 My name is Jeffrey Schloss. I'm the director of the Division of Genome Sciences, which is part of the extramural program of NHGRI. I was born in Chicago, Illinois, in April of 1951. My folks lived in Hyde Park when I was born, near the University of Chicago. Shortly thereafter, they moved to the south side, and when I was four, we moved to the south suburbs. About 186 south, so it's 186 blocks south. It tells you exactly where it was. A town called Flussmore. We had a substitute teacher in fourth grade who brought in, and I love this because it's really how these things happen. She brought in a microscope that could project the image onto a screen. And she brought in paramecia, paramecium. And I was just blown away. I was fascinated. It was the first time I'd seen anything, really probably seen a microscope and seen living organisms under a microscope. And I can't say that I have any impression of school before that. And actually, frankly, not much after that in elementary school, but I really thought that that was cool. And I would say that it just really made me interested in biology. And to the extent that when you're a kid, people ask you, what do you want to do when you grow up? That was probably the first time that I had any kind of an idea. Now, of course, growing up where I did and in the environment I did, I didn't know anything about being a PhD. It turns out that one of my parents' friends was a chemist and was a PhD, but I didn't know that until a good deal later. So of course, if I was going to do anything related to biology, that meant you're going to be a doctor, right? Because that's what Jewish mothers like to hear, too. I undergrad at Case Western Reserve in Cleveland and grad school in Carnegie Mellon, which was not known for its biology, but they had a small biology department that they were rebuilding. A lot of people had retired and they were rebuilding and doing some new hires. So we had a few of the older faculty and a few of the newer faculty, including the person who I had done undergraduate research for in Cleveland, he'd actually moved to Carnegie Mellon. In between, he had a sabbatical at Cold Spring Harbor. So that was my first exposure to Cold Spring Harbor. I spent the first six months moving and setting up the lab from Cleveland to Pittsburgh and the second six months of the year in between undergraduate and graduate school, actually working at Cold Spring Harbor Lab, which was very cool. We lived on the lab premises. We were a cell biology lab. We were a lab that was looking at proteins that are responsible for motility in non-muscle cells. And we set up one of the first microscopes that was used for scanning electron microscopes that was used for biology. They'd been used for metallurgy for a while and they'd hired this physicist to come in who actually understood the instrument and we set up a microscope down in the basement of the animal house, which is the one across the grass from the cafeteria. I got my PhD formally in 79 in May. I actually was done after the fall term. So I moved on to my postdoc at the end of 78, but that was the 78, 79. I got my PhD and moved on to a postdoc at Yale. His name is Joel Rosenbaum. He's known for his work and, mostly for his work in Climatomonas. He's done other things too, but about half the lab was working in Climatomonas. And there were other things going on with cells in culture and I actually went to do some mammalian cell culture work looking at some of the other fiber systems related to motility proteins in some of the cells that they had growing there, but within about a year I switched to Climatomonas and from doing a lot of microscopy and some protein work to nucleic acids. I was just interested in the project. We had very little exposure actually to nucleic acids research in the biology department I was in for my PhD, in part because it was a very small department. It's sort of historically oriented. So the lab I was in was almost entirely a microscopy lab and we were moved in next to a protein biochemist who worked in muscle proteins. So it was really a good opportunity. We sort of translated technologies and ways of thinking about the biological questions back and forth between those two labs. So I spent a lot of my time as a grad student trying to isolate protein complexes from nonmuscle cells, try to get them in something of a native state and relate the state of the protein complexes to what the cells were doing that we could observe microscopically. It was sort of a function molecular correlation type of thing. But we had very little nucleic acids work and in fact we had, as they were hiring some people in the last year, so I was there, people would come in and talk about their research using restriction enzymes. So that was my first introduction to restriction enzymes which I really hadn't heard about. And we all thought it was very funny. They were talking about bam and bagel. This was one thing that was a little, so it was very pragmatic but it was a little discouraging and it's one thing that I've sort of pushed against in my career at NIH was that we at one point went to some of our professors when I was a grad student and we said, we'd really like to start a journal club to explore some of the rest of what's going on in biology research because the department is pretty limited. And they said, that's great if you want to do that but just make sure it doesn't detract from your focus on your research. And in one sense they were absolutely right and in another sense they were absolutely wrong, particularly in the limited department. But that really, I would say that one of the themes that I've tried to promote at NIH has been multidisciplinary, interdisciplinary research. We were a very accomplished light and electron microscopy lab. We had, I think, quite deep insights into what was going on in some structural functional relationships with what you could look at in cells that a lot of the rest of the cell biology and even microscopy communities didn't because of the way we used several different kinds of light microscopy with very careful electron microscopy. But it was all microscopy and it was sort of a way of approaching the problem and then to bring in the protein work and the protein biochemistry and start doing functional assays along with the microscopy that was sort of the first step. We wanted to broaden even further with this journal club and I actually don't think we did it. So 84, I left for University of Kentucky in Lexington, an assistant professor job at the University of Kentucky in Lexington in biology. Well, first of all, I have to admit that I did have to look at the map and see exactly where was Lexington because I had been growing up in Chicago and moved to Cleveland and Pittsburgh and New Haven with a little side trip to Long Island. So I had five or six interviews and it just seemed like a really vital department. There were again a bunch of older faculty but a good solid core of younger faculty doing molecular work in several kinds of molecules. We had nucleic acids, proteins, lipids, but it seemed like a very good group and seemed like a good job offer. Well, at that point actually it was far from family because by that point my family had all moved from Chicago to the West Coast. So I wasn't close to family but seemed like a good move at the time. It was largely a good experience. It was only a not a good experience to the extent that I didn't get tenure but that turned out to be a good thing because then I met Jane Peterson at a cell biology meeting while I was looking for jobs and the rest is history. She had a 3x5 card up on a bulletin board looking for program director in the genome center at NIH. Obviously we had postdocs and grad students in the labs where I was and one of the postdocs from the lab in Cleveland I kept in touch with and she knew I was