 The last up is the CIDR concept, and Larry is going to give the presentation. While we're getting set up, some of you know I wear many hats here. And one of the hats, I'm going to put on my LC hat for a second. And just, I didn't have a chance to come up when we brought up resilience. I just want to point out that there is an Office of Behavioral and Social Science Research at the NIH that did a very long series of grants or supplemented long series of grants on resilience. I would suggest I think David Kaufman is not in the room, but you ask him more about that if you want. I can tell you that we did not participate very heavily because there was not a genetics or genomics component to it. It seemed in early days to start looking at things that might be more behavioral. So just to go back to what Malia and Kyle and Hal had brought up. Now I'm going to switch hats. This is a renewal of a concept that Council has seen before. We have a format for renewals, which is why you have a very short concept, a few paragraphs at the most. I'm going to break some of the rules in that this program is not necessarily an extramural program or an intramural program. It's an NIH, trans-NIH program. So I'm going to give a little bit more of a lengthy intro into the history of this program and what it is and what it does. And the title was The Center for Hearted and Sees Research. Many of you in the room and on the line know it as CIDR. I would guess that quite a few Council members have either used or been reviewers for this program. So there's a lot of familiarity online. Here are the concept questions. I've clearly written them as questions that are a little bit hard to say no to, but that's part of my job up here. Feel free to say no if you disagree with me. And really the concept is should the NIH ICs, again this is a joint project with lots of institutes, support in a program to review, and I put carry out in a TALIS because it's not just run samples, as you'll see in a few minutes. Carry out large-scale genomic research projects for investigators. So the investigators that some of you have taken of this program know that investigators put in a proposal and they subsequently get the generation of the data from their samples at no cost to them. It is a cost to the NIH, but not to the investigators. And I'll talk about how we do that at the end for those that aren't familiar. We write this program. This program has traditionally been very flexible, so that is not a rigid program where we decide today we're going to offer Sanger sequencing and we're going to offer Sanger sequencing for the next seven years. Obviously that would not be a good thing ever. And then it is also a program that is structured so it does not require to continue should it not be needed. So there's no out-year funding if there's no out-year need. Just want to highlight those. It's a program that's been going on since 1996. I've been overlooking the program for the last 16 years. And the aims of this program is to speed the discovery of genes that contribute to health and disease. It's mostly looking at germline genetics, although we do somatic sequencing as well. It's currently supported by 10 NIH institutes. I think the low might have been four at the founding. It's gone up to 13 institutes have come and gone over the years as to where they think their portfolio could use this program. We have a very structured way for institutes to enter as members and to exit. You'll note the last one's not really an institute. It's the office of the director. Sometimes a round chair referred to as Building One, but Building One has a significant research portfolio these days at the point where they're actually technically a large institute. So they are also a member. Why would we have a multi-institute program? One is to share expertise, governance, resources, share research projects. It's really easy for institutes to pay half of a project and the other is to pay another half of the project. It's actually very, very easy as our grants management people know because they're not involved. And we also can do economies of scale. So if you're trying to think of what the model is, this really is like a co-op for those who ever were in a co-op. It's their shared effort and shared benefits from being a member of the co-op. There's a governance that's set up for this program that's been going on continuously. There are staff meetings, so myself and Barbara Thomas and one other HCI person meet every other week with the group that holds the contract for this program to help keep an eye on things as well as discuss future directions. There's a site or access committee which I'll describe in a second. It meets six times a year, so applications can come in six times a year. Every other month people can have an application. And then there's a board of governors which is kind of like a secondary view. I shouldn't use that term because it's technically not. It helps to oversight for the program. They also meet very briefly six times a year for looking over what things are put into the queue, so to speak. The program is managed by the NHLBI Budget Office along with the NHLBI Contracting Office. That's inside deals for how the money gets moved around and the paperwork gets done, which is a non-trivial issue. The