are some of the conclusions that we will need to figure out how to fold into the final version of this document that we've all been wrestling with and I promise you we will try to do our best to capture some of those very interesting ideas my task here and it won't take me half an hour it will be something less than that but we're not quite sure yet how much is to try to pull together where we have come from over the course of the last say 38 hours or so and particularly where we might then take it next in order to try to distill this into an exciting compelling document that conveys where genomic research could most usefully go next I couldn't help as we've been here at this meeting to think back on the last time we had an early house meeting to plan the future of genome research the month was May of 1998 it was a tumultuous time the goals of the genome project as outlined in the original Albert's document were within sight of being accomplished if we turned our attention in the most vigorous and energetic way towards scaling up of sequencing and a host of other related activities and yet at that very moment there was quite a visible threat to the possibility of the genome project ever actually being able to go forward in that fashion because of other events which I'm happy to say now seem to have sunk fairly far back in the fog of history at least for most people's attentions and here at this meeting we really haven't paid much attention at all to that particular unsettling that happened those four and a half years ago because I think it's fair to say whatever that was all about things have turned out pretty darn well and we now have in the public domain a very far advanced sequence of the human genome a variety of other genome sequences pouring out on lots of other interesting organisms technologies that continue to advance many goals that were not part of the original Albert's plan such as a detailed study of genetic variation already well on the way towards producing extremely useful data sets that are going to advance our ability to understand how the genome works ways in which we are studying genome function which we're not anticipated below those 12 years ago in which are already now yielding fascinating observations that allow us both to understand biology and to understand mechanisms of disease so thinking back on that time four and a half years ago it's amazing how far we have come in that interval but I think this is also important to reflect on because this has been a very different meeting than that one was and not just because of the historical events that were happening in May of 98 but because the goal here that we've been dealing with for these two days is a very different one in 98 and prior to that in 93 when we were meeting together to try to lay out the specific five year milestones we were very much guided by an overall blueprint of what we thought the genome project could and should accomplish and essentially we are now in a circumstance of being able to say it worked it worked remarkably well it's coming in sooner than expected and we have that foundation of the building that is now represented in the house that Jack built or whoever is building it that we are now using as our symbol of genomics of the future so it's been said before but I want to say it once again this meeting is really much more than what happened with that much smaller group back in 1988 that Bruce Albers chaired that tried to put together a much more ambitious and dramatic kind of plan about how to move a whole new discipline into the future we have the opportunity to do that again and I think you all have been doing it remarkably well and I want to say a few things about what I think we have taken away from this some of the major themes hopefully without repeating things that have already been said because I think we have had a fair number of summaries in various forms already that have been quite eloquent and articulate first of all let me say with regard to the document I think you did make it very clear that what we brought to you in advance of this meeting while it had content in it that I have largely heard mostly endorsed in terms of the specific areas that we called critical elements the ones that were the focus of NHGRI anyway most people have felt that is a pretty good representation of where the exciting science opportunities are on the other hand it was very clear that the formulation in the presentation did not do a service to the excitement that we all feel about this and which we would like to convey to the rest of the scientific community and in that regard not to be defensive but I think our document went through probably too many revisions before you saw it in fact I don't know what draft we were on but I know it was well into the 20s by the time it came to you and many hours had been spent on that and many people had had their fingers on it and it had benefited from the output of all the workshops that we told you about those dozen or so over the last year but that meant that all of those particular outputs felt some need to be included and many of them were perhaps in a way that did not achieve the prioritization that a document of this sort might have and in that regard this looked a bit like a document written by a committee and we all know the definition of a committee a committee is a group of individuals who represent a cul-de-sac down which good ideas are lured and quietly strangled so we may have suffered a bit by that particular kind of occasion but I think we've recovered from it quite nicely all of the day by the fact that you all are not shy and retiring and you have pointed out quite accurately opportunities to fix this I told you the story on Monday night about Wolfgang Pauley and the postcard he sent to his friend the postcard containing a diagram which was basically the only the frame of what might have been a wonderful work of art Pauley at the time being somewhat irritated by his colleague Werner Heisenberg who said he developed a unified field theory of whether all the forces of the universe except he had a few details to work out and Pauley's comment