looking for jobs and she said, you really ought to look at NIH. There's some really interesting jobs at NIH being a program director and I had initially dismissed that but then when I met Jane it seemed like a reasonable opportunity and I came here within a couple of weeks I think after meeting her at the cell biology meeting and interviewed and they liked me and I liked them so I moved here. The other main parts of the program where Betty was leading the sort of functional genomics types of things and I think it was clear at that point already that Elise Feingold would be coming in. She started formally a little bit after I did but I think she already had some knowledge of NCHTR because she was in the Grants Associates program and moving from I guess a postdoc position to an extramural position and so I think it was probably already formulating that let's say that she might work with Betty but I'm not absolutely positive about that and then we had the informatics and then we had Elsie. It was pretty small. Very vaguely I would say I was aware of the Human Genome Project. Being an assistant professor I was pretty busy teaching a fair amount. An assistant professor in a biology department teaches whole courses unlike med school where you teach a couple of classes here and there very often not every place so I was teaching a couple of courses and had a couple students in the lab and a couple of graduates and a couple of grad students and I was sort of the chair of the graduate recruiting that is screening at least. I did a fair amount of service even as an assistant professor which may have been part of the problem but so I was pretty immersed in local activities so I was fleetingly aware and heard of it but I can't say that I remember any debates within our department for example about whether the federal funding agencies should be supporting this Human Genome Project nonsense or something like that. Our battles were about very old school physiology and mycology versus modern molecular approaches and how the distribution of teaching responsibilities and other kinds of resources should be distributed within the department faculty meetings and so forth. Well I mean I didn't have any history with this organization to see things changing. Michael was, I mean I'm sure he still had his lab he was around some I just remember him as very personable just a real calm, pleasant guy. I think he thought that the project was in good hands with the leadership of Alka and you know and Mark with Jane and Betty so I don't remember him trying to steer things very much. Clearly there was a search going on but I mean certainly he came, I know he came to some meetings I can't remember if it was weekly or something like that but so I can't say that I remember him as a strong influence on directions we were taking and again I mean the grant, I hadn't any contact with this community before that. Right so I didn't know if there was turmoil because of Watson meeting and things like that but I wouldn't have, I can't say I noticed that very much. The other thing, so in terms of the portfolio I did not have a lot of grants the grants that I was involved in with the centers branch was probably five or six at the beginning I was brand new at the whole business so I really had a lot of ropes to learn. My first grant, so my first day on the job was a Sunday as it always is and there was a pre-council meeting that night and so I went to the pre-council meeting and I had spoken to people a little bit beforehand and there was some discussion of a Drosophila physical mapping grant application I've probably mentioned this to you before Jerry Rubin's physical mapping grant for Drosophila and that was the first grant that was assigned to me so that was cool because I got to hear the council discussion of it obviously that was brought by Jane and then basically work that up as a new grant and see it through its paces but I said one of the things you had asked me about earlier was whether I worked closely with Jane I would say yeah we worked, we were constantly working together but that was actually true of the whole group we were highly interactive and so while I was working most closely with Jane because we were running that centers program and that was the first time I really thought much about milestones people had specific numbers of markers they were expected to map per month I think and then I had to learn all of the stuff about NIH grants from the grant from the NIH side I had applied for grants, I had a grant but it's kind of different from the NIH side and so I had a lot to learn oh I was saying I had just a few grants so this was a center so it wasn't as if I had 20 or 30 RO1s and therefore had a lot of contact with more typical NIH grantee community of academic investigators working on their individual projects instead it was this very sort of highly focused program where people were generally at a more advanced stage in their career they had a pretty big grant from the center's program and they were grinding away to build these maps I would say she was a good guide and mentor she gave me plenty of rope I don't think she let me hang myself but she let me have independence so I don't know why I don't know if that meant I was doing a good job or if she was just purposely saying go do it, it's the best way to learn we never tried to come back and figure that out, talk about that but I found her very easy to work with at that time but as I say when any time we were formulating anything anything we were taking for every application that came through we all talked about we all talked about all the applications that were coming through to NHGRI and we could do that because it wasn't very many program directors frankly weren't that many grants there weren't that many people doing this kind of research at that time we had the genome study section which was formulated with a different perspective than most of CSR review and I think I did realize at the time that that was going on that it was quite different it was focused on the goals of a specific project and so we interacted closely with Cheryl Corsaro who ran that we also had our own review office and as far as I remember the review officers SRAs, we called them at the time and the program staff were pretty much together all the time we almost always ate lunch together because there was plenty of stuff to talk about and we just met in the conference room we easily all fit around a small conference table over on the 6th floor of 38A and it was a very close group and so yes I worked with Jane but I felt that I worked with the entire group well I mean Mark brought the perspective of someone who and I was conscious of that he had been in industry and so he brought probably additional rich perspectives and so he hadn't been at NIH for that long I think at that time I don't actually remember him having been away that much so that's interesting I remember him being much of a presence and he was excellent with crafting language as he has continued to be and so you could struggle with sentences fix them so they just they flowed better and they conveyed the meaning better and he would always ask you whether he was changing the meaning away from what you were trying to get at so he's very good about that he didn't just say this is the way to do it I guess the main impression I had over all those years it wasn't just that year was that this was not a top down and you're a grunt because you're new so you're going to have to do this and this thing and then bring it back and then we'll view it and tell you what you did wrong it was none of that we just worked as a team well so I should say that Francis was one of the center grantees and we had as I said we were the centers were supposed to be mapping certain numbers of markers and so forth and we had two grantees who were not seeing