 The purpose is to produce high-quality finished genomes at approximately current costs in the next three years, about 25 to 50 human, and select human non-primate genomes. And I'll talk about quality in a few slides. Also we would expect grantees over time to develop more specific definition of what high-quality finished genome is and to use state-of-the-art methods to maximize quantity and quality. And whenever NHGRI says use state-of-the-art methods, what we mean is we want them to push the state-of-the-art without actually becoming long-term technology development centers. So why? For human, the reference is good but can be made better. There are still gaps, not a lot of them, but there are still gaps. There is incomplete representation of structural variation. We need to move from a mosaic representation to one that represents haplotypes faithfully. And we all know that we can better represent world populations in the reference. When this came up at the workshop, it was considered of such high priority that we should consider moving ahead, even though costs are currently somewhat high individually for these. For non-human primates, primate genomes are key to understanding fundamental questions about human genome evolution. But the existing, almost all the existing primate references are not really reference quality. They're not finished. They could be better. So what is high-quality? Well, certainly well beyond the current short-read-only methods. It's possible to state quality in absolute terms. You can ask for end-to-end contiguity, for example, fully haplotype resolved. You can ask for zero gaps. Out of that, you could have specific lists of known difficult regions and ask that they be 100% resolved. It's also possible to state high quality in terms of the gain in biomedical utility with reference to the uses of the human reference sequence, for example. Practical issues need to be considered because there is a moving state of the art. Whatever language we propose, we have in the FOA, we'll need to ask applicants to consider all of these things because we don't... They all actually do have to be considered in any definition of what's high quality. And we have to give some leeway to encourage creative proposals. There is some relationship to other activities. The NHGRI Genome Reference Consortium Award already funds among its activities, production of five to seven very high-quality human genomes. Some of those are expected to be essentially end-to-end contiguous, so there will be some regions that won't be resolved. And any further effort will need to coordinate with that, not just in terms of production, but in terms of integration into references. There is no overlap for the primary genome part of this. We are certain to need some community input regarding the selection of samples in both cases. We propose this as a cooperative agreement mechanism for a resource, and we think this will require about $2 million a year for three years, and we would expect to make one possibly two awards. And I have to say that this amount of funds, again, comes from our best information at the time of the July meeting of what a current very high-quality reference genome, haplotype resolved, would cost $200,000 to $250,000. But again, I have spoken to people since then who think it can be done for substantially less, and I hope it can be. So I'm going to stop there, and once again go around to the folks who I asked to take a look at this in advance. Lawn, do you want to start? In this case, you wouldn't limit the call to groups that aren't funded under the consortium already? No, this is independent, and this would expressly not be... We would not attempt to make this highly coordinated with the core GSP. I think it's just too much. It needs to be an independent effort. In the past, actually in the distant past, we had these together, kinds of functions together in the same kind of grant, and actually this kind of activity, the high-quality genome activity, needs its own focus to such an extent that I think having it be part or associated with something bigger, it has to be a separate grant. Having it too associated with something more wide-ranging will detract from it. But I don't think that having a grant from the same institution that also has one of the larger grants is a particular bar. It's a very highly specialized expertise. I don't have a whole lot to add to that. I think the questions on this were overlap with other programs which you touched upon, the costs, and in the overall climate of the funding environment, you go one or two. It was less of a scientific case, I think. Anyone on the phone? Oh, sorry. Carol. It's okay. Yeah. What about annotation of the genome assemblies? Is that out of scope for this, or would that be considered in scope? And then reagents to actually then leverage the genome assemblies to do more experimental work. Is all of that out of scope? So what do you have in mind for annotation? Well, you could do computational gene predictions using all the available stuff you have, or you could create CDNA libraries to get better gene models using transcripts or RNA-seq or something. There's a lot of additional data that you could generate that would really improve the quality annotations of these genomes and make them far more useful, but not for this amount of money. Yeah. So I think that goes, in fact, that goes beyond the way I was thinking about it. But what you said made me think not just about the utility, but also it's possible that you could do some quality checking that way. It's possible. And maybe insofar as it was quality assessment, yes, but not annotation. This is really a pretty vanilla production effort. Carlos. So it's being done as a contract, right? No. No, it's not being done. It's a cooperative agreement. Cooperative agreement. For a research resource. I see. So does that mean that, for example, a company that's got an innovative technology solution to do the sequencing, couldn't fit or they could? Yeah, they could. They could. Okay. So we're basically sort of putting out this RFA at $2 million a year to buy 50 high-quality genomes. But there is more to it than that, right? If someone says, I can do it now for whatever per, that might not be as interesting as an application that said, you know, these are the critical factors in getting this feature of high quality. And it is, in my hands, it's more efficient because I'm going to use this combination of techniques. Yeah. I'm not wedded to any particular method, for example, and I'm going to try these. There should be some leeway for that. I see. I see. But at the end of the day, because it's not as tightly integrated with the rest of the program, right, if an applicant or a group were to come in and say, okay, we could deliver 100 high-quality genomes at this price and here's what we're going to do. And, you know, then that's what we're sort of looking for. That's what we're looking for. And of course, they should have some well-thought-out plans for how they're going to pick the individuals with populations they're going to come from with reference to the existing reference and lack of representation or lack of utility or improving utility along a number of different dimensions. I mean, it goes without saying that the diversity component here is going to be really, really critical, right? Particularly if you want to have a bunch of the HLA alleles in those complex regions. Yes. Carol. So you mentioned the GRC. So is it anticipated that they would take stewardship of these genome assemblies? Are they going to be properly resourced to do that task? Because what they're doing now is extraordinarily important to the existing reference sequences. Yes. What they're doing now is extraordinarily important, I agree. There will have to be an interface because if these are good quality, I can't imagine that the GRC will not want to use them. So I can also imagine that there's some boundaries that this may go across for the way the GRC operates right now, so I have to be dealt with. Yeah. So that would be a separate, but again, that would be a separate resourced activity if they don't have the sufficient resources to take these on. Yeah. There wouldn't be additional resources. We'd have to figure out a way within the amount of money we have to get this done if it becomes very important. Yes. I was going to say, I think this is a terrific initiative and it's right on target. Small investment for a big payoff. Val or David, any comments? It's a good initiative. I think it's a good initiative as well. Just to clarify, all these genomes would be independently assembled. Is that correct? I'm sorry, Val. I don't know. I think this is up to the applicant to propose what the best way is. Of what the assembly would be? Yes. If joint assembly improves the quality, then it should be joint assembly, right? Whatever the best possible reference that can be produced. With evidence that you've produced inaccurate. Yeah, yeah, yeah. It's all part of their application. Yeah. I think this is really, really critically important, right? Everybody bemoans the fact that we don't have enough really good high-quality genomes. The other sort of group that would be good to bring in is the sort of genome in a volatile consortium who's been thinking a lot about these issues. I think there's a lot of complementarity between what's being funded here and what they want to do. I think we're ready to vote. Can I have a motion to approve? In a second. All in favor? Any opposed? Any abstentions? Thank you. All right, Adam, last one. And again, we will not be voting on this one.