was well he could paint like Titian except he had still a few details to work out and that's why the canvas was blank at the time well I don't think we presented you with a blank canvas but I think we presented you with a canvas where the images were perhaps a little scrambled and not as appealing to the eye as they might have been and thanks to you we now have the chance to go back with our paint brushes and redo this image in a fashion that looks hopefully more like a classic of one of the old masters I don't know if it will be Impressionist or whatever but it will have I think much more substance to it and for those who look at it a chance to inspire will be greatly aided by all of this so a few comments this new version including all the elements I talked about at the beginning of the morning such as a preamble that really goes into some of the passion and the excitement will be concise to a level that the current version is not in fact we had been wondering because we do want to publish this next April in a prominent place and obviously the document you saw was much too long to appear in most journals that I know of we were already beginning to wrestle with the fact about well we'd have to have the web version that had all the details and then a shorter version that would get published I think that's wrong I think we should shrink this back based upon your good advice into a concise document and the document will be the document and it might have some sort of undergirding of historical information about how we got there but it needs to be self-contained it need not tell the reader who goes to the journal well in order to really understand this you'd better go to the web if we have to do that then I think we probably failed in our mission to excite the scientific community about where we're going this is and this came up several times this is not a five-year plan this is a plan of unspecified duration just as the Albert's report really was although it had to lay out some possible timelines we're in a different circumstance this time and therefore we ought to couch this document in different terminologies and so we I think are going to be well served by having goals that are very ambitious but not necessarily all of them tied to exactly the same time interval now don't get me wrong we must not shy away from defining specific deliverables when they are well suited to the goals that we outline but we shouldn't force that kind of specific deliverable timeline when it isn't a good fit to the goals or the grand challenges that we think are most exciting for the future of the field there's a danger in being too specified in terms of a top-down kind of view of what the future of a field such as this is with all of the speed with which it's going to move and so there will be some of those specifics and some of those timelines when we are able to do so in a fashion that is I think likely to inspire good scientific work but we're not going to overdo it in this particular instance I think again the content of what you read in the original plan the core points will generally be preserved because they did get a positive review for the most part but again the formatting is going to be resolved in a fairly major way as you saw in this document which seems to have held up pretty well throughout the course of this morning which I take it as quite a miracle with a group of people with high and strong opinions so it must be really good then and we will undoubtedly want to stick with that with perhaps some tweaking around the edges and again that will involve in the document focusing with more of a bright light on the core issues that are the central activities that NHGRI needs to make as their highest priorities with some of the other opportunities for partnership of which there are many and many quite exciting included but probably moved to a particular that they're somewhat different and perhaps not outlined in as much specificity as the core elements I think some of the things that we didn't really quite resolve which are going to be very interesting for myself and my other colleagues at NIH and in fact in other agencies to wrestle with are some of the opportunities that do seem very much like they are core potential future actions of genomics but probably are beyond the ability of the NHGRI to do all by ourselves and we've had many of those in the past we are right now engaging in this haplotype map effort which is very much along those lines with the strong participation of 18 institutes at NIH besides ourselves as well as participation by the private sector and we're going to need to continue to pursue those kinds of models for some of the things here and I will just mention two and one in particular because I think it was a subject of some confusion this morning and didn't get discussed as much in this meeting as the document as perhaps the opportunity might present and that is this whole business of providing to academic researchers the opportunity to identify affinity reagents that will allow you to probe a pathway or to identify a particular protein in a fashion that most academic researchers do not have the chance currently to do and that would be by the availability of large small molecule libraries of various sorts and I might say I don't think that is a technology which is in any way considered to be static and while one of the ways we could do this as put forward I think quite accurately by Alan Williamson is to partner with the private sector to see whether access to such libraries could be obtained I think the ways in which we make those libraries and the compounds that exist within them are going to be changing and there's probably many possibilities in the next few years of libraries that are going to be more powerful than what anybody currently has and if we could collectively through NIH resources work on making that kind of capability accessible to an investigator who is trying to understand a pathway we would both I think open a whole new window of approaches to biology and at the same time provide an opportunity to at least take the first step towards development of therapeutics