it that way and Tom Kasky was one and Francis was the other and they were coming much more from a medical perspective than the other projects that were probably more I think more from molecular biologists as opposed to physicians and so it was kind of funny that here was this guy who if you will wasn't towing the line and he was the one who was recruited of what the program was not only trying to do and he was recruited to be the director so that was kind of interesting but no I mean he came in and he seemed like very much like a regular guy but very smart and clearly very compelling so in that way I think we I didn't really know him before his wasn't one of the centers that I was directly responsible for that was James so I hadn't interacted with him a lot but I think we could pretty quickly see that this was a guy who was going to be somebody who would go in front of congress and if he remembered to when he traveled here while he was still going back and forth to bring a good pair of dress shoes because he didn't always do that so when he had to go to a meeting I've seen him remember him having to borrow a pair of shoes from people because he hadn't brought a pair of dress shoes I don't know if anyone has mentioned that maybe I made it up it was around so you know I started here in 92 Francis came in 93 and it's by 96 because I know that I helped write with Jane the RFA to preliminary to the scale up it was sort of pilot projects toward large scale sequencing so this was a some of the mapping goals had been met not all of them by any means but the sequencing that was going on was sequencing in order to define STS's sequence tag sites it wasn't really the idea yet of starting to march through chromosomes which was the way people originally thought you were going to have to sequence the genome as you know and so we wrote that RFA and then some time and I'm pretty sure that we got the grants those grants started and I remember having been involved in setting up the first round Robin because there were questions very early on is what's the sequence quality and how are we going to figure that out and all the sequencing was based on backs at that time so it was very easy to exchange templates between centers obviously you wanted to make sure that people weren't working on the same backs so that we weren't duplicating effort but and the backs were mapped so you could say okay we're going to work on this pool of backs and they're going to work on that pool of backs but to have the comparison we'd set up a round Robin of exchange of backs and then people were going to sequence them and there was supposed to be a comparison of the data and it was at that and it was in that era that I there were two people here who had been running the technology development program Bob Strasberg and Carol Dahl and they there were always questions about how much emphasis you should put on different parts of the program that's normal in any kind of research program and Francis wanted was very much we got to start sequencing and the logical argument could be made we didn't have the technology to do that yet and we were near the scale and we need a serious investment in sequencing technology and so I wasn't part of those part of those discussions in any detail but I it's pretty clear that Bob and Carol really wanted to try to push the technologies they were much more interested in pushing on those technologies because they saw huge opportunities and they were in touch with the biotechs and the research community that could do that and they saw big opportunities and Francis wasn't apparently willing to put as much investment as in that as they would have liked to see I don't remember that we had very explicit discussions about some of this but that's my perception of how this went and they moved on to another institute that was where they could have a bigger investment in technologies I don't know extent to which it was sequencing technologies because they moved to cancer and the cancer supported all sorts of molecular technologies over the years but we certainly still had the sequencing technology programs here and so it was they left and we needed to fill that slot and basically Elka and Jane and Mark offered that to me I could continue working with the centers or I could take on that different portfolio and that was really interesting and attractive to me I liked working with Jane but this was a chance to be a little bit more independent and so I just moved I switched my portfolio and that was when we hired Adam Felsenfeld so he actually implemented that round Robin as I remember this yeah so I mean you know this is always kind of rolling evolution the ABI was already in the business of these slab gel sequencers there were with the fluorescence detection the throughput and just the number of samples you could put on those was very limited the sample the lanes were wavy there were all kinds of problems with them because they were still individual people pouring slab gels so I inherited an RFA that was on electrophoretic sequencing to improve that basic methodology and I just look back at some of the things we issued shortly thereafter so this is in 97 was this low cost high accuracy DNA sequencing technologies RFA in parallel with a more generic just genomic technologies program announcement where that was that program announcement all those applications were just out to the community and just send us your interesting genomic methods including as you read the new this 97 RFA including additional improvements on those electrophoretic methods but what we're looking for here is what else is out there what are there are there some other ways to sequence DNA other than the chemistries and the separation methods and so I look to see was some of the things that we funded under this and you know there was still we didn't want to do slab gels but frankly didn't want to do capillaries because that was already coming on to the scene capillary electrophoresis but so these were trying to integrate other kinds of detection methods electrochemical detection methods into the capillaries or figuring out way or using mass spectrometry so you'd get rid of the gels all together or integrating sample prep because these were all still very separate processes you had to grow up your bacteria and pick your colonies and with toothpicks generally at that time it was starting to move toward robotic pickers and robot and arm robots to pick and place arm robots like are used in automotive industry or to assemble integrated circuit chips to assemble sequencing reactions you know using little pipeters or things but this was so we had an application couple of applications there to try to integrate the sample prep with the separation step and with a different way to detect the separated bands so these were the kinds of things that were being funded at that time there was a nanopore that wasn't the first time we'd funded nanopore but there was a nanopore application and then another idea using exonucleases where somehow you would get the exonuclease to chop off a nucleotide at a time and the products of that would flow out onto a surface get attached to a surface and then be detected with some chemical means on that surface and then you'd move presumably your microscope whatever along the surface and read off what basis so this was we were definitely trying to push it that way to different ways to sequence already in 97 so trying to address this question sequencing with Sanger reactions and electrophoresis even capillary electrophoresis was still expensive and relatively low throughput obviously the advent a couple of years later of the higher throughput capillary of the commercial capillary systems where you could actually do 96 at a time and not worry about lanes waving into each other and so forth and do automated loading as opposed