 We're going to do a low-tech talk about technology. I guess my funny little story to continue where Barbara was is last night, Francis comes up to me and thanks me for agreeing to give this talk and about 10 minutes later, in sort of the spirit of the good cop, bad cop, Jeff Trent comes up to me and gives me his condolences for having to give the technology talk at this meeting, so we'll see how it goes. In 1900, David Hilbert, the famed German mathematician, presented a prophetic paper at International Congress in Paris regarding his vision for his field in the 20th century. In his introduction, he noted that every real advance goes hand-in-hand with the invention of sharper tools and simpler methods. I would like to focus my remarks on some of the sharper tools and simpler methods that we require to translate the output of the human genome project into a true revolution in biomedical research. My point of view will be from the perspective of an end user who believes that the systematic and comprehensive approaches that have characterized genomics programs need to be translated into a framework that can enable scalable, more quantitative biological research. I'll address some of the technical and methodological changes with respect to tools, resources, and processes that must be addressed in order to arrive at the next level of understanding. Given time constraints, this will be a bottom up view of technology that is in no way exhaustive and will intentionally defer discussions of a number of obvious genomics technologies to other speakers or breakout groups. In addition to discussing technology per se, I'll address a number of sociological issues, as I call them, relevant to the ongoing problems associated with technology development, implementation, and access. I think today the principal goal of the Human Genome Project has been the enumeration of the objects of biology, defining genes, genomes, and their products, as well as describing the rudimentary behavior. These efforts, directed at building the infrastructure in the form of resources and tools, have been motivated by, I think, our collective desire to deconstruct biological processes into the molecular components. I believe that the key challenge for the coming century, so this is the 2220 talk, I believe the key challenge for the coming century to be to establish complete molecular descriptions of biological processes that are sufficiently quantitative and dynamic to enable predictive modeling or simulation of that process. To achieve this end, we will need to move away from the study of individual molecular components as the fundamental unit of research. Instead, the new focus will become macromolecular assemblies of increasing complexity, defined as physical entities, as well as functional units. That is to say that our attention will need to move towards pathways, networks, molecular machines, organelles, and eventually the cell itself as a unit of work. In order for it to achieve what will essentially be a four-dimensional description of biology, our technology must enable us to be not only more quantitative and system focused, but also require that we abandon destructive methods for studying cell biology and gene function. Such capabilities, quite frankly, are well beyond our current technologies. And the future developments that will enable us to directly view biological processes at this molecular level are predicated on our ability to interface with other disciplines. We need to proactively leverage developments in other fields, such as material science, MEMS technology, nanotechnology, optics, as well as polymer and small molecule chemistry. It is clear to me that the near-term goals of the Human Geno Projects will be to complete the catalog of the molecular objects. This catalog will not only include genomes, genes, and gene products, but information on natural variation and reversible covalent modifications. It will also be important to expand this cataloging effort to include other cellular constituents, for example, metabolites. A key driver for investment implementation of genomics technology has been the attempt to reduce the time and effort an individual scientist spends creating and assembling reagents, resources, and information necessary to design and perform an experiment. The logical extension of the current efforts will eventually lead to elimination or trivialization of the classical tasks associated with molecular biology research. I think freeing the hands as well as the minds of the community to focus on a new class of problems. There are a number of technology-related initiatives worth considering the near-term that will significantly enhance biological research and essentially force us to confront these more significant technology challenges. For example, while there are current efforts focused on assembling collections of full-length cDNAs and deploying them into universal vector systems that readily support a variety of tasks, I believe that we ultimately need to develop efficient and cost-effective methods for complete gene synthesis. Such tools will simplify sample tracking issues for agent distribution and storage, as well as enhance our ability to manipulate gene and genome sequence in support of experimental objectives. In a similar spirit, we will need to consider systematic efforts to create expanded antibody or antibody-like reagent sets directed against all proteins and all essential molecular determinants. These efforts will also challenge our ability to correctly express large numbers of proteins in both large and small quantities, a skill set that cannot be underestimated. Given the relatively large size of antibodies and the difficulty of introducing antibodies into cells, there are significant limitations in their uses as biological tools to quantitate or monitor in vivo intercellular activities. Alternatives will need to be explored, and it is likely that there will be a collection of approaches required for regenerating a robust genome-wide set of reagents. I think it's some areas worth exploring, include approaches for directly labeling proteins such as chimeric protein strategies, such as GFP fusion strategies, methods that enable us to systematically incorporate modified amino acids, for example, fluorescently labeled amino acids into proteins, or improve methods for in vivo strategies for covalently modifying a protein using chemical methods. Clearly, the delivery of such reagents into a cell with accurate incorporation into the cellular compartment of choice with appropriate level of activity and the maintenance of normal biological function represents significant issues that will need to be addressed. Another alternative to antibodies would be the application of, for example, small molecule chemistry, polymer chemistry, or nanotechnology to help us develop reagents and tools to selectively follow, explore the functional activity and dynamics of molecular components in biological processes. Such reagents could be used to indirectly label proteins by selective binding, by selective binding, blocking, or activation of protein activity, or monitor protein function by acting as a substrate or sensor or reporter of a selected activity. These represent a very powerful subset of tools that are generally not available or accessible to academic researchers and are well worth considering. The fundamental strategies for studying gene function require us to modulate or alter gene or gene products activity or expression. It is