 All right, so let's get back to business now. As you well know, any institute of NIH is required to get concept clearance before they can issue a funding opportunity announcement. We use our advisory council because it is a public advisory group for that purpose. So there's a series of four concepts that we're going to be working on this afternoon. Three of them are a cluster together, part of the genome sequencing program. Adam's going to start us off. He's going to give a presentation, a summary of the work July 28, 29 workshop on the sequencing program, and then he's going to slide into his concept for the common disease variants. Yep. Okay, Adam? Thanks, Rudy. You'll note on the cover slide for this that this is the condensed version of this talk. You on the ECB should have the longer version. It's about nine slides longer, I think, just to get into a little bit more detail of the high-level points. Given how much interesting and significant I thought was talked about at this workshop, I can only give you a very high-level feel for it. And I hope those of you here who were at that workshop can interrupt and point out anything that you think isn't adequately emphasized here because I'm going to have to just go at a pretty high-level. So what was the workshop about? It started out very general, actually, although with the background of our current programs, talked about what questions, asked about what questions and opportunities in genomics can be addressed at scale, starting with sequencing but not limited to it, and to consider options for future NHGRI programs to address. And at scale has a particular meaning and was talked about at the workshop, so it's not just the answers to the scientific questions that require scale to address, but it's also more what we call foundational aspects, comprehensive data resources, developed technologies, approaches, project designs, analysis methods, policies, file formats, any of those durable goods that will live beyond the effort itself or add to other efforts in the community. And of course, the supplies that they are highly managed for us. We knew that we wanted to acknowledge our history as we went into the workshop but not be bound by it, and we included topics proximate to sequencing that could raise new scientific opportunities. And why now? I think you all know why now? The technology is changing, the availability of infrastructure, including data and analysis tools are more available to many in the world, the community is changing, a different mix of people. There are many more clinical applications. I remember four years ago they were hardly any, and now there are many. And we are in a different era than we have been with federal budgets the way they are, and discussions ongoing at NIH about central planning versus investigator-initiated work. This diagram is just to illustrate the workshop scope, and I'm going to come back to different versions of this in this talk and when I talk about the concepts. The workshop scope, of course, centered on disease gene invariant discovery but had a substantial element regarding genome function, that was discussed as well and completely in bounds, and also a substantial element of clinical applications of sequencing, perfectly in bounds. These two technology development are in parentheses and related informatics are both in parentheses because we didn't include them as standalone topics in the workshop, but we were pretty sure that they would come up, and in fact they did. And you'll hear more about that in a bit. So the major topics, I've divided up into four. There are other ways of dividing it up, but the first, and I'll go through each of these, genetic architecture of health and disease at scale, so discovering variants, conferring risk for common disease, and discovering the genomic basis of Mendelian disease. Second large topic was integrating genomic variant discovery with function. The third was clinical genome sequencing at scale. The fourth, which was actually not a main topic, it ended up being a breakout group that was comparative in evolutionary genomics, and there were some conclusions about that as well. Towards the end of the workshop, we discussed possible features for implementing programs in these areas. Everything, again, everything is online. If you haven't seen some of the talks, I recommend that you take a look at least at the top-level plenary talks and maybe some of the concluding discussions. The meeting was, yeah, there was a Twitter feed. So just at a very high level, scientific opportunities were discover genetic variants underlying human disease and healthy traits, improving understanding of genomic variation through functional genomic studies, continued work in comparative and evolutionary genomics to aid variant interpretation, and evaluate the clinical utility of genomic sequencing and approaches to implementation. In addition to those opportunities, there were two sort of comments that were made a number of times during the workshop and a number of contexts. First, we ought to foster a virtuous cycle between discovery and clinic. And second, we can't do it all, but there are plenty of places where NHGRI can lead and contribute to efforts to ensure collective sequence data is a powerful resource for the scientific community. I'll just go now into the, I'm sorry, because this is such a high level that I feel like I'm skipping over important points. In the area of genetic architecture of health and disease, there were a couple of topics that came up, defining genotype-phenotype relationships underlying human-inherited disease and healthy traits across the spectrum. This, of course, requires scale, requires multiple studies to be done, large sample sizes, requires probably looking at a variety of study designs. One comment that came up a couple of times was it's better to do a few of these comprehensively than try to do many, many of them partially. Another main topic was to enable the knowledge base needed to interpret genome sequence variation in life science, drug