key is we do it once under one program instead of under 10 different institutes. Right now the services that are offered under this program include consults on the study design, the meat of the molecular stuff in the lab, the SNP genotyping, DNA sequencing. I'll show you the different forms there. Low-pass sequencing is italicized because it's likely to become a service soon. We've been beta testing the use of this. I should highlight that this is a program that is designed for production. So when we bring a service online it has to be bulletproof. It has to work every time, all the time. It is not a program that does extensive weeding edge research into the molecular techniques, and that's because sometimes the projects are 200,000 samples and you don't want to be doing technology development on a sample of 200,000 samples. The program also has the ability to do some statistical analyses and data wrangling, which I think is one of the big undersold parts of this program that I'll tell you about in a few minutes. So the question that we're asking you today is, is CIDR still relevant? There are commercial service providers, there are academic service providers, there are local academic-based cores, and then there's offshore suppliers. So generating genetic data, whether it be sequencing or genotype or even methylation data, has become a commodity. It was not when this program started. This program started to centralize micro-satellite genotyping in a high-efficient way. For those of you that are old enough to remember that, there's a bit of an art and some economies of knowledge that were really important. It is clearly not the case today. If you have a credit card, you can get your genome sequenced. If you have a bigger credit card, you can get 1,000 genome sequenced. So how does this program stay relevant in the light of that easy access to material? And this is how I kind of think of commodity providers. This is no insult to people who run cores in these labs, samples come in, if you need bricks, they will sell you bricks. And that's what your credit card would get. Under this program, we also provide the investigators with some design information, study design information, suggestions. Independent of the people giving study design, there's a review process that looks after the genetic aspects of the projects that come through. This augments peer review from the institutes that usually focuses on the ascertainment of the samples, the phenotyping, the quality of the data, the reason that you want to look at this phenotype. What the CIDR Access Committee does is it looks at, is there a good chance of success for using these in a genetic setting, or a genomic setting. And just to show you the current committee, now is chaired by Catherine Lunetta, and there's some names that some of you might recognize. No one who's sitting in the room happens to be currently on the committee, but I know several of you have been. The committee takes these proposals for short, few pages of an application and reviews them. As I said, we meet every other month so that the turnaround can be quick. In addition to providing the review, the program also provides data quality, both upfront. So the program is designed to analyze the samples for best success. The point that we drive home on that is that if a sample fails, the investigator or the institute is not charged for it. So there are a bunch of upfront quality control issues that make sure a sample, if you're going to put it on a whole genome sequencing, is going to return very good data. Then afterwards, the program is heavily involved in moving the data to data repositories. And I can't underscore how important that is for having full data sharing. And the program has had a lot of experience doing this since the GWAS policy came out when the program was doing GWAS studies. Just to give you an example of some of the interactions with the investigators so that their study design, we help design, the program helps design how the samples should be arrayed in place and does a sample pre-check. Once the pre-check is done, there are checks for the sex of the participant, call rates, pedigree relationships, check known and discover unknown pedigree relationships. There are contamination metrics that are measured on each sample. And the study is Q-seed and looked for batch effects. As some of you may know, there have been some very high-profile studies that were batch effects, not case control. So the current procedure is to randomize the samples after they get to the lab. Then the molecular and the phenotypic data emerge in first-past association. The program does not do the final analysis. It does first-past association kind of studies when applicable to give the investigator an idea of what's going on. And then as I said, the data that are generated under this program are directly deposited to DBGAP or other databases as appropriate. I will tell you somewhat of a mean thing we do in that the data is deposited and the investigator doesn't get the data until it's deposited. So that ensures almost 100% compliance with data sharing policies. The data is not open access to everyone, but the program scales the data and annotates the data so it can be ready to go into DBGAP. Influencing is a big thing these days for