to the contraptions that people were building the pneumatic contraptions people were building in order to load 48 or 36 or 48 or 96 whatever samples onto a slab gel where the slab gel was less than a millimeter in between two glass plates I mean that was it was all very hand labor so and this was one of the issues is that it's how long are you willing if you want to do something really different how long are you willing to invest in it and presumably we'll come to that sometime later in this discussion in the next hour or so but if you want to change the paradigm sometimes you know the methods were completely different it was no longer molecular biology everything else was really based on molecular biology and on if you were getting sorry capillary separations where that was known technology it just hadn't been used for single nucleotide resolution DNA separations but mass spec was well established methods it just hadn't been used in this application the electrochemical detection of other kinds of molecules but the question was could you bring some of those things together and apply them to this particular problem where there was clear if you could do it you could conceivably have an impact and then there was a question could you do it in the academic setting and then could that be engineered to a commercial product that would be robust and cost effective the biotech sector I would say is sometimes innovates to the level of actually a completely new idea but there aren't that many completely new ideas they may put together two ideas that are reasonably well established but nobody's ever thought of putting them together that would be a really important thing for biotech to do but some of this some of these methods if it's never been tried for that particular kind of molecule those are things that if you can incentivize them you can get academics to think about it's not to say that biotech wasn't thinking about it too and we supported things in the SBIR program too this would happen to be an RO1s and R21s and things like that mechanisms and at that time I think this was not probably not calling for small business grants not yet but because back in those days we could combine yeah this was R21s, RO1s and program projects we could combine mechanisms we can't do that anymore we can do it just separate program announcements but there are a lot of innovative people in academia too who are more into this kind of methods developments, different communities it's not generally the biology community so my initial grantees were mostly analytical chemists you get a few people who were physicists or bioengineers but in those days this was more from the analytical chemistry community that knew these methods for other purposes and could start developing them it's a little hard for me to know I've heard stories from different investigators about who did exactly what but it's pretty clear that the capillary electrophoresis that ABI commercialized was in partnership with Hitachi and I don't know who did that original research leading to that whether those were Japanese investigators or American investigators but I know there were investigators in the United States who were working on that same problem this is another just general principle is you know we've said in some of our program announcements, RFAs and so forth that we want to fund whoever is the best place to do this kind of research and so NHGRI has funded a lot more work sort of R01 research in company settings than most institutes do because for the tasks we thought were important to get done the commercial setting was actually a really good one but academia also can innovate in those areas so I feel that what we did with the academic grants there was no real possibility there was certainly no real possibility of direct commercialization because this was stuff that was going on in academic labs now some of them may have thought that they could put the plans up on their website which is what happened with some of the spotting for microarrays in the early days but it was pretty, you know these methods were going to have to get commercialized we're going to have to get the licenses we're going to have to get the license we're going to have to get issued from the universities to companies but that's fine so what we were trying to do is sort of feed the early pipeline there's NHGRI academia NHGRI really generally can't afford to fund the commercialization of these instruments but of these systems I should say but anything that we could do in academia with where there's a lot of expertise to add knowledge about the components and whether that was some of the dyes there was a lot of dye research going on part of it was done by Jing Zhu and who's gone on to Columbia and done a lot more of that in Rich Matthews' lab at Berkeley they developed some of the original energy transfer dyes it was done in academia and I can't tell you whether ABI was working on that in parallel these kind of information gets around but I do firmly believe that information that came out of the academic labs helped the companies so there were some experiments maybe they didn't have to do because they saw the results published from academia and they could say ok we've learned from that and now we can go do our own thing so maybe the exact path to their commercial product wasn't funded by NIH but what was learned from the academic labs and that's true for example of these dyes a lot of the polymers that are used in the capillary systems certainly the software probably not a lot about the fluidics because that was a different kind of engineering so I'm completely convinced that it's appropriate for NIH to fund that kind of research to feed that pipeline of technologies the other thing that it absolutely does is it educates the students who then go get hired those students were getting snapped up the students from the labs that were getting funded to do these projects were getting snapped up by those companies because even if they weren't going to bring their specific project and their dye that they had worked on they knew how to do this kind of research yes it is absolutely what was somewhat different about that is it was a bigger lab and he had a sort of different approach to how he was going to do this but he really hired on a wide variety of expertise and it was an environment unlike many others where you could do this kind of research where you really brought together biologists who understood biologists, biochemists surface chemists electrical engineers, mechanical engineers because if you were trying to develop really novel devices you couldn't do you couldn't do it piecemeal you had to have all of that together to be able to tackle some of the problems if it was going to have any chance of getting to an end point it's not to say he did all of these projects perfectly but the basic approach of having all that expertise in one place and where the people could switch around to work on different projects and you had an environment where people were willing to try stuff and fail and there would still be a way to support them to do another project that was that took advantage of their expertise as well as what they learned from their failures in the other projects so you had some long term support and a lot of innovations came out of that group often by somebody going there because they knew that the expertise existed there it doesn't mean that all the ideas necessarily originated with Ron Davis's group that said and there's been a lot of of biotech spin out from Ron's lab because of the environment and the kinds of projects and the ways of thinking about what's needed to advance biology and medicine by groups like that oh that's so loaded I mean I thought I think we had a general feeling that it was pretty annoying it wasn't like he was thinking of something of something