discovery, clinical prediction and diagnosis. And the third main topic under this, under this area was to include a range of human diseases and populations to expand discovery, define architecture and broaden access as a matter of social justice. The second large area of discussion, integrating genomic variant discovery with function, several topics to define the molecular, cellular, organ and organismal functions of coding and non-coding genome sequences, to develop tools to manipulate sequences at scale, to develop methods for faithful large-scale functional characterization of sequence variants, and to systematically catalog molecular components and their interactions across cell fates and cell states. There are a couple of details that I think are worth mentioning that happened that came up during this. There was a, we often wonder whether the best way to pursue understanding variant function is to develop assays and sort of wait for the variants to be implicated and then follow them up. And then another way to approach it is to think about what catalogs of systematic, systematically developed functional information you can make in advance. And of course, both of those are important and the workshop discussions reflected that. The third big area was clinical genome sequencing and the topics that came up there were to define clinical context in which genome sequencing improves patient outcomes to enable rapid, robust detection of all clinically relevant variation in a single test. To make sure this is this idea of a virtuous cycle again, to be able to use or enable the use of clinical sequencing data in research, to improve approaches for determining pathogenicity, and to identify effective methods for implementing sequencing into routine medical practice. And again, the idea here was that NHGRI needs to support catalytic research that we can't do it all, but we have an important role. In the final area, comparative and evolutionary genomics, there were several discussions. One was that we ought to produce some high-quality de novo sequenced and assembled genomes, especially for the human reference. The numbers that were thrown around were 25 to 50. These are needed for a variety of reasons, mainly that it's not always possible to map to certain regions newly sequenced genomes. Another idea that came up was that structural variation is extremely important, and this would enable better characterization of it. To obtain nucleotide level resolution of every conserved element in humans and understand specific genomic changes in human and primate lineages, this is obviously very specific kind of recommendation, project recommendations. Third area was to leverage model organisms for functional genomics. And finally, further develop the informatics infrastructure for display, alignment, and distribution of the information. Since the end, again, the discussion turned to implementation advice for variant discovery and clinical sequencing, that we would need flexibility, nimbleness, and clear goals that increasingly will need partners for large efforts and not just partners for providing samples, but more direct partnering, that will need more transparency and outreach, especially if we need to work with collaborators across disease communities, that we need to design any programs so progress is measurable with clear short-term and long-term goals, and that small and large efforts are or should be managed to be complementary, that they're different things and they both have their place. So under the idea that NHGRI cannot do all of this but needs to have an influential role, a lot of concepts came up like doing exemplar studies, doing methods and technology development, building foundational resources, and fostering standards, enabling integration, and other areas. One interesting statement that was made by probably two or three people separately during the workshop that I think all of us heard was that most of the sequence data in the future won't be produced even by NIH funding, NHGRI funding, but by NIH funding, and that we ought to take a role in ensuring that the data will be useful. So overall impressions, our impressions were that the opportunities are as important and as exciting as ever, that what we heard is in line with the NHGRI strategic plan, that there was increased emphasis on complementarity between variant discovery and function between discovery and clinical sequencing. This is the facet of the workshop that's actually hardest to summarize in a list like this, and I do want to emphasize it because we really do think that there were places where those boundaries were being blurred productively. There's still high value in large, highly managed resource generating, technology advancing consortia, again we can't do it all, and we have to explore new ways to establish partnerships, co-funding, any mechanisms that we can come up for sharing costs for these projects. So I'm going to come back to the workshop scope just to orient you for the next slide. I'm just going to talk a little bit about how that maps on to current programs. We obviously have current programs in all these areas, including currently the large-scale sequencing and analysis centers and the CMGs. We have a technology development. We have ENCODE, GGR, functional variation, a number in the clinical area, and a number of informatics related to sequencing. But this sort of is the future. This is what was talked about in the workshop, and this is the workshop wish list. I'll just go over the points again. In the middle is the disease, gene, and variant discovery across architectures and designs. Clinical applications of sequencing and all the functional stuff. We heard this about a recommendation for discussion about comparative and evolutionary genomics and production for gold, very high quality genomes, and maybe some methods development to go with it. And of course there's this ensemble of activities that also figured prominently in the workshop, and I'll talk about those in a bit in the next presentation, along with everything, along