 Thank you, Dr. Gibbons, and good afternoon. So this brief presentation really has a dual purpose. As a follow-on to the last one, I want to provide some details about opportunities for synergy between the NHGRI Genome Sequencing Program and the HL TopMed Program, and of course, this is at a different level than the last talk. Also this presentation is going to be a bit of a vehicle for an update on the Genome Sequencing Program, mostly focused on the Centers for Common Disease Genomics. So I think it's useful to just start off with, by way of introduction, a little bit of a framing, although most of that framing has already been done. In an oversimplified view, and I probably, after the last talk, need to rethink this a bit, but that it is quite oversimplified, I think HL has an interest in any categorical policy, has an interest in specific diseases and genomic resources of importance to their constituencies. And NHGRI has a complementary orientation, it's been discussed. We have an interest in general lessons, how best to discover disease variants across multiple examples and disease architectures. We also have an interest in development of new methods and approaches and community resources that can be used by all. Another point about background is I'm going to only talk about what's going forward from now. In fact, there's been, at least going back now, four years' collaboration with NHLBI, starting in 2012. There's been co-funding and supplements to previous, the previous incarnation of the Genome Sequencing Program, both to the Centers for Mendelian Genomics and to the large-scale sequencing program. I know most of you know this. You've seen this many times, but just as a reminder, the Genome Sequencing Program was revised, reconfigured and re-competed in 2015. New awards were made in 2016. I'll spend most of the time, again, on this new CCDG program, this big circle, interaction with TopMed, but there are opportunities in each of the others, and I'll address them in turns very quickly. The Centers for Mendelian Genomics, I don't think I need to go over this in any detail again, aims to ultimately discover all the causal genes for human, Mendelian and monogenic diseases. NHLBI provides $2 million a year in co-funding for projects of high interest to them. Centers for Common Disease Genomics has a bit of a different mission, two-fold mission, to create a reliable paradigm for identifying variants contributing to disease risk and protection by exploring multiple diseases and disease architectures, looking at multiple designs, looking across the allele frequency spectrum, studying multiple populations. This requires studies with sufficient power to understand when we're done, including done with understanding associations in non-coding regions, which is a particularly hard problem. The other purpose of the CCDG program is to continue the foundational deliverables, and by that, really, I mean improvements in the state of the heart, lowering costs, improving quality, improving analysis pipelines, and so on. The CCDG program effectively works in working groups, as many of these do. Right now, there are three disease working groups. The first one is where all of the, almost all of the specific phenotypes of direct interest to NHLBI resides, cardiovascular working group. The next that are approved and are ongoing are one in atrial fibrillation, that's entirely NHLBI co-funds, early onset coronary artery disease and hemorrhagic stroke. There is a working group on autoimmune inflammatory disease, which has approved projects in inflammatory bowel disease, type 1 diabetes and asthma, and one in neuropsychiatric disease, which has approved projects in autism and epilepsy. All told, current plans over the initial two to three years are for about 80,000 whole genomes and 60,000 whole exomes. Project plans, as Eric Green said this morning, are posted at the coordinating center website. And this is the breakdown of the allocations so far to each of these areas, and you can see roughly half the capacity is up with cardiovascular. A quick look at some details of the atrial fibrillation project. Again, this project was actually proposed, oh, I forgot to mention something. Each of these projects were originally proposed in the applications, and they were reviewed. But in bringing together the consortium, they had to be refined and reprioritized, that the result of that is those posted project plans. And AFib was one of them, but wouldn't have been, couldn't have been prioritized without the co-funding, so we're really, really grateful for it. New sample, you can see the breakdown. It's going to be about 6,000 samples, the breakdown of cases and controls. That's new cases and new controls. New samples will come from 21 studies. African-American samples will be about 40% of the new work. There is already a whole genome sequence data for 2,900 cases done under TopMed, and the plan is to use other cardiovascular disease project controls for analysis. And the co-funding from NHLBI is $5 million a year for two years. I wanted to mention this in particular because this project sort of has dual citizenship as a, it has dual citizenship as both a TopMed project and a CCDG project. And that raises a lot of, it sounds easy to say, but it raises a lot of practical issues, which immediately makes us think about data and how to handle the data, because these data have to get to two different groups with two different sets of expectations that all has to be worked out, and it has to be technically smooth. In the CCDG program, the data flow is pretty simple right now, we're just at the start. This is all, I should say, this entire plan has been worked out and implemented by the CCDG data working group. Harmonized processing will be done at each CCDG. Right now the centers are testing out their harmonized data processing pipelines to make sure that if they get the same inputs, they get approximately the same outputs. There will be combined variant calling with a different, slightly different emphasis at each CCDG center, so some will emphasize single nucleotide variants, some will emphasize structural variants and the results will be brought together. Right now we plan to share data directly within the consortium for initial QC and analysis, including with analysis centers. And of course data will be deposited in DbGaP for wider sharing with the community. The data flow in TopMed lends itself to collaboration pretty easily. I'm just going to go over this again very quickly. TopMed sequencing centers send data to an informatics research center. The data are processed, VCF files go to an exchange area in DbGaP for TopMed consortium analysis, they also go to DbGaP for community analysis, and then the underlying data go into NCBI SRA for the long term. It's extremely straightforward for CCDGs to integrate with this. In fact, the CCDGs already, as you heard already, are producing TopMed data, so they have the right relationships already established. And the PI for the TopMed IRC is also a co-investigator for the CCDG coordinating center, so that works and is already integrated with the data working group. Moving on to the analysis centers. Again Eric Green introduced these at the beginning of the day. Their intent is to come up with novel and creative analyses of the data produced by the genome sequencing program that will cut across individual projects, grants, and programs, and is again not disease specific. They are also