in circumstances where you do have a drugable target but perhaps the disease is not common enough to inspire much attention outside of academic circles and I frankly from a personal perspective I think that that is an area of which is a real paradigm shift possibility from the perspective of many people who are currently working in the laboratory including people who think of themselves as fairly sophisticated genomicists and it's one that we really ought to pursue with a lot more attention and intensity than has been possible up until now another area that I think we are really going to have to work on together across the NIH and in other agencies as well is this a very large cohorts and again this I will say perhaps is a little bit of my own hobby horse but I cannot imagine are getting to a point five to seven years from now and looking around and not having available the opportunity to discover whether a genotype phenotype correlation is really valid or not it's going to be wonderful to have disease specific cohorts collected it was pointed out today that many of the ones that have been collected so far are probably not done in the fashion that's going to be ideal for what we want to do next with genomics but all of those disease cohorts are by their very definition going to be biased and they should be biased if you're going to look for the genetic contribution to diabetes you don't want to just collect any old case of diabetes you want to load up your cases as much as you can with those that have the strongest hereditary contribution so they're going to be the ones that have affected family members they're going to be the ones that have an early onset and then if you're lucky enough to find a variant that seems to be associated with disease the immediate next question is going to be okay in an unbiased situation what is the relative risk contributed by that allele and if you don't have such a collection that's been done in that fashion and ideally one ought to just do this with a very large cohort that would be sufficiently large to look at most common diseases then we'll all be sorry and if we don't get started on that very soon we're going to find ourselves empty-handed obviously other entities are taking this on the Biobank and the United Kingdom being an example and one that's right now mired in some controversy but I think it's fair to say it's a very good idea and it's the kind of resource which if one can work through all the practical and ethical issues would provide something that would be a great benefit to those who had access to it I might just say that the Biobank in the UK if it happens may or may not be easily accessible to US investigators and frankly will not have the kind of diversity of participants that will help us very much at all with questions that relate to health disparities and if that is in fact one of our major goals in the next few years at NIH and I think it is and it must be then we can't count on our good colleagues in the United Kingdom to just sort of take care of this for us we're going to have to make an investment of our own finally just to wrap this up before I introduce Dr. Zerhouni I want to say the focus of this next plan on applications to health is not window dressing I know it's made some people anxious at this meeting because in fact the glory of genomics over the course of the last dozen years has largely been in what many people would describe as basic science after all the sequencing of the genome our most visible flagship certainly would meet in most people's minds that description but we now have an opportunity in the field in a direction that draws us much closer to direct applications to human health recognizing that we probably can't do a lot of that alone but that second story of the house is a really important one it's not just there as a political statement to try to protect us from being seen as perhaps somewhat less in the middle of things than we were 10 or 12 years ago it is a specific intentional effort to change the mindset to make it possible for people who do genomics not just to write in the first paragraph of their grant application why this might be relevant someday to a disease but actually to be thinking about it and to be thinking about it in a regular ongoing way and to be provided with the tools and the environment and the colleagues to make that a reality in a fashion that up until now has only happened in some places at some times and I think if we collectively can come to that estimation that reality that would be a great service to the public who of course is expecting all of this and in fact to ourselves because I think that's where a great deal of wonderful science is going to occur a couple other things about the roof or the penthouse or whatever we're calling that third story I want to associate myself strongly with Barbara Wald's comments in her presentation although it wasn't particularly relevant to the topic she was talking about I like to the comments a lot I think we are in a circumstance here that bears some attention I am one of the strongest supporters you will encounter about the LC program and the wisdom that undergirded its founding when the whole project got underway and I think it has been a very successful experiment up until now in terms of connecting up research on ethical, legal and social issues with the science that moves the genome forward it is still an experiment it is still in progress it has happened over the course of the last few years where LC has taken on more of a role of the weeping profit the Jeremiah in the circumstance whose utterances largely are interpreted at least by most of the scientific community as predicting doom as basically pointing towards all of the reasons why there is something to be afraid of here and this notion of trying to couch the LC agenda in a more proactive way that emphasizes the positive opportunities is one that I think really deserves some close attention at this current time else we will end up in a circumstance where there is even a polarization between two communities that have so much to learn from each other and here I don't mean