certain that we will see an acceleration and expansion of such activities dedicated to generating genome-wide collections are deploying systematic methods for global gene knockouts in model organisms or cell lines. The advent of RNA interference and improvements in anti-sense technology are examples of scalable technologies that are available in the here and now in merit immediate exploitation. In addition, I think we can anticipate improvements in gene and protein switch technology that will enable regulated and perhaps more importantly, tunable and reversible modulation of gene and protein activity. These features will be based not only on classical molecular biology methods, but will increasingly employ small molecule agonists and antagonists as a source of the tunability. Disruption of function by any of these methods plays significant demands in our ability to quantify the dynamics of gene and gene product activity as well as the spatial and temporal dynamics of synthesis and turnover. This technical gap is really the Achilles' heel of our current biology, with extract to extracting the type of information we will require for more complete description of biologic processes. Tools such as mass spectroscopy or chip-based profiling paradigms that can globally monitor in parallel changes in aspects of the activity of molecular components deserve increased intention, but are only part of the answer. Such technologies are limited by cost issues as well as by the current status of our separation and purification technologies. Furthermore, these processes are examples of destructive testing, and it's very likely that the new biology begins just beyond the reach of some of these technologies. Before we leave the discussion of immediate technology goals, I wanna consider some near-term issues in cell biology, as I believe at least over the next 20 years, the cell itself will become our primary unit of work. I believe the fundamental issue is actually the cell lines themselves, and we clearly need a better biological and genetically characterized resource. I will suggest that the current revolution in stem cell biology will represent the logical place for the development of a more robust set of cellular reagents that accurately reflect underlying biology. Another area that needs to be addressed is the translation of cell-based assays into single-cell formats, enabling a more quantal approach to biological research. In addition, we'll need to improve our ability to manipulate, separate, and sort single cells or small quantities of cells, and I suspect that this is an area where men's technology will have a tremendous and perhaps immediate impact. I'd like now to turn my attention away from issues relating to reagents, tools, and process in order to consider what I'll call a sociological issues surrounding technology. One of the most important consequences of the genomics initiatives has been the introduction of high-throughput process technologies to the discovery phase of research. The increasing reliance on mechanical, automation, instrumentation, laboratory information management systems has had a significant impact on the workplace, making issues of operation, organization, and diversification of an intellectual capital quite frankly part of the competitive mix. I think we would all agree that one of the current facts of life is that ongoing access to a broadly defined and well-integrated technology base is a competitive advantage. Furthermore, narrowing technology life cycles and increasing expectations of productivity suggests that technology is not only indispensable, but quite frankly it's increasingly disposable. As a consequence, when contemplating technology investments, we must not only consider the development or acquisition of technology, but almost will pay attention to the issues surrounding the integration and maintenance of such a platform. This emerging reality is another, a number of important implications. It is clear that groups and investigators will need increased levels of resources and increased flexibility and freedom to operate with respect to deployment of both financial as well as intellectual capital. In the case of larger centers, in order to maintain critical mass, longer term stable sources of funding, not necessarily tied to grant cycles who need to be identified. The system does not change. It is very likely we'll witness emergence of quasi-private research institutes, loosely affiliated with major academic centers. Such centers will develop their own endowments and will not only have more direct control over the resources and intellectual property, but will have increased flexibility to develop partnerships with the private sector. As a consequence of the universal, it will have significantly reduced its access to technology, risk losing some of its best faculty and students, and the gap between the haves and head knots will only increase. The expanded need for capital creates a number of interesting challenges as well as opportunity. One of the ironies of raising very large sums of money is that the ultimate decision makers are not only forced to choose between a vast array of diverse options, but are unlikely to be able to judge a funding opportunity on its technical merit. As a consequence, psychological factors such as credibility, vision, popular opinion, or conventional wisdom have a significant impact in decision making. In such an environment, public relations and positioning become essential tactics, and we really can't ignore these issues. For example, do you think it is easy to raise money for biological research in the context of a technology bubble when the public is focused on human cloning, bioterrorism, or genetic pollution? Another example of how public positioning or guidance by the NHGRI or similar groups can significantly impact technology development is respect to the private sector investment. The single largest source of discretionary funds dedicated to novel technology development can be found in the venture capital community. Venture capitalists are not a collection of futurist or technology gurus, but are highly-tuned censors of the leading edge of emerging consensus. As a consequence, clear articulation and programmatic vision by groups such as this will influence the course of investment, creating not only a pull on technology, but a significant multiplier effect on investment. Finally, from the perspective of technology development and distribution as well as access to capital, it is important that we pay increased attention to the interfaces between the public and private sector. For example, an area need of immediate attention is the management of technology licensing intellectual property. If you view this as a problem of asset management, you can hear the level of experience and the resources available to the university offices that manage the endowment of the university, to the resources and personnel associated with the Office of Technology and Licensing. One immediately understands the source of the well-recognized problems, such as establishing partnerships, ensuring flow of information and technology, as well as encumbering or fragmenting intellectual property portfolios. In summary, I believe the near-term technology investments and developments represent logical extensions of current efforts designed to compile and catalog molecular objects, as well as to eliminate traditional tasks associated with molecular and cellular biology. As we complete this program and