those of you that are influencers. I just, that's why I put influence. The ECHO program from Building 1 is using all the facilities and the program itself. Kids first in the Include program programs you may or may not know about from the office of the director are partially using the review process, the exact review process, and the All of Us program is using some of the inspiration from this program as well as the same infrastructure for genotyping. And so it has had, the program has had influence beyond just this site or program itself. To give you an idea of what a long list, a long number of years in the business shows over here, it's like you got shifted on this computer a bit. It's the Institute, the number of projects and the number of samples, unless you're really fast in adding this up, I'll just do it for you. We've been 500 some, 505 projects. This is out of date because it was a couple of weeks ago. And over 1.2 million samples have passed through this program since its initial start. That's led to about 2,000 publications. Also out of date, not included, I think on here, are secondary publications where other investigators have used the data that has been deposited in dbGaP to do additional work. There's 133 dbGaP postings. I found it hard to do the exact calculations, but either one in 10 or one in 12 dbGaP data sets with molecular data came from this program. So it's a pretty big impact on the data that are out there for folks to use. It's used by lots of investigators, 173 over the duration, 93 different institutions. I didn't check that. It's a lot of institutions. 122 phenotypes and publications have appeared in 500 different journals over the years. The mechanism that we're proposing under this concept is that the investigators use this process called an X01 application. Or again, they write a short proposal of what they want to do. It gets reviewed. They will not get funds from it, but they get the service. The back end at the NIH does is called a research development contract of the IDIQ type, which means indefinite delivery, indefinite quantity. Basically the contract says we can order X amount over an indefinite amount of time, or any amount over an indefinite amount of time. And that term in the past has been five to seven years. There is no commitment to use the service if it's not needed. The funding and the administration, at some point I will have to put a cap on how much we expect to spend. There are different reasons for putting numbers in a different spot. This one, based on past experience, will probably be less than $100 million as the cap. That doesn't mean we're going to spend, we're not committing $100 million. And then NHGRI supports, doesn't put funds in it unless we have a project, but we support the program in kind through my effort and effort of a few other people in the institute that I'll show you on the next slide. And our working model going forward is to have this as a full and open competition. That depends on whether people want to continue with this. Just to show you the concept and operations team, you can see there's several individuals from NHGRI that put in different bits of effort for this. NHLBI is the, ironically NHLBI is not a member of this program, so they don't send samples through the program, but they have a unit that does the contract support for us. And then there's also the Board of Governors. Rudy and Ann Fitzpatrick are in parentheses, because when I try to think of things that are a little bit wacky, they either tell me I can't or cannot do them and keep us legal. So again, just closing with the questions, I'll leave them up. And I've already forgotten, I know you're still the discussant, and I think Laura is the other discussant. Laura Beirut and Lynn Jordy are going to start us off. So Laura, are you on the call? I am on the call. So I just want to say that I am fully supportive of this concept. I have been an investigator who's used CIDR, and the services that are given there are really superb. And I think the important component of it are those other type of wrap-around services. It is not just genotyping or sequencing. It has been kind of the study design, the interaction between CIDR and the investigators, help with data analysis and submitting to DBGAP that I think really makes this unique and not able to be reproduced as a commodity. I also want to say that I've seen CIDR transition over the years to newer technologies. And I anticipate that they will continue to be transitioning to newer technologies as other methodologies, sequencing, whatever the next thing comes. So I think CIDR has been very productive, has been designed very well, has worked well, and I support it moving forward. And I think the Institute's supporting it gives evidence of its success. Thank you, Laura. Lynn. Well, I agree with everything Laura just said. I was actually on the CIDR review committee way back when all they did was SNP arrays. So like some of the other folks in the room, I have a fairly long history with them. One question I have, Larry, I noticed that 10 of the 27 ICs contribute to CIDR, given that I think genetics and genomics have successfully invaded almost every area of medicine by now, is it possible that more ICs could lend support? There's not a stock answer to that. So