to be just saying that the LC agenda has turned in this direction I think frankly the scientific agenda perhaps for one reason or another has also begun to develop a bit of an attitude about the way in which the LC agenda is getting in the way and frankly I think that is also unfortunate and needs to be fought against in particular we've had this comment from several people about the way in which LC restrictions, or at least that's what they're interpreted to be they're really IRB restrictions they're really a common rule restrictions are getting in the way of people being able to do clinical research that they would like to do and I think it is well taken that it would be a very high priority topic for LC to take on an effort to try to map a pathway to get through this particular quagmire but I think actually the answer is not to eliminate protections against misuse of tissues unless somebody assumed that that's really what's being advocated for that will not work we must not lose the confidence in the public that we are carrying out this research in a fashion that is absolutely sensitive to their concerns about privacy and misuse of information frankly I think the principles here are fairly well worked out it's the implementation that's gotten really quite scrambled up and if it's possible it has to be possible for scientists and LC experts and public policy folks and yes even the government because HIPAA is involved in this who frankly need some help trying to sort through all the differing messages that they're getting if we can in fact get that all together and chart a course that maintains those principles but deals with an implementation that makes it possible for research to go forward in an ethical responsible way that would be a wonderful contribution and I think I agree with what's been said that that ought to be a very high priority right now for this community that we should all be proud of of having built and put together and speak to each other and can work together towards that kind of goal which is a benevolent one with much opportunity for public good well that's enough of that little bit of a speech let me finish my remarks here by just saying in a much more practical way what's next for this planning process you have given us wonderful ideas and I think speaking to Alan and Eric and Mark as I called them on Monday night and we will continue to call them for the next few months the three of them and myself will make an effort here in the next couple of weeks to try to capture the major elements of what we have heard in this meeting and reformulate the document accordingly and I'm not too worried about how to do that actually I think it's pretty clear how to do that in a way going gosh what did they say and what do we do now I think it's pretty clear what that pathway is we will call upon the leaders of the breakout groups because you all were wonderfully helpful in this process to vet what we come up with and see whether you think we captured it and we may call on various other people as well as we need specific expertise but soon after that we will then get our council subcommittee those six people whose names I mentioned to you right together to see how they think we're doing because again the ultimate approval of this planning document will be with the advisory council in that ultimate moment where they approve hopefully not disprove but approve this is going to have to come up in February at our council meeting in our open session at that point then it will be probably edited a little bit more and sent off for publication where the rest of the world will have a chance to see what we all did in April in a wonderful coincidence with the 50th anniversary of Watson and Crick in the completion quasi still a little bit but less quasi than before of the sequence of the human genome which at that point will have all of the chromosomes in the form now already enjoyed by 2021 and 22 and many others yet to come in the very near future so that's the pathway we're on and again some of you will get a chance to see these revisions we probably won't ask all 160 of you to do so because then we probably fall into that same situation of too many cooks in the stew but I think we've heard you and I hope you also will recognize that we have recorded your thoughts in all of the ways that our staff have been able to do during the breakout sessions and all of this and the plenary material which is on videotape so we can go back and recover your wise comments and by the way I would like to be sure all of the people who presented this morning make their power points available to the 3Gs so that we have those as well because there's some very good things in there and I don't want to lose them so that's the pathway we're on and again I would like to say this has been an incredible kind of gathering a gathering where we managed to bring together I think probably the largest intellectual collection of people thinking about the genome in a very long time some folks who were here were in fact I think quite impressed to see the diversity of expertise but also the depth of expertise represented by this particular set of attendees and it's been rough and tumble and that's exactly the way it always should be when we're talking about the genome and I think the outcome of this is going to be a dramatic improvement on where we thought we were three days ago and I hope you will all feel by the fact that you were here a sense of ownership because that is an intention that this is not just a plan of some government agency this is a plan of the genome community jointly generated by all of you together here at Early House and then put out for the rest of the world to hopefully have them take ownership of it as well so thank you for your very very important contributions to this and I would now like to go on and introduce the final speaker of the meeting my new boss Dr. Zerhouni Dr. Zerhouni has been here this morning and listening to all of the inputs it's probably a good thing that he didn't prepare his remarks ahead of time based on the documents I gave him last week in fact Dr. Zerhouni is the kind of