with the concept. So I'm just going to stop here. I'm going to apologize for going through everything so fast, because I don't think it does. It didn't do the workshop justice. And ask for people who were at the workshop to please comment. Yeah, sure. I'm curious for this 25 to 50 gold standard genome sequences that you were talking about, the reference genomes. So were there discussions around what kind of population representation there would be among those 25 to 50 genomes? Not specifically other than that. That was an extremely important consideration is what populations to choose. Eric. So Adam, I thought you did a good job summarizing at a very high level. And I also want to congratulate you on the written document. It was a lot to put together you and Caroline Hutter. I think that's an outstanding job. Yeah, Caroline really led on that. So I guess I want to comment on one of the points you made. You commented this idea of discovery to function and discovery to translation or clinical application is the hardest to define the boundary. I believe that was your words. I wonder if we're trying to do something that frankly we shouldn't be doing. We shouldn't be defining a boundary. I understand from the Institute's point of view from a pure funding perspective, there's a reason you need to. But I think intellectually defining a boundary is a mistake that indeed the science gets pushed forward at that interface of discovery to function and discovery to translation and we shouldn't be pushing ourselves too hard to define that boundary. In fact, we should be using this opportunity to build bridges between discovery and function, discovery to application. I wonder if you would comment on that. Yeah, I think I think I hope I didn't come across as saying that the boundaries should be defined. I think the words that were used the workshop in summary, I think by by Eric Green were that the boundaries are being blurred. And you know, as I consider it in context, it's exactly that. It's it's they they ought to be blurred or that there's utility in blurring them. Joe. Yeah, I thought it was interesting and important. I think that came out of the discussion about it wasn't, as you said, listed in the main topics, but the idea that we were finished with the comparative genome aspect and the gold standard kind of genome, it was really came out of the community expressing the need that, you know, these were areas that might have been left behind and thought of somewhat passé. But in fact, the reality is, is that I think those things emerges as really, you know, continue of continuing importance and trying to understand, you know, the aspects of the genome that are really not finished. And I thought that was an additional valuable bullet point that you had there that really wasn't by design in terms of the program. You know, the program, the weight program was structured. Yeah, Joe, are you talking about the, the higher quality genomes or just the comparative? Both, actually. The fact that we, you know, there are important reasons to capture structural variation in terms of understanding disease as well as the comparative aspect in aiding and understanding disease that these comparative genomics really has an additional role to play. And I think we left things at 2X or 8X or something and really there deserves to be deeper consideration of comparative genomics for understanding of disease. Right. And this, this came up especially in the context of primates and understanding lineage specific elements in primate lineage. But it came up in other contexts as well. Eric. This hasn't come up, so I'll say it now. It probably needs to be stated that the, the LC component was not part of the workshop by design. You know, except for Howard, that the community still committed to the LC component. And the Institute is, but, but, but by design, it was not part, it was, it was not part of the agenda. I think that's important that the people watching this don't think we forgot the LC component. It was by design put on to another symposium or topic. To echo those points, I do think it's interesting and important that issues related to social justice came up through discussion. And I think that's a really important point where we need to ask, I think it's important to continue conversations about how the diversity will be represented among those conditions and the study populations. Yes. Also two other points that, that were not emphasized. The importance of a good design and large sample size, which obvious to all here, but it did come up in several presentations, as well as exactly that issue of ensuring that those large numbers represent a broader base than they currently do. Any, any others? Yeah. You, you touched on this, but over and over and over again during the meeting, the adjudication of variants came up as perhaps the central challenge that we have ahead of us. And from a clinical standpoint, what seemed to come up generally was, ultimately, we'll have to grapple with actual outcomes. Yeah. Caroline or any other NHGRI staff, do you want to comment? Do you want to fill in any details? I don't know. Is this on? No. Okay. No, I think that you did it. Again, this was the really high level and we point people to the additional documentation for the details. We had the internal debate about how into details to go today and we really wanted to stay more high level. I also do want to give feedback to people who were at the meeting who've seen a draft to let us know if you have any specific things that you think could be better represented and particularly if there's anything key missing that we're not capturing that you want to be sure is getting into the workshop report. I think that the workshop report was really well done. And it's the high level. Adam, you did a great job, but the report is actually really good and captures. I think all of those subtle points that were discussed that were really important and interesting too. Yeah, it's hard. It was hard to decide what was a subtle point and what was a critical point.