brand new he was just going to do it faster better cheaper which has clear impact but it wasn't oh I'm going to sequence human genome that was already going on what he did so there were two main innovations one of which I think originated with his group which was the idea that you didn't have to start with maps that by using paradens and and then not only an idea but a demonstration of the value you could get from doing that molecular biology and following it up with the right kind of software and he hired really good people to do both of those and then his affiliation with ABI to really jump start the use of the high throughput capillary array sequencers and so the impact so that was those were all really good insights important insights into what it might take to do this better faster at least and reasonable quality and probably you know there were people within the public human genome project who were sort of frustrated I think about the degree of emphasis on quality that if we could sacrifice some of that we could go faster but there were very strong voices strong and well respected voices about the imperative to get right at least to a certain level and so I mean I think I think the most important thing he did was light of fire under the public effort it did and this is another place where Francis was quite effective in selling this to get more money for the public effort as well as the way he was able to galvanize international collaborators to try to improve efficiency and reduce redundancy and you know buy into the public data release because this was a very different thing than what Greg Venter was going to do you know and so so brought more money brought perhaps better collaboration cooperation and say okay we can accelerate this even with pretty high standards there was a big push to improve the throughput of mapping using basically old fashioned methods but automation I mean the automation was a big it was a big deal in terms of the applying automation to biology experiments I don't know if it had ever been done before but it was certainly done for the human genome project so it allowed better control and it allowed probably use of smaller volumes and you know better tracking of what you had done so you could see where errors might be you could fine tune and take variability out and so forth so there was automation throughout I don't think Craig particularly contributed to that part but that was an ongoing thing in the centers that we were supporting and I'm sure at Sanger and so forth so I mean the lighting of a fire was probably good the role that Ari Petrino played was a very interesting one to try to to try to bring the two to try to keep the whole thing from exploding to the extent possible to bring the two together at least for the wrap up I mean Craig is an extremely smart guy his style was so different is so different more of the entrepreneur and startup type we have some of those in the public sector too they managed to come across a little bit differently though doesn't mean that they're any less entrepreneurial and have any less vision but they operate differently and may I say piss people off but not as much as Craig does he seems to thrive on it in my opinion where I don't think that would surprise anybody whereas I think some other people will do it though they may not thrive on it as much pissing people off well he also was unabashed about reinventing facts so I mean there were things that we knew inside of NIH that he would explain them very differently I'm not going to go into the details because it's probably old enough that we could and it's probably out there but I'm not going to be the one to I can just say that there were I know at least one very firm example where he explained things in a very different way from what the facts were and so I wouldn't be surprised that he did that in other circumstances but he's not the only he's not the only one who does that and I'm not only referring to current politics so I'm talking about genomics researchers I have to give him credit he's innovated in a lot of different areas of biology I think he's often over claimed what he's achieved but he's very innovative and he's thinking always a few steps ahead of what could be important and I think that's great so the genome project was although that way it wasn't emphasized at the beginning it was still about variation it was about human variation and so the only variation we got was in the back overlaps when the backs happened to be from different human libraries and so you know you could I'm sure there were people who thought we can just get that with microarrays I'm certainly wrong about this but it always seemed to me that if you could get all the kinds of information you wanted from one assay it would be better now that could be a too high bar and I completely appreciate that so you know people would sequence and they would run chips and they would run you know single nucleotide variant chips and you know and you could do all these different chips and each of those things is in aggregate could much cheaper than if you just decided you had to do very very high quality sequencing of everything but if you could get a single technology to give you all of those answers at once then first of all you know just the sample handling you could do it once you don't do it eight times or whatever the number of technologies you have to apply the data analysis should be simpler and if you ever want to make this routine like for medical use you probably don't want to be running all that different number of assays just for the sequence it's not that you're not going to run you know blood chemistries for all the other things for patients you don't want to be running a whole I don't think you want to be running a whole battery of tests just to get the DNA sequence right so that I think was in the back of my mind from a very early day we certainly knew and actually in this 97 RFA we talk about you know there are going to be lots of other things that need to be sequenced for which we're not going to have all the resources like the mapped back libraries the other ways to do sequencing that can be cost effective that can be applied very widely across agriculture and microbes and I don't think we were talking about microbiomes at that time but certainly microbes agricultural important organisms and and model organisms for biology research so so I mean that's that actually I was surprised when I look back at this today in preparation for talking with you about some of the statements that were here it was all it was all there this was nothing new when we got to the planning for after the human genome project and had that series of the huge series of workshops as you know one of which was sequencing and re-sequencing the biome and that was how it was framed there were a lot of other sequencing to do including human variation sequencing and we've got to have ways to do this more cost effectively and accurately and efficiently and quickly you know so there were all the all the ways to make it hard cheaper faster more accurate right well I mean the quality standards were originally set in part for for STS's right is if you didn't have an accuracy so that you could design a primer where you needed it a pair of primers where you needed it to span a region and that was an you know it's an amplicon in this case it was for a mapping STS but you couldn't do the same thing if you had a region where you were looking for variants that were related to to disease if you couldn't design confidently design a primer then your accuracy was too low so right so it had to be accuracy had to be pretty high right but I mean so I mean I think there were costs of the I'm going to come back to something you talked about there were some costs of the acceleration that was prompted by the salara by the venter