 So I don't know about wisdom, but I'm going to attempt to impart a few Lessons from what I think now can almost be called history I did have a some sense of deja vu about this discussion it reminded me of Policy forum I wrote in 1995 entitled a time to sequence the Context here of course was quite different, but the message the same We had been through a period of intense technology development and mapping And what we had was not perfect, but it was pretty good And I made the case here that it kind of was time to sequence and I I think I got the timing About right. This was necessary to make a major shift of resources of focus and so forth and that seems to be the overwhelming Impression that I've gotten. There's a kind of consensus that it's a time to sequence I also had here just a somewhat humorous way that I started this article reminiscing on how Eight years before My my views toward the human genome project had first appeared in science and in the form of this one one word quote in retrospect The human genome project was a big project, but it actually wasn't huge What we're talking about here at least as I would like to define it is huge and is going to require Learning all of the lessons of the HGP and and many others and is going to involve the collective effort of that an Enormous number of people with highly diverse skills From time to time I'm asked to give talks about the so the right and the wrong lessons from the human genome project This slide captures just a few of what I regard as the right lessons some other day I'll discuss the wrong lessons which are the most commonly cited ones I think a key point is that Policymaking should really focus on goals and not detailed means and I think that as we move on from this workshop We really have to try to crisp and our ideas about what we're trying to accomplish The questions of how to accomplish them are gonna have to be worked out in pilot projects and going forward So my herding cats Point is that at its best the the human genome project was a cat herding operation And I think that's very much the way that this kind of endeavor Should go and the point here is that there has to be a certain amount of organization, but not too much I Think the human genome project always got in trouble when it was Too tightly organized to tightly manage from the top down But of course complete chaos doesn't work either if you actually have goals I won't be labor the third point But one thing I learned I've learned in the HGP is to sort of be aware of predictions of IT crises Because they actually tend to solve themselves and I think they saw themselves because we're bit players in the IT world Streaming high-definition video into every home in prime time is a problem So mind-bogglingly bigger than the ones that we're discussing that And has such strong consumer demand behind it that we will continue to be able to piggyback on the that Technological sector and I was pleased to see that that Daniel and Nancy are two main commentators on The subject really I heard them saying more or less the same thing One area in which sort of top-down activity is absolutely essential is In quality control and her laboratory data standards and so forth and it's not a problem that you can solve In one go you need a system a process We've heard for example quite a lot about the encoded a project not not directly, but it's hovering around our activities and Having just watched it. I was not a participant I think really its greatest success is that it took a bunch of functional genomic kinds of assays that scored very poorly when the project started on these kind of criteria and and really developed Strong interlaboratory standards you could actually recapitulate much of the encode data now today Relatively easily and with much less expense mostly because we know how to do it now So let me focus a little on goals rather than means because I do think that that's what we really need is some kind of I I don't think we perhaps have a strong consensus about goals here But that's where I think we need to focus the discussion My own view is that what we really need is a long-term big-picture plan To do what I call modernizing the discovery model that guides my own medical research So this is rather grandiose kind of goal But notice a couple of things about it is the emphasis on modernization I think it's going to be critical in this period of Tight resources where we're going to have to be doing this I think that we just need to learn to think about this project and to sell this project Not as a big new add-on initiative, but it's a modernization of Activities that the NIH is already very heavily involved in and we gradually have to start doing them somewhat differently And many of the bullet points like the ones Eric's just showed You know indicate a lot of different ways in which this enterprise needs to be modernized but the other point is that We're really talking about a discovery model that applies to much of what the NIH does at least in its basic and translational research and That's a different way of formulating the goal. It's not as easy to sell as as That we're going to have a major decade-long initiative about a particular disease or whatever But I think it's what we need to do and that the risks are high that will be distracted by more parochial goals There's not time Now to really talk about How this modernization of the discovery model might occur I would like to refer people interested in In the subject to a recent National Research Council report called toward precision medicine is issued last fall I served on this committee. It was chaired by Charles Sawyer's and Sue Desmond Hellman two prominent cancer researchers and this report made a major effort to lay out The a game plan a long-term game plan for modernizing the discovery model of biomedical research and to move toward What is referred to here as precision medicine just as a brief aside the committee these words turn out to be important and And the committee self-consciously chose this this term over personalized Harold Varmas actually had the nicest comment on that that I've heard in which he captured I think exactly what the committee was thinking He said his father was a GP and he said, you know my father practiced personalized medicine He knew his patients names. He visited them in their homes. What he practiced was precision medicine He could not tailor the treatments that he had available to particular Sort of biological aspects of those individuals. We want to do that with more precision We get too caught up in the whole personalized idea and it does lead away from evidence-based medicine from it Leads back towards anecdote and we're seeing a lot of that right now And I see it as actually a major risk in the say decade or so ahead This report is full of diagrams of this general sort and you'll you'll notice their similarity to Many that were shown in this in this meeting with the information commons the knowledge network and many things filtering into this or feeding into this and Various sort of virtuous cycles that one would like to set up We've covered a lot of this ground at the at the core of the committee's recommendation is a Shift towards more emphasis on the study of patients that are embedded in the ordinary course of health care there is going to be tension and I think Patricia did a particularly nice job just a couple of talks ago and capturing some of these tensions It's going to take time, you know to really launch an endeavor that is modernizing this discovery model and Other things are going to go on at the same time But I think that 10 to 20 years from now The standard discovery model and biomedical research will be largely focused At the discovery stage on large patient populations that are embedded in the ordinary course of health care And that these various mechanisms that we've discussed of then pulling people out of those environments for more specialized Purposes will will be the the main follow-up mechanism