NIH director who was preparing animated power points sitting next to me this morning while you all were talking so he's very much on top of the technology of immediate response in this kind of gathering I must say from a personal perspective his arrival on the scene not only about six months ago has been a wonderful wonderful thing for NIH Elias comes to us from a previous role as executive vice dean at Hopkins he has great and deep experience in how academic institutions of very high caliber operate and sometimes how they don't operate and have to be fixed and he has brought that expertise both in terms of science and in terms of management of a large and complicated organization to NIH with great energy it has been an absolute delight to have him present in our midst for these last six months one of his very first priorities has been to try to chart a course for all of NIH in the next coming years which is now being referred to as a roadmap and he has vigorously brought into that set of discussions a number of senior figures in the scientific community some of whom have also been at this meeting and it's interesting to contemplate how that exercise which basically tries to look at all of biomedical research overlaps in many interesting ways with what we've been trying to do here in a much more focused way in the field of genomics and so I was particularly pleased that despite his many commitments he was willing and able to clear off this morning to come and be with us to invite him to come and tell you what he thinks about all of this and anything else that's on his mind. Elias. Thank you Francis you know it's interesting when you become director of NIH they invite you a lot to give either opening kind of remarks or closing remarks and I'm wondering always what's best actually opening remarks are not bad because nobody has talked about the subject yet you welcome everybody and closing remarks are really a lot harder especially with an assembly like this one you've done a lot of work I was trying to get a little briefing about what happened during the past two days and I heard that there were a lot of fights and I said that's good and especially following my predecessor Harold and also recognizing all the good work he's done to bring genomics to the point genomics are today and also looking at how great he looks I'm becoming optimistic again and especially that after that I could probably take nine years in India to do but the most challenging thing was really to follow Francis as you know Francis has an analytical analytical mind that is very comprehensive and it's very hard to follow Francis do I have these slides on? So what I wanted to do really was not not to summarize or try to say something that would be over repetitive but try to say something to share with you the lessons I'm receiving not just from this morning but over the six months that I've had at NIH and share with you my own approach to what we're talking about here and some of my concepts that relate fundamentally to the maturity of this field and what it is that you can extract from it that will teach us how to go forward it is a challenge because I think what we're experiencing here is a meeting of you know experts and you've gone through all of the issues and one of the things that I find quite interesting is the boldness of all the proposals and the depth of everybody's comments so let me just share some comments with you on an informal basis and again as as Francis was saying I put these slides together in the coffee break so forgive me if they're not complete or if they're misspellings my first reaction really coming down from Washington I say what am I going to say what's worth saying after your meeting and I said my first reaction would be to be speechless it's probably the best thing to do but for those of you who know me I can't remain speechless for too long so clearly what you've developed is a bold and comprehensive plan and I'm really genome-wide I understand David Botstein wants me to use that word so I'm using that word but really the interesting thing is that I have a little bit of difficulty with and that's the analogy of the house in fact I think in genomics what we're seeing is the result of a seed that was planted many years ago and if you think about that image that seed has grown and in my mind I see it more as an arborization what you're witnessing here at early two is an arborization of the field into completely different importance and directions to me it's more like a neuron actually think of a stem neuron that has evolved and has grown its axon and now it's developing synapses and dendrites to the rest of the system so in fact you're clearly dealing with a formidable problem a formidable reverse engineering problem which is at the core of biology and it is going to permeate through these arborizations we need to tackle it together and it's going to be much larger than NHGRI it's going to be all of NIH all of the world actually and we realize that now the reverse engineering problem you know and all the analogies about parts list and how parts function and so on you've heard those I'm not going to repeat them but it is a powerful analogy because frankly it is the most formidable problem mankind has ever known of reverse engineering at the single cell with a billion molecules and billions of cells and you try to compute that you realize that we've never faced a problem like this in the history of science in fact as we are disturbing our environment as we are changing the ecology of the earth and our own relationships to the environment mastering this problem may actually be the future or the lack thereof of our societies so it is a national priority in my mind and I say that very very clearly because if you do not master life sciences in the next 25-50 years what you're going to see is a humongous impact because of the cost of disease and the cost of disease worldwide on political stability societal stability our ability to fund education and as you know an educated public is a public that supports knowledge and all of those things are incredibly dependent on your work and you understand the