begin to address issues associated with creating more quantitative biology, a significant number of technical issues will need to be tackled that require exploration of a ray of emerging technologies associated with fields traditionally outside of biomedical research. And quite frankly, I think we have to eliminate the trivial tasks before we can, before we'll get the community to focus on these more difficult tasks. All of this technology development is associated with significant costs and will continue to change the fundamentals of our scientific culture. So let me end where I began my remarks with a quote from David Hilbert's famous Paris address, as long as a branch of science offers an abundance of problems, so long it is alive, a lack of problems foreshadows extinction or the cessation of independent development. Thanks. Opportunity for... Said less than 30 minutes. Questions, reflections, reactions. Like in the world where you only have 15 minutes to pitch anything. We'll start with you in again. If people don't know, I always have a question. So from my perspective, one of the interesting technologies that we want to get our hands on is... Sorry, got to stand up. Small molecule technologies. And you've mentioned some of them in passing. And of course there's a lot of existing small molecule technologies all buried inside these wonderful pharmaceutical companies. And again, we don't want to do the same thing as pharmaceutical companies. We just want molecules that are going to mess up the cell or going to bind things or going to do all sorts of things. That's what you think. Okay, maybe I'm wrong. But I can't suggest that we go off and steal as we can with our academic computational side stuff from them because that's probably illegal. But there seems to be an opportunity of coming to some partnership with a very big existing technology base that currently is very much kept behind that is not accessible to academics. And I was wondering if you had any thoughts about how to unlock that potential. Well, I think you make a very good point. And that's one of the points I want to make sure out here is that I think whatever people call it, chemical genetics, chemical genomics or just old-fashioned pharmaceutical chemistry, these molecular tools are the most available tools we have for sort of modulating gene activity and doing all these fun things like creating biological sensors. What people lose, it's actually not a big day-hour as you think to academics to get into chemistry these days because as much as we take pride in sort of the genome project, what we forget is the same thing happened to chemistry in the 1980s. So the development of comaterial chemistry, the application of high-through-the-process technologies, the screening, et cetera, et cetera, all happened in the 1980s. And a lot of the drive for genomics technologies by the pharmaceutical injury street in the 90s was really arising the fact we have all the screening capacity, how do we get to the next level. What has also happened is that a lot of the robotics and things that are necessary, both on the screening side, the assay side, and also for the development of these small molecule libraries, 10 years ago essentially required 60 or 70 people inside a company like Glaxo to build. And they've all become commoditized. There's a whole set of companies which are called tools companies we do. So for an investment on the order of 10 to $15 million, because I've actually done this, one can actually set up a really tremendous capacity to generate diversity, to screen, and bring online the sort of necessary know-how to really work in a focused area. So I think it's a false presumption in our part that this is an area that we can enter into. And almost every academic campus has access to medicinal chemistry know-how and expertise. So I think it's really a question of will more than anything else. And I think for example, in many classes of proteins we're interested in, the know-how around building selective inhibitors or agonists, G-protein coupled receptors, increasing areas of kinases and proteases is so well advanced that getting relatively selective inhibitors to use as biological tools is something that happens in months. I have many examples of my normal organization where we actually made tools within a week for doing gene knockouts. I don't know what technology enables you to do that. So I think it's just a matter of deciding you're gonna do it and doing it. You mentioned some of the issues that relate to intellectual property in our research or the university research in particular, and how relatively speaking technology transfer offices are not necessarily the places where the people of greatest wisdom and intellect collect. I'll stand by that. And my question to you is can you think of strategies that could be deployed either by the public sector or in a more decentralized fashion that would force university presidents really to pay attention to what's going on in technology transfer and perhaps to come together to come up with some sort of policy for the academic sector on what they should be doing with respect to technology transfer. Because my sense is that technology transfer has sort of been pushed by lower level people and university presidents may not have been paying attention at a fairly, or direct attention to the issue. And therefore you have a constant push to get a certain more intellectual property rights and more and more and more and then compete with other universities to assert more and more and more. So I wondered if you had any thoughts. Yeah, I think by goal aside, so I don't want to even step on that landline, I think the fundamental problem is asymmetric warfare. You can't have the Marines on one side and the Boy Scouts on the other and expect there to be a dialogue of equals to create a system and I think that's as simply as it gets. If you go back into universities and you ask them how they managed their endowments in the 70s and 80s, they all realized that they were doing worse in the market. They didn't diversify portfolios, et cetera, et cetera. They brought in professional money managers and suddenly we have universities with multi-billion dollar endowments because they brought first line people who get paid the same way, get paid at Goldman Sachs for investing MITs or Harvard's money. If you don't do the same thing in the technology licensing office, which I would guarantee you is just as much of an asset, probably a more durable asset than finances, then the market will find other ways. That's why I mentioned quasi-private kinds of things because people on their own will figure this out. They will raise their endowments, they will manage their money and they will manage intellectual property. So I think the way the university have to do is take it seriously and they can't do an ad hoc. I think one of the greatest dangers is that I think there are concerns for by some universities, particularly medical centers that they're not deriving enough capital, enough cash from their technology licensing. One of the consequences is they end up doing stupid deals which seems to be very front-end loaded on the cash side, but they actually do two things. One is they do not participate in the long term and they actually create a situation which all of us have lived through the Kree locks and many other technologies of fragmenting and encumbering technologies in a matter because no one thought it through. And so I think if you had parties who were