and tiger and salara which was that I think we could have gotten a better quality sequence the first time around it might have taken a little longer it might have been more expensive but we've done a lot of backtracking I think because because of the race now you could say that the opportunity cost and just get it done once and then let the whole world work on those ways to improve the quality and this is a philosophical argument and you can argue it either way but I think we would have gotten a better quality sequence we didn't get the small indels and we did a bad job on the structural variants on the larger scale structural variants and copy number variants because of the way the sequencing went to shorter reads and this was a big discussion there's no we can get a lot more data faster if we sacrifice read links if we tune the way we use the capillary sequencers to more runs per day but that means we use shorter read links but that meant we lost a lot of kinds of data that I think we could have gotten earlier and so now we've had a backtracking we're still struggling because with the next gen technologies got way worse in terms of being able to collect these kinds of data and so it's fantastic to be able to sequence a genome for approaching a thousand dollars but you're not getting the information you need you can do a bunch of molecular biology we're jumping way ahead obviously where you can do a bunch of really carefully thought out and beautifully implemented molecular biology using 10x technology now and some of the others like that dovetail but you're still, it's surrogate but we don't have the real thing yet we're approaching the real thing with some of single molecule methods which with much much longer reads it remains to be seen whether those will ever be able to be implemented at the throughput and cost that we really do need which is of course then why I would say we need more research but at least we've seen the vision of what you can get with even low quality relatively low quality single reads so people are complaining now you talk, oh you could do Pac-Bio yeah but that costs 20,000 or 25,000 dollars to sequence a genome well it depends on what you're comparing it to are you comparing it to you know $1,500 sequence on Illumina which is still pretty much without some of those add-ons to do the longer range assemblies or you're comparing it to $10 million of the Human Genome Project $15 million for a high quality draft Human Genome Project sequence it's come down unbelievably even if it's $20,000 but again myself box the other thing is there's been a much bigger investment in Pac-Bio commercialization both by Pac-Bio and now by Roche and it's come a long way but it's still not where it really needs to be for the ways people would like to be able to use that sequence but it is single molecule and there's information you can get from that you can't get from from the ensemble measurements including some of the directly getting the better assemblies but directly getting the methylation and that's all not not all worked out yet but it's again it's many steps that you don't have to do you don't get methylation for $1,500 you have to do body cell flight and stuff like that so I think that vision is still worth pursuing it doesn't mean you stop and don't do the research using the methods we have today doesn't mean you stop developing those other methods but that kind of that kind of refinement I think can actually go on in biotech more so again many of the ideas the concepts behind Dovetail and 10X came out of academia they weren't some of them were funded in our organized programs and some of them were just funded by RO1s that the professors applied for and that's good that's why we try to have both opportunities but a lot of that can be done with ideas that emerge the original ideas as well as commercialization emerge in biotechs but if you want to develop a completely brand new way to approach the problem I still think that we need both but I think there has to be a source of funds for academics to do that it can't all be on a three-year timeline it can't it can't be all on a three-year timeline it also can't be in an environment where you can define from the beginning how it's going to be commercially successful which I think biotechs and companies really do have to worry more about so part of the challenge of rolling out these technologies is each time you have a new commercial technology the bar keeps getting higher so the bar for PacBio got higher over the time that PacBio was being developed because of the other next gen's that were coming out and the fantastic engineering that got done in these companies I'm not knocking it at all these brilliant people and working really hard and they've done a fantastic job but this was a different vision but the bar kept getting higher and then similarly for nanopore if you were going to have a biotech come up with the idea of nanopore sequencing today and what was going to be their path to achieve the scientific results and the commercialization success in order to be a commercial the commercialization success to displace other technologies I don't think that's going to fly in biotech and so you need other ways to do it it can be there are a lot of contests coming up now and that's one way to fund things is that whoever wins gets the money to take it the next step somehow they have to get the money to develop that initial prototype but I think it's I've used the term here it's an ecosystem and you need lots of members of the ecosystem and you need evolution and it's fine some things will die off and some things that could have been really good will die off but it's because they didn't fit the niche one of the things I want to mention and so either you were going to get to it or not but there's a question there was 454 and solid and Illumina and what did we fund did we fund any of that did we matter did NHGRI's investment matter well there was some development of sequencing by synthesis for years before that got commercialized and it was just sort of going along sort of slowly and people were struggling with it in a number of the other projects that we were funding and so we were aware of the concept it wasn't really taking off it was ending up being hard because the chemistry is quite difficult actually 454 was the first one to get there could avoid a lot of the chemistry it was pyro sequencing but that had its own limitations as we saw by sort of the demise of that platform but it really was the first one out there and they had a market and they contributed a lot particularly to microbiome science that people could do with that platform that was very much expensive or impossible with other platforms but we did fund well so I mean clearly as you know we funded PacBio pretty much from the moment they were a spin out from the Cornell lab as a small business grant up in New York and then they moved to California we continued funding them for a while and then they came in and got some large grants in our bigger programs but as we were formulating what would become commonly known as the $100,000 genome and the $1,000 genome we were already being approached by some of these entities and we funded a program project grant to 454 and you know we had been funding their predecessor Curogen for a while before that so there were all these other companies that we were funding before Affymetrix and Curogen and yeah in the early earlier days of the genome project even early before I was here so before the early 90s maybe late 80s but certainly early 90s so and this was all again it was just part of this bubbling pot I think is perhaps a way to say it is that we were stimulating that there was an industry there which is the other important thing there is going to be a market for some of this stuff which is important if you are going to have