there are some ICs that are now starting, technically the smaller ones that are starting to move into more genomics and genetics in their profile and they're interested in joining. It takes a bunch of back and forth and explaining to them. And then there are other ICs that are large enough that they can just do their own program. In fact, NHLBI, way back even before SNPs in MicroSatellite, had a competing program as well and did have a comparable program for NHLBI grantees up until about four or five years ago, I think, 10 years and then switched into TopMed as being their effort. So there's not one answer to that. We institutes have come and gone over times for various reasons. Sometimes it's the institute director's view, surprisingly, and other times we have institutes that are interested in coming back. So it's one of my jobs is to advertise. One other comment I'd like to make is that while some people do view sequencing and genotyping as a commodity, it's not a commodity that every investigator can actually afford. So that along with the expertise from CIDR, I think one could argue that that helps to democratize access to genetics and genomics. Yeah, and we've had internal discussions as well as with the Board of Governors about maybe pushing this more for trainees and junior investigators and non-highly resourced academic organizations. The requirements, as I understand it, and someone will yell at me if I'm wrong, is that you don't necessarily have to have an active grant to receive a CIDR funding. I think you do have to be in an organization that can have all the infrastructure to receive government grants, I don't know about NIH grants in particular. So that is a way to democratize it. And we are now at the stage, especially with DNA sequencing, where there are economies to do smaller projects. Originally, the projects had to be really big in order to make it efficient. For DNA sequencing, you can answer things with smaller samples. You don't have to do 200,000 or can't afford to do 200,000 whole genomes. I think Gail and then Nancy. We've got Gail, Peter, Nancy, Iftikar. Thanks for that great presentation. I think it's a very valuable program. I don't have any reservations about it. I think one of the real values, the many values that have come from this program is data sets that were very well phenotype, well characterized, that had no genomic data. The investigators then adding genomic data and then that becoming public data. And to that end, I'm wondering how or if you work to make people who are not in genomics aware of your program because with your analysis and design services, it doesn't have to be a genomic study, right? And that's some of who we're looking for as people who have a great data set without genomics. So how do those people find you? Each of the institutes has a liaison. So this is part of why I'm up here and why we're broadcasting is to advertise a little bit more. Each institute has a liaison that if people have projects, they're supposed to go to that liaison. We also go to the extramural program directors periodically and tell them about the program. But I'll be honest, we could do better dissemination. There's part of the problem is you don't want to do so much dissemination that we then can't scale the program up large enough. But we could be better at it. One of the things that you're referring to took a bit of a switch, and I don't know if Barbara wants to comment or not. Barbara Thomas runs the Cider Access Committee. And we realized a couple of years ago that we had two flavors of applications or it could anticipate two flavors. One, I want to find the gene for X or I want to find the genes for X where I want to add genomics to my well-characterized phenotype. We call those resource applications because they're not necessarily as hypothesis-driven and we have revamped how we do some of the reviews to be sure to accommodate resource applications where you want to add either whole-snip arrays or genomic sequencing to a well-phenotype. There you have the potential to potentially answer 15 questions instead of one big one. So spread the word is the other thing I would ask. Peter, go ahead. Yeah, thank you. In general, I'm very supportive of this kind of center, but I have two questions about things that maybe can be improved moving forward. So one of the things you said was that the submitters are encouraged to submit to dbGaP and yet of the 500 projects only 130 did so. What was the problem actually? No, it's not. First of all, a lot of those, that's since 1996, a lot of them were not dbGaP-style projects. Every project that the Institute has made a commitment to submit gets submitted, it's 100%. And a related question, so as a user of data, so I'm a bioinformatician, oftentimes you find that the data could have been improved by the use of consistent standards and so I'm wondering if CIDR could include a component that would promote essentially data fairness. Have you thought of that sort of thing? So consistent standards on... On clinical data such as ontology... So on the phenotyping side. PhenX and things like that. We do push PhenX whenever we can, but part of that comes in from the, usually there's a parent grant that's already established. We could be better probably at the initiation of the proposals