seeing each other as it equals, I think the market's efficient enough to come up with methods because the private sector needs the public sector and the public sector needs the private sector. So I think it's actually a relatively simple thing. Jeff? I think it's one of those journeys of a thousand miles begins with a few step problems that if we can't systematically, and I use the word tunable because I agree with you, I don't think it's a binary system, but be able to in a tunable fashion turn things on and off, which I think ultimately is one of the beautiful things about using small molecules that you probably have a greater dynamic range, particularly in cellular kinds of situations to study those problems. And I think if we solve that problem first, we can begin to address the other things. And I would actually err on the side, this comes back to the question of thinking about more selective inhibitors, because one of the biggest problems in drug discovery, quite frankly, is unintended targets. I mean, it is very hard, quite frankly, to design a small molecule of three to 400 molecular weight that only hits one kinase and doesn't hit another other things. And so the reality is that selectivity from a sort of model system point of view is probably better in the near term. And if we can solve that problem, I think the more complex problems begin to come on. Jeff, can we hear your comments on completeness, and specifically with the reference to technology increments, but across the whole range of molecular objects too, not just the human genome? Well, I think, you know, I guess people who work for me know I say this probably every single day is I think perfection is the enemy of good. And I think Dr. Wold said it correctly is that it's a bang for the buck issue. How complete, how complete you want to get, probably depends on how valuable that last yard is. And I have no problem with deferring goals of completeness to a better technology. I think a lot of us make very pragmatic decisions. And I think that was the big emotional issue around the drought sequence versus the complete sequence was really accepting the fact that the current status of technology meant we could get a 98% solution fairly immediately and we'll defer to the future of the finishing. And so I think there's a need for pragmatism and not dogmatism in those kinds of problems. That's sort of my view of completeness. And I think in terms of the broader notion of completeness, the reason I threw this notion of metabolites and other things is that I think we have too narrow a vision of what in fact constitutes molecular objects. I think for most of us it's genes and proteins and even proteins is only begrudgingly accepted in the last couple of years as a bimolecular biologist. And so I think getting broader is actually very, very important because at least as we've done a lot of work on looking at two, 300 components in networks, you're being forced to look at other things that are going on in the economy of the cell. I mean, when I come to this point, I always remember in Judson's book on sort of the history of molecular biology, it makes interesting points that the first 50 years of last century dealt with the flow of energy and then we forgot all about that and we dealt with the flow of information. So at some level, some grand unification has to happen if we're really going to get complete pictures. So I think completeness has to really be translated to comprehensiveness probably. Since you have the ability to be outside of the grant system to some extent, even though you've been on the review side and I have the same opportunity as an ear-mirror investigator of not participating, I was particularly intrigued with the component of your talk that really might say that our opportunities, especially in technology, and I think we all know this, are limited by the current mechanisms of funding. You sit on these study sections, you look at them, they're very difficult to fund, they're very difficult to pull together, and it is almost forcing individuals that want to marry these technologies to go to places, as you mentioned, which are either quasi-academic, sit somehow between these, and what that does, in my opinion, is marginalize the government's role in stimulating the process. So have you thought of ways to suggest to people like Francis in the front row and others, institute directors here, how they could help the team building or other approaches where maybe the R01 component is only one of the measurements for us? I mean, I think you, the problem is is, I think ultimately, I think here's the problem that I think we're facing, is that, and this comes back to the compression of technology life cycles, is anyone who sat on any of the sequencing study sections knows is that by the time the grants got there in response to some RFA six months ago, the technologies changed. I sat on the study section where, all of a sudden, 9612 capillary electrophoresis was the technology to regore, and every single grant here was irrelevant, and that is happening continuously. And so that's why I think ultimately we have to think about providing larger sums of discretionary money, and that are unlinked from the grant cycle. I think this is, whether you call it endowment, call it what you want, but we have to think about, particularly in these larger centers, way that people can smooth their expense, make forward-looking decisions. I mean, let's be frank about it. If you really wanna get bold about technology, you gotta buy technology, and we realize that 99% of what we do at any one time is probably wrong. It's just that we don't know which 99%, and so that's really the sort of challenge that we have. So I think that's ultimately what we need is we have to figure out a way, whether it's through private funding, whether it's through government mechanisms to create discretionary funds that allow people to take the chances. We used to talk about in R01 grants that we work on our next grant, but that really, that's okay when it takes $10,000 to start a pilot project. But I just know some work that we've done, for example, what I'll call cellular genetics, thinking about how to knock out every gene in cell lines. Well, that required us to bring in-house a lot of sort of micro-injection equipment and things like that. I mean, we decided on a Thursday, someone was in Germany the week later, and two weeks later, everything is set up, and you started doing the experiment. If you have to wait 18 to 24 months, why do it? So you can't do that $10,000. You need a half a million dollars or something just to get started. So I think we're just gonna have to fundamentally change our view of the peer-reviewed system, and maybe ultimately what we have to do is distribute responsibility and trust the quality of our investigators to make wise decisions and be accountable for them. Jeff, I hear with the interest your comment on the raising of capital and the managerial part of science, and I understand how in the type of position you are occupying, you have to think on a large scale. But I'm worried about the loss of individuality, and when I think back, we wouldn't be here except for three individuals at the present time, and Maxim and Gilbert and Sanger. And those were individuals. They weren't big conglomerates of people making advanced high throughput technology. So how do you envision using the resources of, you might say, the commercial section to generate the individual ideas rather than the