any kind of venture investment but then so Curogen was the predecessor to 454 and it was all Jonathan but so we funded a program project grant and what that was they already had their first platform was going to be rolled out this was before it was rolled out and actually I think at the time they were a little bit nervous about even submitting the application because people didn't know a lot of the details about how that system was going to work but what we funded was the potential for scale-up because they couldn't sequence genomes on that original platform which was their business plan was that earlier platform and I'm sure I know because I've been told over and over again by some of the people working with them that they were able to try any number of variables in their engineering because of the grant and even if those particular variables those paths didn't work out for them it helped them to zero in on the ones that ultimately did work for improving that platform so those are the kinds of things that NHGRI support accelerated their accelerated their path toward their next platform in large part by giving them more flexibility to try things and fail which is okay if we accelerated the development of the platform and also stimulated the field we stimulated competition in the field and completely convinced of that we definitely we funded work in George Church's lab which is where the the allego-ligation method of sequencing was originally developed and then picked up by a spin out from the road Agincourt, we funded Agincourt and we gave them additional money to try, it was a supplement to accelerate that development and that became the solid technology when they were bought out by ABI so we certainly funded early work on that it's impossible to know when these things are bought up by a bigger company they can bring they always bring a re-engineering team and they may re-engineer that platform and by that I include the instruments the software and the chemistry to go into it they may redo everything but that doesn't mean that the investment to show that it was even possible to do this thing wasn't worthwhile and we had much less direct funding of the Illumina technologies we funded Illumina early on when they were a bead hybridization company in fact very early on when it was, people didn't believe that they could decode the beads with oligo hybridizations so helped that get off the ground subsequently we funded a bunch of grants mostly through SBIR to help them with their high throughput oligosynth development of their high throughput oligosynthesis we didn't fund much of the development of the sequencing technology though we funded a project that spun out of Ron Davis' lab and again I think you know a lot of this but I'm saying it so we'll get it on on this history video a project that started in Ron Davis' lab that was being run in part by Moustapher Nagi to develop a sort of a mega-scale pyro sequencer and this was one of those situations where Ron felt that he just didn't have enough money in his NIH grant and couldn't put together money from other grants so he spun it out as a Vantome as a small company and a Vantome applied for a grant from us and before we were able to process the grant all the way through they were bought by Illumina and that became the MySeq not directly but a lot of the fundamental technologies behind some of the detection technologies I think and some of the engineering because MySeq isn't pyro sequencing but a lot of the technology concepts contributed to the development of the MySeq and I know that because that was in their press releases so I'm not over-claiming I'm always very careful about over-claiming the other thing is we funded any number of companies to develop sequencing technologies that never really developed sequencing technologies and we knew that this would be the case sequencing is a really high bar it's really hard to do but there's a lot of other nucleic acids analysis that you need to be able to do for mapping or for not genome sequencing but for forensics for sample prep and I have at least one or two companies in mind for each of those and actually some of these were on my slides that I when I gave my council talk of these were initially funded as sequencing technologies they weren't quite up to snuff to do sequencing but they ended up getting spun out to other applications and commercialized so again I think that the funding if there's a question of should the federal government which we seem to get frequently should the federal government be investing in this kind of technology development I think it's a no brainer and I think we have a lot to show for it I tried to avoid calling it the hundred thousand dollar genome and the thousand dollar genome but I failed because it just got too cumbersome we were talking about two orders of magnitude and four orders of magnitude because it all depended on what your starting point was both for what your dollar starting point was but also for what kind of sequencing you meant and so you know it's different to do sequencing de novo versus resequencing and we said whatever you're doing we're looking for a two order of magnitude cost reduction in five years and then a four order of magnitude cost reduction and it just it did become more convenient and people would it was a faster way to communicate to say the hundred thousand dollar or the thousand dollar genome but there's nothing magical about the thousand dollar genome right it's just a nice round number I learned actually at the sequencing program planning workshop I may have told you this in 2014 of an earlier annunciation it's not the right word expression of the concept of the thousand dollar genome happened at a workshop that Bob Robert Sinsheimer organized at UC Santa Cruz so I don't I have I didn't I didn't find exactly what the date was of this but and people who were there who told me about this were Bob Waterston and Dave Hausler and they actually used that phrase at that workshop and apparently it's in some sort of a report from that workshop I mean I first heard it I think in our planning for the first bookend for the first bookend meeting bookend meetings being the two large planning meetings at early on either side of a series of maybe 12 or 14 or 15 topical workshops one of which was sequencing and resequencing the microbiome so that was like 2001 and I don't remember if I heard it in the planning or if it was at the workshop I know we had some breakout groups for technology development and and it was being used thousand dollar genome was being used there I don't know who first brought it up there different people have different stories about this Dave Schwartz who has been a grantee of ours for a long time who is developing a mapping developed a mapping technology said that he thinks he threw that number out just as something completely ridiculous just saying well what should you know somebody said well what should it cost and he said oh you know a thousand dollars so everybody could laugh one of the ways I heard it expressed and expressed it to press at the time when we started the program was that this is a cost that's reasonable in a medical setting there are imaging tests that cost a thousand dollars that get done on people all the time and so you could envision using sequencing as a practical medical test though it's not a cheap one it's not a screening test but you know I think anybody who talks about it today would say that a thousand it's a great and important milestone to achieve but it's not where we ultimately need to be and again it comes back to something we talked about later is it's only as good as the quality that you get for that cost so I mean would I love to see fifty dollars sequencing but it actually includes long range all the long range phasing information that you'd like to see as well as methylation