asked to help with certain cross program analyses, so work on common control sets, and helping us understand when a project really has reached a point of diminishing returns. Just in practical terms, two of the analysis center PIs are already independently involved in TopMed analysis, and one of them, Dr. Shi Hong Lin, has already organized a joint analysis retreat that's coming up in a couple of months. And to end with the coordinating center. Coordinating center has a dual role also, administrative and logistical support and outreach, but also is taking a leadership coordination role with this issue of common controls, specifying common controls. I think this is going to be potentially extremely interesting, a practical area of interaction between TopMed and the CCDGs, because already all these plans, all the CCDG plans incorporate using controls from other studies, including TopMed studies where they can. So that's really all I wanted to say. The genome sequencing program is an opportunity for collaboration. We're already off to a great start with NHLBI and TopMed. It's easy to find, as was pointed out during the questions just a few minutes ago, that there are plenty of common interests around specific disease phenotypes. The program itself is designed to add new projects over time, and I just want to say finally that this is not the only collaboration. We have collaboration ongoing with NIA. There's co-funding independently. There's funding going to the New York Genome Center for very highly complementary work on autism, and the Centers for Mendelian Genomics also have co-funds from NIA. So I think I'll stop there and ask for any questions. Right. Yeah, so thanks for the presentation. These cross-institute collaborations around human genome sequencing are great, and I'm wondering if you can give us a little perspective on the history and the context of what this TopMed proposal or program now is. Vs IVs, sort of the history of NHGRI. So of course the one that I now think of is TCGA was a previously launched collaboration around the particular disease. Was that the first or were there others, and I guess most important to get to the point is, are there lessons that you can learn from those previous efforts for how what you just described might be designed? Yeah, there are so many lessons that it's from the very low level to the very high level. So yeah, of course TCGA is a great example of very successful and long-standing collaboration. There have been others over time. We have collaborated with NIDDK on some of the diabetes work, and that was very productive. We see many different kinds of successful models. So at some level, each of these is its own thing, but you can see some common features. The most important lesson that I've taken away from all of these together, an overarching lesson, is about the data actually, and it came up just in the last talk. That without a really robust data resource and really well characterized data that's available for analysis, this becomes very hard because it has to be, such a resource has to be kind of recreated with every project, and that's a burden and has to be done slightly differently. The technology is constantly changing, and it ends up getting split up across many different projects. So that's something that in an ideal world, I think that's my sort of number one, this would make it easier lesson. The other is that there's no substitute for just bringing people together. It's an enormous amount of work. People have different, each of these collaborations works differently, and each IC we collaborate with has somewhat different expectations. Maybe what I would add to it is, and I think Adam answered it correctly, there's sort of these very concrete things, especially around data and some of the routines that we just have learned the lessons so that any collaboration we go into just sort of must have. But then the truth is, and I don't mean this at all negative, is that every institute has a slightly different culture, including ours. And every organization has a slightly different culture, and there's, boy, huge cultural differences between even different NIH institutes, and I'm sure people think the same about us, which is fine. And so I think it's that sort of the style with which the extramural programs do their work, and then bringing together these groups. So I think if you've seen one institute, you've seen one institute, because everyone's a little different, and so every time we have different relationship, new collaborations, those elements have to be handled differently. And yet the working groups all end up looking fairly the same and having somewhat the same names and same functions. So we have that template. I think that's very helpful. So the working groups are working? Yeah, yeah, they have to, yeah. Nobody wants to be on another phone call where it's not useful. So, Adam, as I recall when these programs where we're discussed here a year and a half ago or two years ago, the idea was we were going to be phenotype agnostic, and the decisions about which phenotypes to pursue, and I think that you've probably chosen the right ones, was going to be driven a lot by what other institutes were going to step up to the table. So Gary would argue that therefore they should all be cardiovascular. With all respect to Gary, I don't think that would be a great decision. But can you talk a little bit, or do you feel comfortable talking a little bit about interactions with other institutes and why other institutes aren't on this slide? So, first of all, I think it does take a little time. I do want to point to NIDDK because there was quite a good synergy. We are working closely and continue to work closely with NIA on the Alzheimer's E-Sequencing Project, and that does, that will continue. So, and again, NEI has contributed to the, I, yeah, I don't know. Can I make a point that most of these projects were in the applications, were peer reviewed? These are the projects that the applicants worked very hard to bring together. Some of these collections were huge. So, this is not a criticism of you, it's more a question about the culture around other institutes and why they are not investing in making them. So I think everyone will have a slightly different story, sort of what I was answering before. I would point out, for example, that when you have an institute that doesn't have permanent directorship leadership, they have acting leadership. There's a certain conservatism that probably they're not supposed to make long-term decisions, and I don't blame them. I don't think they should be. Notice in my director's report we have two new institute directors starting child health, mental health, that could be very relevant. They have a pretty significant interest, one would imagine, in genetics and genomics, and so that'll be some very exciting discussions when they both come on the scene. As just two more examples. I should also say that we've had good interactions with NIMH as well. Any questions from our phone participants? Anything else from the council? Okay, thank you very much, Adam. Thank you. We're just a little bit behind schedule, but basically in good shape. How about if we reconvene at 1.30, new council members just latch on to one of the old hands, they'll take you up on flight to the cafeteria, okay? And there'll be people in this room, please take purses and wallets, things like that, but you can leave your laptops. See you at 1.30 then.