largest scale one? Well, I'll make it what I think would be a controversial comment. I believe big science guarantees the life of small science, and this is why. I recently gave a talk at the Miller College of Wisconsin, Howard Jacobs invited me, and I talked about some work we did in angiogenesis about a lot of, let's say, a saturation genetic screen across a particular pathway, which was quite impressive because they used the collection of complete knockout set of genes in Drosophila and complete knockout in zebrafish. So Howard raised his hand, and he said, well, how many people did that project? And my answer was three, and that's the size of a small lab and a small group. Why did three people do this project? Because they had access to an infrastructure which actually could generate tools, resources, and reagents. So I don't think every bit of science has to be done in sort of an industrialized process. Quite frankly, industrialized processes are great for sequencing the genome, but the kind of feedback and correction systems we need in biology or not. But quite frankly, small investigators are equally enabled by large science. So I think what we have to do is find ways that people can remain affiliated. And I think if you look at universities, MIT, I think Mental College of Wisconsin, Bailer, the NHGRI, where there are big centers that have all these things, I think the unintended consequence is that the small labs or the creativity and the drive, the things, actually benefit disproportionately in many ways. So again, big science I think guarantees small science as long as we short access. And somehow, when access doesn't have to be, every investigator has the $100 million budget, it means they have access to some center that actually does. My question is sort of about the completeness issue. Right. In its larger context that you really brought up, I sort of feel like that the technology, we can have a very robust feeling about how that's gonna be developed. And then the completeness aspect comes down to, okay, so we've got all of the genes, we've got all the transcripts, we've got all the tissues, we've got all the embryonic stages, but from which organisms? And how are we going to do that across a diversity? And do we need to think about, we clearly can't do that over all organisms, but can we, what about stem cells? Can we make stem cells from all organisms? Or how will we actually provide a finite resource that will allow the breadth without saying that the whole ecosphere is our laboratory, without making some strategic choices, choices that need to be made sooner rather than later, and all of this in the face of the pragmatism that you're so devoted to. Right, well I think the choice is ultimately a question of economics. If sequencing costs a millionth of a sense of base, then I think evolutionary biologists would go out and sequence as many things as they can find and develop reagents and resources for everything. So I think from a pragmatic point of view, I think people have to make their, I guess I'm arguing for your own choice. I mean, I'm very lucky in some sense is that I understand the boundary conditions in which I work, and so that's how I can direct my choices. So what I think we're trying to, when I talk about technology, I think it is the preservation of choice ultimately. And completeness should be, it's a point of view question, what is complete for me and what I need to do my task is linked to the process that I engage in, but everyone here has their own scientific sort of point of view and that's where completeness is. I don't know, I don't think we can come in this room and come to a definition of what complete is. I mean, you know, there are whole mathematical disciplines dealing with the concept of completeness and we can't even approach that level of clarity. So I think it's really point of view and an opinion, so I don't even know where to, you know, draw a line in what you're trying to predict. In the last 20 years, I would have said there has been a significant shift in the culture in science. I think you standing up here in fact is a perfect example of what you've been able to do with your company. And I think we saw that perhaps most profoundly when Francis Collins would not cave in on the access of the data that was supported through the public taxpayer in this country and that is something that we've benefited from. At the same time, there was another culture that has become increasingly stronger in this country and I think the generation looking behind me has pretty much experienced a significant shift in that culture from when they were trained as scientists to now what we see as the business of science and that when we talk about technology transfer and we talk about what's going in the academic environment, I think we are really seeing a significant change of not the Boy Scouts versus, who was on the other end? The Marines. The Marines. Yeah, I'm not sure I would have made that analogy, certainly not today, but rather a grayness of who used to be the good guys and who used to be the bad guys. And part of this is because we grew up many of us in the 60s and the 70s. So did I, so I mean. And so did you. You're wearing blue jeans, right? So, and I don't think there are necessarily good guys and bad guys now because I think we've changed some of our definitions and we've changed some of our context. So what I need to ask you is I need to ask you to look 20 years from now and to tell me what's gonna happen with that culture because when you said we should basically give a blank check to either investigators to do what they think is fine, I think the public would buy that if they thought that they would get a financial benefit from that. But if we give, the public gives us as scientists a blank check and instead we're fighting about whether there is then going to be access to the information. I don't think the trust of the public is going to be what it is today. And in fact it isn't today, and what is it today? So I want you to look 20 years from now and tell me where do you think our public policy is going to be then in terms of the access to the technology and what impact is that gonna have on science? Well I think one of the good news is that if any of the gene patents actually issue in 20 years they're irrelevant because of GAT. So I think time in some level is on our side. And I just want to make it very clear, I'm not advocating a blank check, what I'm advocating is that we think very deeply what proportion of someone's research budget is ultimately discretionary. And just because in nature of the science we do, it can't be the $10,000 we sort of scrolled away from the American Cancer Society grant but it's gonna have to be a larger sum of money. And ultimately you are accountable how you do it. There's an investor in you, I mean whether it's a private sector investment or the government you're gonna have to stand up in front of your peers and decide. So it's not a blank check, it's really an issue of freedom to operate that respects the current dynamics. The grain is, I don't know if it's a bad thing honestly. I don't think that, if we act like Franciscan monks in terms of our science, the public's not gonna care. And given the bucks that we need even in the academic sector to do this, the public has, this is my point, when I made this thing about, when I made the point about trying to raise large sums of money and sort of the ironic thing that the people make a decision about where to