and maybe a few other things you know whether you could actually incorporate chromosome folding and folding into that I don't know but I think we could always continue striving for that because we're not only going to need to sequence people once that was was sold that way early on but it's just not true for any number of reasons particularly if you're looking at epigenetics because epigenetics is probably going to be a very effective readout of environmental effects perhaps a better readout than having a lot of environmental sensors that people will wear because that's actually so what's the outcome of all the environmental exposures what's it done to your genome and if you need that information you're going to have to sequence people over and over again calling it comparable to an imaging test that you could do for a thousand dollars isn't satisfying so I don't know what people actually were thinking at the time they thought it was a very bold goal and if you could even boldly approach it it would revolutionize the industry which of course it's done but without the bold goal so now you've got your archive of Francis quotes about you know architects make no make no modest plans that's not the word that was used in it it was a great idea I think it was a wonderful idea we didn't originate here at NHGRI but it came from the community which is great because it's actually better for these things to come from the community but I think it was a good one for us to grab on to we were a very small institute still in the in this was mid 90s when with roughly eight program directors and NIH formed this bioengineering consortium this was largely I think due to pressure from the radiology community and also somewhat from the bioengineering community that felt it wasn't getting it was they were having hard time getting grants from NIH getting support from NIH in part because NIH was so focused on hypothesis driven research and these people were trying to develop devices and instrumentation that were critical for medical care but they couldn't tap into the into the pot of money this growing pot of money and so at least in part perhaps wholly in response to that pressure the NIH director's office and this was during Varmas's directorship formed the bioengineering consortium and it was led by the deputy director for extramural research who at the time was Wendy Baldwin and they needed they asked for a representative from each of the institutes for this engineering stuff and well that you know I was a cell biologist by training I am a cell biologist by training but I was the closest thing we had because I was the leading the technology development program shortly after I I think it was formed in 96 or 97 shortly after I took on that role but I was the closest we had so I started attending those meetings and I learned so much from colleagues some of whose portfolios really were about bioengineering most of them were not most of them were people like me who had some part of their portfolio that could be related to developing new methods but through the ways of thinking at we developed some program announcements you have to have a product of a trans-anite panel right so we developed some program announcements that allowed people to explicitly come in with design-driven grant applications as opposed to hypothesis-driven grant applications and we stressed in several of them bringing together a biologist or clinician with an engineer so it couldn't be just some wacky I'm being facetious a little bit provocative wacky engineering idea there had to be some solid connection to a biologist or clinician who would want it and so the development of the engineering idea would go hand in hand with the use of it and so that was some of where some of the ideas that led to SEGs came from interdisciplinary also the idea is as we had with the centers program you had some small RO1 grants but you also had opportunities for people to put together teams that you couldn't sustain or support on a on an RO1 grant and also the idea actually that you should be able to design a project that was not going to go on forever but with larger investments in order to support these larger more complicated teams you didn't want to have to fund that for the next 30 years 40 years for a person's whole career so design a project where you're going to have an outcome and tell us where you're going to get to in 10 years because that's it that's the limit and probably some fairly heavy component of engineering in there with the biology so all these things were formulating at kind of the same time it was through beacon that I got assigned to be one of the NIH representatives to the national nanotechnology initiative this was percolating largely out of NSF and DOE and probably I think DOD maybe a little bit of NIST at that time I'm not sure how important NIST was in the very earliest days that there was this new sort of science emerging that was very hard to get enough funding for and I think people just thought that this was a really good hook on nanotechnology and the idea of control of matter at the single atomic level which was again at that time something of a dream there had been some very rudimentary demonstrations that you could laboriously with an atomic force microscope move molecules into a pattern that spelled out IBM or something like that but then there was some more philosophical but physics based books coming out at the time and Feynman's discussions about the room at the bottom and so forth and there was enough to build on there to build ideas so they form this thing and they needed some representatives from NIH and some of my colleagues I didn't know what was going on in the background some of my colleagues at BCON thought I would be a good participant in that I don't know why but they did and I wasn't, it was sole rep there was a much more experienced bioengineer from Drexel who had been at Drexel and he actually went back to Drexel a few years later who was working in one of the institutes and I'm sorry I don't remember which one Dov was his name I can't believe I've remembered that and he was the sort of the lead on that but I was part of that team and then when he left I became more of the lead so it was all serendipity the serendipity of my getting to take on the technology development portfolio here through that getting involved in BCON through that getting involved in nano and us thinking about what we're going to do after the Human Genome Project and the idea from the SEGs came from Maynard as you need to come back to original research ideas something that's not top-down organized so we formulated sort of an envelope for it but the actual research that people would do under SEGs was completely up to them not up to us, we built an envelope they had to work with of the two your slugs of money and interdisciplinary and things like that What's really funny is that that's so ironic because one of the things that Maynard has and I have spoken at length about he continues to email me about how he hates interdisciplinary work Well, I'm not sure he called for that but we thought that that would because all of the successes of the Human Genome Project up until that time completely depended on biologists and biochemists and engineers and informaticians and if you were going to make any of these fancy systems work you had to have surface chemists because you were trying to work with very small volumes and so you had to have some needle or something that would be able to collect the sample off the bottom so your surface in the tube had to be just right so the sample would go down to the bottom without a centrifuge so that was a lesson that I was actually bringing in part to things I did in Beacon and then subsequently to SEGs from the Human Genome Project so it was really interesting because they all fed off of each other