on pushing that and there is probably an opportunity to add that to part of the review process. I hope that would be a good thing. Nancy Cox. Yeah, thank you. I just wanted to add a reason, so I remain very supportive of CIDR and a reason that many investigators find it really useful to try to get CIDR applications funded within the context of a grant they're getting funded by an institute. It really helps the budget planning. So you can often avoid the 500k cap issues in projects that would have some genotyping or sequencing that would otherwise put you over the cap when the project goes through CIDR because that part of the project doesn't go into your budget. It means that the institutes have to be more thoughtful and careful about how they allocate their funds because knowing when those costs are going to be used in the context of the given grant years, but it can really help the crazy things that happen when people end up creating batch effects because they have to budget the genotyping or sequencing over five years of the grant instead of having it all done in the first year or the last year or whatever. And so I think CIDR has really improved the quality of a lot of projects and reduced the batch effects by its existence. And that's not a trivial thing for people like us who also use a lot of data from public resources. Nancy, thank you. I won't comment on getting around the 500k cap, but I would comment on another grantsmanship place that plays a role in that investigators can get approval for the project via the CIDR grant review before they put in their RO1 or their funding for the rest of the project. That does give them a stamp of approval, at least for the genomics genetics statistical side of their grants. A lot of investigators do it the other way, too. They get the grant and then they put in the application or do both simultaneously. Because the turnaround is so fast, you pretty much get an answer within two months depending upon where you are in the calendar. You could get it as fast as two months. It does allow people to get feedback very quickly. I was wondering whether you ever had discussions about clear certification for CIDR? Yes, it's a little bit in the weeds for the concept. I will tell you the current existing historical one, we wrote the ability to clear certify into the previous contract. I think Nancy knows a little bit about this as well. I'm all for doing it. We can specify to have it be done. The problem we did with some of the CLIA projects is that folks send a mistake it for a core or a clinical diagnostic lab. Really, the vision would be you have 1,000 people or 5,000 people or 10,000 people and you want to get pharmacogenetics genotypes on them to follow them later. So you need all 10,000 of them at the same time. What doesn't work is I'm going to send you three samples this week and three samples that week. That's better done in a clinical genetics lab. So we could clear certify the process, but not necessarily use it the way clinicians would want to use it as a genetic result. That is not exactly a commodity, but there are other places to have molecular diagnostics like that done. So it gets too much detail for this concept, but that could be on the table easily to have a pipeline that could be CLIA-certifiable. I think it would be neat because you're so far catering to discovery, but there's more and more translation genomics initiatives, and if everything aligned, that could be considered in the next iterations. That's why I wrote it in seven years ago, but it turned out we didn't have very many projects that requested it. Judy. Can you just summarize what your track record has been with respect to underrepresented populations in your various studies? You know, I don't think I've done calculations of that because we do the samples that come in from the proposals. The page, one of the few NHGRI projects that went through was the page program. I'm not even sure we always get, obviously we could do that from the genetic data, but I'm not sure we even always get that phenotypic data up front. So it would be an interesting retrospective thing to do, but I'd much rather just make sure we do better going forward. There are many studies that are 100% underrepresented, but I couldn't give you a tally. And so, for instance, we did a very large project for NCI that was 200,000 samples at CIDR, 400,000 samples total. I'm not sure we know, and it was a large international sample, I'm not sure we know the ancestry or the ethnicity or anything on those individuals. I think the opportunity for this comes in offering this and advertising it amongst other institutions more so than, and to be fair, a lot of the samples that are done were collected a long time ago. One of the which was one of the intramural programs from Debo that was highlighted. Last call for questions or comments. Okay, can I have a motion to approve the concept? Second. All in favor? Anyone opposed? Anyone abstaining? Thank you very much. Thank you, Larry. Okay, we're due for a break. I want it to be full 15 minutes. How about if we reassemble at 3.30? Note how Rudy predicted the schedule and got us within three minutes after four clearances and a report. I'm telling you, this guy is a magician. Don't expect that every time. I am so impressed.