spend it, it is ultimately the public. I mean to give the size of the budgets we're gonna need for the NIH and other kind of institute, it has to compete with almost any other big ticket item. So the public has to believe good things are coming out of it and good things are benefits to human health, they are learning and quite frankly, they are things that stimulate the economy and those things are tied together. And part of the reason there is this grainness is that investigators, this is the reason I eventually had to leave Harvard when I was there. At some point you have to decide whether you're a player or you're a spectator in this business and if you can't access capital through one source you're looking through another. And so I think part of the grainness has to do with the kinds of compromises that you have to make an investigator just to do and that's why I think the quasi-private road is one that's probably gonna happen because the market forces are forcing people there. So unless the cost of research gets significantly compressed down to we can work again in small budgets, it's gonna be hard to divorce ourselves from the grainness and I just think it's an inevitable consequence of how things are and this has happened in other areas. The only difference between what happened in big science and physics was they were more relevant to the Cold War and the defense industry so the money cycled in another way. Here we're cycling in what is predominantly a private sector activity which is related to healthcare in the pharmaceutical industry so it just addresses a different set of markets and I think that's the reality of having big expensive technology-based science so I think the grainness becomes more gray, quite frankly. So 20 years is going to be grayer than it is? I believe I think it will be grayer than it is today and I think part of it is there's increased cycling back and forth and it's interesting. We think if I ever told you who the biggest investors and venture funds are, they're the endowment funds at all your universities. It is tea of craft. I mean if you actually look at the composition of your investors, it's all you people and I actually think if you stood up here and asked how many people actually are on the scientific advisory boards of one or a number of pharmaceutical biotech companies, I would think that some fraction of people's discretion and income are derived from it every way so I think it's a very gray area and only gets grayer and the cycling back and forth between is remarkable. I hire people, they're choosing between an academic job or a job in biotechnology and they're a couple of years and their second job is between biotechnology academics, I think that's why it's gonna be gray. Sort of to get back a little to the technology. Sure. There are two things I think that might have been missed over a little bit. One is that there's both observation and manipulation technologies and I think you addressed that very well at the cellular level. Right. But I think for this group we also need to be thinking at the other two higher levels of biological inference, the individual and the population. Absolutely. And the technologies for analysis and manipulation at the cellular level may not necessarily be the same as the ones at the population and at the whole organism level so we need to be thinking about three different areas of targeting our technology. That's a good point and I purposely said I was doing it top, bottom up and would go there but I think the problems just get harder as you get. So that we can start to attack some of those problems. The other one is that actually Francis showed this slide about Tom Watson's comment about how many computers he could imagine, the other Watson, how many computers he could imagine. He was actually correct. At the level of capital investment that was necessary for the way those computers were built at that point in time and the fact that you needed 10 PhDs in electrical engineering to run the darn things, there really was a market of that and what has changed since then is the way that we build our computers and the way that we have the knowledge base for computation as operating systems that are inherent within the machines. So I think if we break out of the mentality that biological instrumentation must be big and must be capital investment, we might be able to get back to the idea that individuals have the capacity to change rapidly just like you can buy a new computer every 18 months perhaps. That same notion of changing our biological technology is a function of how we construct the machines rather than the machines themselves. Yeah and I think what you're saying, the sort of mainframe of the PC mentality you'll begin to see right now in instrumentation. I know several companies or for example have built desktop cell sorters with the size of your computer that probably will cost 50 to $60,000 compared to the equivalent Beck and Dickinson type of machine and I think you'll completely see that trend but that's sort of a natural process of technology, decreasing footprints and costs and I think that will result in the commoditization. I think ultimately at the sort of cutting edge, let's say breadboard level of technology development, that's still gonna remain expensive but it's also why we think we need the private sector because you're not gonna be able to take the device you invent in your laboratory and put it in a shrink-rope and put it in a box and sell it for $15,000 or whatever. So I think you're right, I think that's what's gonna happen and there will be a greater distribution of technology. One of the questions are availability of technology and what species would do we do? Right. I mean do we do everybody's favorite species? And just as a suggestion is that if we follow your model which is, if you'll allow me to paraphrase it, a few labs having most money and the tools that you could develop a ship time model. In other words, if you're an oceanographer you can't afford that ship to go to Antarctica. What you do is you get two weeks on a ship. Right. And so one model I would like to suggest is that a ship time model. That is, you lease me your laboratory space for two weeks and me and five people come and beat the hell out of your equipment. I think it's already happening. For example, one of the biggest advances in extra crystallography and we take advantage of this is the use of synchrotron radiation to generate structures that you can use smaller crystals. In the Bay Area, both Berkeley and Stanford have synchrotron lines, you pay for a subscription and you bring your crystals and you go there. They still do big time research but everyone benefits and I think that is a model that will work, Tom. I'm sorry. I want to go back to the sociology issue that you brought up and that Karen elaborated on and that others have elaborated on. I urge you to actually cast your net even more broadly than you already did and make a couple of points here. One is that the suggestion that the universities get their acts together better in terms of how they think about tech transfer is something that we need to acknowledge is not a costless event and that it is probably, it probably is not true currently that the academy is this pillar or bastion of totally altruistic inquiry. But I think that we need to recognize that as we move more and more to the issue of a more entrepreneurial academy, which we very clearly are in this era, that it does raise profound issues for