big picture here so what you're doing really in my mind is an arborization that is permeating the entire process the thought process of all of bioscience and one of the things that I keep hearing and I kept hearing this morning is this issue of overwhelming scale and complexity and it's too complex too many data points what do we do with them well let me just give you a perspective that comes from my personal training and as I was coming this morning I was thinking about what example can I give you that will make the point I'm trying to make and the best example I could come up with is my combination padlock when I go to the YMCA to swim I have a four digit padlock and if you know your mathematics you know there are 10,000 combinations so it could be 9999 or 0000 and everything in between so you have about 10,000 possibilities very complex but my code is 6398 I'm telling you what the code is you need to have better sequencing capabilities but anyway here's my code 6398 so it's one of 10,000 now you're facing the problem of figuring out what was in my mind when I made that code and why is that code that code and could you really open my locker if I challenged you to do that well you could really there are 10,000 combinations you still will have to use the brute force approach that you hear and that is that accumulate all the data put databases and find correlations well but if you knew how I came up with that code I can show you how the scale and complexity of the problem can reduce itself by orders of magnitude so here's how I came up with this code the way I do it so I can't as I said 10,000 possibilities so I don't forget the code I have a little routine in my mind when I make these codes and I say alright I'm going to take the first number and I divide it by 2 so I fix 6 divide by 2 becomes 3 and I say the next operation is multiply that number by 3 so it becomes 9 and then I remove 1 remember it because otherwise I would have forgotten it but you you're going to say no but you see Eric is a good mathematician so Eric is going to say well but how does that really help your memory well I'll show you why it helps your memory now if I ask you the question how many combinations there are in 10,000 possibilities that would fit multiply by 3 minus 1 anybody has the answer Eric? half 10 okay who says less than 5 you vote now 5 or less than 5 come on Eric 3 why is it 3 because obviously you have if you start by divide by 2 you can only have an even number and so 4 divide by 2 is 4 divide by 2 is 2 and then my 3 is 6 and then 5 and then you can have another one too you can start with 2 2 and 1 and 3 and 2 because you divide by 2 multiply by 3 and remove 1 but what about 8 can't work because 8 divide by 3 is 4 multiply by 3 is 12 out of range is too much animation here but there are 3 possibilities as Aravinda said and a huge reduction in complexity I'm sorry? no because 0 divide by 2 is 0 0 multiply by 3 is 0 0 minus 1 is minus 1 and I don't have minus 1 on my padlock so what it is that I think the challenge in front of you is in my mind is that you have succeeded in essentially growing a new field and you have an identity crisis or problem or issue that you have to resolve is the field a discipline or is the field a tool for many other disciplines and that to me is the challenge the complexity you're facing is why are you scientists to define operating rules of the system because every operating rule of the system you can find reduces complexity by orders of magnitude the second is that you have to understand the inter elemental functions and I've heard that about functional genomics and understanding connected gene elements that are not forgive me I'm a radiologist so it's not my field but you see what I'm trying to say there's a lot of genes that interact through functions that you could quantify and then you have to be really in terms of understanding scale and complexity you have to be able to perturb the system in a way that is measurable and that you have observations that connect with the perturbation and that is why you're hearing these ideas about molecules for research molecular libraries that will allow you to probably change the parameters of the system so you can extract from there the inter elemental functions and most importantly I think you need sufficient data density I think in my mind what I think the field is facing is truly an information theory problem and let me just show you this now again these are rough graphics let's see if I give you a data point like this circle here I'm out of uncertainty around this you weren't sure exactly where the point was but it's certainly a point in the universe of data and I asked you what do you think this represents there's no answer to that but I'll give you a second one and then is this part of a line or is this part of a circle or two eyes and Mickey Mouse and I'll give you a third one what is that tough to say and the problem in information theory is that you need both data density and data quality in fact if I showed you what it is that I intended to form here I could ask you to give me the shape I had in mind and you probably come up with ten answers but if I gave you more data points all of a sudden you realize I was thinking about a rectangle and the rectangle to get to the rectangle you really need to have data points that have a lot less uncertainty around the point and many more points how you accumulate that data in genomics is going to be your challenge and fundamentally to me I need to understand how you want to go about it because I know one thing if you use brute force it will cost us ten thousand more dollars than if you use intelligent force and that's one area of the science that I think you need to promote and that is that modeling and capability for reduction in complexity and defining the key parameters observable parameters that you need to come up with the solution and you need three things you need density and quality but you also need temporal resolution which really comes to the issue of quantum jump in technologies that we need to accomplish and we will do this now Francis mentioned I have to say something