conflict of interest and for how science is done and how information is shared and other things like that and we need to sort of take that into our calculus as we move in that direction. We also need to take into our calculus that when we move technology or research out of the public sector and into the private sector, that there goes with it a removal from the area of public discourse that I think is incredibly important in our country and I'll just bring a provocative example to place right now which is that if we're not gonna do stem cell biology in the public sector, we will do it in the private sector and what we will lose is the transparency and available to public discourse that we desperately need to have and that in fact the area of assistant reproductive technology has gone some of the ways it's gone because it's been entirely in the private sector and unregulated. So I've been one for arguing that the public needs specifically to address these extremely controversial areas so that we can do it in a way that's open to further discussion and the third point that I would make is and I wanna quote David Botstein on this which will make those of you who know me laugh a lot. But one of the first things I ever remember him saying was that as we move to the big biology and as we move to the small centers, as we move to the huge centers that we will be eating our seed corn and I think we need to be mindful of that harking back to the comment that was made earlier. It is true that I think that the big science will guarantee the small science but I think that to the extent that the big science sort of entrains the younger scientists into the big science system, they're cost-affiliated with that and that's particularly problematic in a time where as we do the biology and as we do the science, we need to think not only about the genomics of individual organisms, we need to think about the interactions between genomes of different organisms, between human and bacteria, between human and parasite and between genomes in the environment. That, I mean, the biological problem is phenomenally huge and so we really, I mean, we need to create the system that allows an even, I mean, I would have painted an even broader skill set than what Barbara painted of what our biologists need and so I just, as we think about the sociology of the change that we have in mind, we really need to look at these profound impacts on all of our institutions and I think they're really complex. Absolutely, so a couple of things that can come out of that. I think in the private sector, particularly if you're a public company, you know about accountability in a way that I think very few people realize there's a good organization called the SEC and there's something more imposing called Reg D and both of those actually have, and both of those make you, as someone said, financially accountable and also legally accountable in a way that you never have to be in a university setting but I think one of the reasons I made this point about psychological components to raising lots of money is that if we as scientists are not sensitive to those, the public backlash will kill you. I mean, I'm old enough to remember when a lot of very smart molecular biologists went to Australia because they couldn't do molecular cloning in the United States. I mean, the same issues are today with human cloning and quite frankly some of the biotechnology companies, CEOs, are doing a very bad job and they only set that particular group of research back both in the public and private sector for many years because remember, you can't buy a car unless you put gas in it. So if you screw up in the private sector and you piss off the public, do you think a venture capitalist or any other sell side investors can put any money on you because there's no way they're ever gonna get liquidity in a public market. So there are self-regulatory things. They may happen through a different mechanism that may be driven by financial incentives but unless we in the private sector, you're in the public sector are sensitive to it, it doesn't happen. So LC issues, public policy issues, yep, it's all part of the same thing. You can't get bucks unless you pay attention to those issues. There's no doubt, there's no doubt in my mind. Hi, so, yes. Hey, Tom. So about the model big versus small, I think it's kind of two questions. One is, I guess when I get up, I can't forget that question. You can sit down. No, the issue, what's big or what's small? There's actually two issues. One is you wanted to dislocate the, bring things away from the funding council, the cycles. But actually, part of the reason of having grants or firefighters is assuring competitiveness. Right. And if you do a big genome standards and so on, and you don't assure a mechanism for competitiveness, you're going to end up with wasting there too. And also loss of innovation, which could money could have gone elsewhere. The other one is I do agree there's a certain level. What's big, what's small? Is it's big having five genome centers in the country? Or is it one per state? Or is it one per city? Or is one for each research institute in the city? Because sometimes you have a feeling some cities want to do that. There's an issue of distance. I mean, we're a core facility in Montreal for probably giving CHIP or genotyping services for 100 groups across Canada. But the groups which are next door obviously get much faster service or the turnaround time with analyzing the data is much faster than when it's in Vancouver. So there's the issues of, and I still think that the better, the Pulitzer Prize papers are going to come up, not just because we do big projects, but because someone who really knows the system has access to technology. So there's got to be a given tape between a big, but so far away or so distance or access would be actually very difficult for people for a real access problem or just a functional access problem. So I just see your thoughts on that. So I think there are two questions. OK, so I think there were two questions in there. Competition and. So it was competition, quality assurance, and quality control. I mean, the bottom line, the reason you need to have some discretionary fund is not only for innovation. It's quite frankly, as I stated in my little overview, was critical mass. How do you maintain a staff unless you give them, I think, Barbies, where it's some certainty, and particularly when you need such diverse intellectual capital to go. So that just has to be there to go on. And we're just going to have to develop different ways of oversight that will penalize people for bad decisions. I think there's one thing around it. I mean, those of us, as Dr. Geyer mentioned, who've been involved in the sequencing program, I don't know how many intermediate-step site visits we do to deal with intermittent problems. And the feedback is there. So the system will respond. My feeling is that we need to think of a strategy that gives every scientist access to this technology. Not every big technology center can afford or should have every single available technology to it, but everybody needs access to it, even as a small scientist, more so as a small scientist. So I think there are issues of distribution of centers, as you pointed out. And I don't know what the optimal numbers are, but I think a lot of it's going to be cost-driven. I think just thinking about distribution of reagents and resources. I made this comment about total synthesis of genes and things like that. Part of that is because you can do, obviously, what you're more interested in is generating variants of genes in conventional molecular biology doesn't work. But if we hark back to the sort of RFLP versus STS genetic marker days, just the physical problems of distributing are simplified if you can make them electronic. So I think we have to think about that aspect too, because I think access to technology is also about simplifying a distribution chain as well. And again, that comes to the sort of private public sector, because, unfortunately, or fortunately, commercial sector is probably a little better if it's incentivized appropriately at assuring distribution. But there has to be oversight of that. So again, I think everyone needs access. I just can't see this. I think this have or have not things will be a very dangerous thing for biology. Yes. Talking just about the public money, what do you think would be an adequate ratio of money for big projects as opposed to investigator-initiated projects? And I heard what you said about, you know, if you have a large-centered filters down to smaller groups, but still, if let's say this proportionate amount of the public money goes to these large centers, the amount that's left for investigator-initiated projects is going to be smaller. And related to the investigator-initiated projects, I agree with you that the time delay between submission of a grant and the time that you actually do it is too big. But I cannot think of a better system than the current peer review system if you do contract or other things, you end up with unintended side effects. So do you have a suggestion for making a full peer review system a little more responsive or less time-intensive? I think we've seen this in the sequence. I mean, I think the NHGRI actually has been a leader in that area. Because by defining, for example, I mean, maybe it's a project-specific thing. But I think in terms of the sequencing project, there's been far more ability to change the rudder of the ship in between grant cycles to move money here, move only there, whatever. So I think, again, it comes back to trusting the people who you're entrusting the money to. That is to give them some flexibility without having to go back to the center of bureaucracy. I mean, there's a big organizational problem. I always hate when process gets in the way of progress. And I think to a large extent that we have these monolithic organizations that are very large that are actually imposing a process that probably makes these things more difficult. So in terms of how much it should be small science versus big science, I can't answer that. I actually think it's not a question. I think the pie simply has to get larger if we are going to find a way of pushing the technology out to a broader number of people. And I think that's the bigger question, is how do we get a bigger part of the pie, not how do we slice some pie into smaller pieces? Among the things I believe you mentioned that the public would like to see from us in applications of the genome project or health care applications, and the problem with getting the private sector to drive that is the vast majority of genetic diseases don't have the market to entice major pharmaceutical companies or even biotech companies. I wonder what you see as the proper roles of the private sector, the government, and academia in taking on that. I think the single most important, the largest selling drug in the world are the statins. And I think it fundamentally was an understanding of familial hypercholesterolemia that sort of led to the whole rationalization of that. I think we're about to see tremendous pharmaceutical drugs that actually will treat Alzheimer's disease because a couple of people a few years ago discovered the presinilins. The bottom line is that the only way that we have today to connect physiology to a gene in a human system is through genetic disease. And that bit of information gives you purchase on the problem that allows you to sort of pursue down an irrationally potentially therapeutic. So in one sense, you're right, big pharma companies generally do not care about rare autosomal recessive diseases. The reality is they probably should care. And the reason they should care is that those are the entry points that really give rise to fundamental biology. And that's, I think, more of an issue in terms of understanding at one level than it is of reality. So that's why I'm forever supportive of whatever research is done in the human genetics community because I think that's the essential information I need to think about developing drugs. Well, I wanted to switch gears from money to research. Oh, good. You mentioned David Gilbert twice in your talk. And we are trying to define what we will be doing in the next 20 years. Gilbert was very successful in defining what all mesigmatism will be doing in the next 50 years. And he's done it simply by formulating 20 open problems. So do you think that genomics really came to the point when we can formulate 20 Gilbert problems in genomics? And so just let everybody work on this. So it's a little bit too early. And if the time really has come to this, then we should really learn a little bit about the standards of the culture of open problems in computational sciences. And the much less known part of Gilbert's work is that he moved to Goettingen, which was the center of German physics at this time, and miserably failed to understand that physics will become the biggest science of 20th century. But what he's really done, he brought the culture of open questions to physics. And physics also followed this paradigm of open questions. So if you really think that genomics has come to the point when we can formulate open problems, then we should learn from Gilbert and probably from physicists what open problem means. For example, let's crack. The combinatorial code of the genome is not an open problem in Gilbert terms. So what do we see? Yeah, I agree. I think part of that, I would be the first person to agree. I mean, I guess my talk or my thoughts is provoked by some of the decisions I have to make in building my current organization. What I've found is that one of the best ways to push people to get to more specificities is to trivialize what they're currently doing. So by eliminating all the tasks that have occupied our time, you create a crisis. And some people will not do well in that crisis. Some people say, well, now I've got to work on the next hard problem, or else I don't have the job. And I think we're just too early. We're at the sort of sticks and stones stage in biological research compared to where we are in physics. We had this discussion last night at the bar. I mean, I think we're back where the pre-Darwinian days, in terms of people thinking about how organisms fall into various phylogenetic trees, and just collecting enough information of sufficient quality that you can actually begin to think of things in a way that is sufficiently quantitative and sufficiently accurate to really have problems in the mathematical sense. So part of the reason for bringing up Hilbert was I always admired that he really came up with the idea of the axiomization of all sciences. And I think really we're at the first stages of building the necessary tools that hopefully will let us get to that point. So I think you're right on target. And that's probably if I can see the motion up here at the place we're going to end. Thank you. I think so.