about what I've been doing for six months so Francis mentioned the roadmap priorities and it's interesting there are four of them and they seem to permeate the entire NIH and permeate all the fields that we're talking about but they're also stemming very very strongly from genomics and those are emerging research technologies and you've talked about them RNAI libraries or very advanced imaging capabilities at the molecular level single cell three-dimensional structure needs to be understood the relationship of structure to function needs to be understood we don't have the tools right now to do this we need to evaluate them and then new pathways to discovery and what I just described in terms of mathematical treatment in the field is in fact a way to have a new pathway to discover new things the idea of comparing the healthy to unhealthy is a means to discovery it's not a field in itself it's a step in the discovery process now the third one which I think is very important to you is misspelled re-engineering of the clinical enterprise I really don't believe at this point that we have a clinical research enterprise in the country that will support your dreams you've heard cohorts you've heard the ability to share data you have the ability to in fact inter-operate clinical trials no, actually other countries have more advanced capabilities than we do so I think one of the issues we have to look at and I'm looking at is how can we re-engineer the clinical research enterprise to serve the new opportunities in science which we never had before ten years ago you wouldn't have had to have that discussion as you've had it today in addition human subject protections are coming to the fore and in addition the privacy protections are coming to the fore and I have trouble seeing how the research enterprise will serve science if it is not redesigned in a way that provides firewalls around these issues of privacy now you thought that gene was a difficult issue and you have Bermuda 1 and you may have to have Bermuda 2 again to talk about it some more but let me tell you the difficulties you're going to have in terms of human data and biological samples are going to be even more daunting and I think you need to be proactive about this and go forward and last but not least is the scientific team of the future and we're obviously on the same wavelength which is clearly an issue that we need to talk one of the sentence I came up with is as we were talking about let's sequence more genomes let's talk about variation really what we're looking at in the next phase is to go from the human genome to understanding the genome of all humans and that is a challenge I think that early tube needs to place on the table because I agree the specificity of the whole of the human genome process was a terrific move and you need to stay there because it is more challenging now to obtain funding and convince Congress, convince the public that we need to have continued funding that the genome is not the end of the road the beginning of the road and that requires public trust and one of the things that I would like to emphasize and I know you have emphasized public trust is paramount and when you look at public trust what is that? in fact the most important definition I've heard of public trust is the ability to predict someone's behavior and the public wants to have the ability to predict our system's behavior the system of research and our thought process the system of science because if they can't, if it's unpredictable then they will lose trust and trust is based also on education because an educated public is a trusting public they understand at the end of the day what are the pros and cons look at what happened in Zambia where genetically modified foods were turned down by a starving population ignorance is the enemy so we have to be hypersensitive to our diversity and I just want to emphasize that and I think you have emphasized the issue of consideration of moral and ethical implications so I know we're on time as you can see I don't remain speechless for a long time but I think I wanted to share where I was in this process as a new director is challenging sometimes I have to say that I have a very well varmed seat where I am thanks to Dr. Varmas and he has done an outstanding job in preparing the field I think it's our challenge to go forward and to develop it further with vision across NIH and really across the entire disciplines of biomedical sciences I'm in for the game I see that Dr. Insol our new director of the National Institute of Mental Health is here and I'm going to borrow one of his sentences when I introduced him to the advisory council he said he came to NIH fired with enthusiasm and was hoping not to be fired with enthusiasm and and I share his view thank you very much thank you Elias for a very provocative thoughtful sophisticated analysis of some of the major issues I think you put your finger squarely on a number of the things that we are facing in the next few years and it's wonderful to have your very thoughtful analysis resonating so deeply what I think all of us are hoping for the future and partnership with you and the rest of NIH is going to be crucial if these dreams are going to come true that's a fitting finale I think a really remarkable couple of days I want to thank all of you one more time for the time and energy you put into this particularly to thank those who performed in leadership roles and for me to thank Alan and Eric and Mark for all of their hard work and for all of the NHGRI staff who performed in various ways as scribes as well as running around doing a host of other things particularly Susan Vasquez for many of the logistical issues that were so smoothly handled one of which is that I have a note in front of me to say that the dullest van that's supposed to leave at 1230 is waiting outside and ready to go and so with that those of you who are wanting to stay for lunch it is now served downstairs please go downstairs and continue your conversations those of you who have to go Godspeed to all of you and we'll try to make the best of all of your wisdom thank you very much