 What I thought we would do is, and remind everybody to use their mics when you're contributing to this, what we'd like to do is just sort of do a little conversation for the next 20 minutes or so while everybody finishes up eating. And so I'm not going to give a talk, but what I'm more interested in is getting sort of a dialogue. Joe Palco, who's in the back of the room, but will join us shortly, I'm sure, raised the issue that I struggle with and that I know a lot of you struggle with, which is how to talk about this stuff. The concepts are hard. They're not related to anything that's part of our common experience. And the genetics language is both, linkage to this equilibrium, as I find puts me off, along with hat maps and alleles and all the other terms that we have in this. So the terminology is really terrible. So I was wondering if anybody has, for a starter, come up with a good metaphor for talking about any of this, or not that I want anybody to give away any trade secrets because it's certainly we'll all steal from each other as this goes along. But how are you guys sort of talking about or describing these kinds of studies? And I would encourage the scientists who, also between gulps, sort of throw in their ideas, because I think that only sort of by having this conversation are we going to get to some kind of common language that is going to help us be able to explain this to my mother, who still is wondering what it is that I do, along with my director. Does anybody have any particular good insights into the way that they're approaching this, you know, descriptively or metaphorically? That's pretty much my experience, too. I mean, I have to say that I really like Dr. Brody's two sock metaphor for trying to show how one snip is somewhat predictive of what the other one might look like. This is pretty hard. Yes, please. Just a reminder, if you could, just tell us who we are, because I can identify you later on tape. This is Addison Greenwood from NCI. As far as looking for metaphors, I never look any further than Dr. Collins. I just... Collins as metaphor? Just steal and crib on repently. But I do find, more often when you're talking with a group, then when you're writing in, the feedback's not very immediate, that I think one of the things we need to think about is the basic framework that we're using, just inherited versus environmental, can set a lot of readers off on the wrong track. Not to say that we shouldn't continue to use that distinction, because if we abandon it, it would be more confusing. But I think that in a lot of situations, the operational difference can be framed after you say inherited versus environmental. I think that the operational difference often has other components, like a temporal component. Most environmental impacts are after you're born and subsequent... Submarine. I think that rather than making these... What we think are simplistic and categorical distinctions that put people in one box or another, I think that as we go through a piece or go through an explanation, the more often we can add different metaphors that sort of pile up the distinction between one and the other, it becomes more cumulatively explanatory. Cool, thanks. Here's another thought. For the first time this morning, I'm sure it's not the first time it's been said, but the notion of a SNP has a different color door on a house at the same address or whatever. And I just wondered whoever... Probably Dr. Collins that came up with that, but has... He doesn't get credit for everything. There are logical extensions of that that are useful. Has that been taken further? Because I'd love to hear if it has. Francis. Not quite like that. Joe Palca has already taken me to task with a ticking time bomb. So maybe we ought to retire those because they seem a little more scary than an odds ratio of 1.23 ought to convey. But I do think the use of the metaphor about searching for the culprit when you're starting with a very large territory and you need a search strategy that's going to get you to the answer. And the answer is to find the right house somewhere in the entire country. And what the tag SNPs allow you to do is to search neighborhood by neighborhood instead of house by house. Because we have learned that the genome is organized that way and that you can in fact get the answer without having to look at every single house and see if that's the one you can first identify whether you're in the right territory using this property of horrendously named linkage disequilibrium. So in that regard, each of the SNPs is really a house and you're assessing whether you're in the place where the causative SNP, because remember, one of these is actually the causative variation that has a functional effect on the genome by modifying the way some gene in that neighborhood is working. But because of the way in which these neighborhoods are all traveling together, you're going to have a long list of other SNPs that are equally impressive as far as their association with the disease, but they're not doing anything. They're along for the ride. So there's a two-edged sword here. The linkage disequilibrium allows you to scan the whole genome without having to do so darn much work. But when you get a signal, then you're sort of slowed down because it doesn't tell you exactly what the functional variant was. It just tells you you're in the zone where it must be. And then you have to do a lot more additional work to figure out which of the maybe 30 or 40 or 50 SNPs that all look equally good are actually the cause of the increased disease risk. And actually, the causality issue is something that I think is something that we really need to worry about in terms of the clarity of the reporting as well and the way we talk about it from the institute's point of view. I mean, finding a SNP, per se, does not finding the cause of the disease. It's really finding the region that something is going on in it. And there's a great deal more work that needs to get done to really understand sort of the basic biology that leads to the therapeutic and stuff. We'll hear from Dr. Gottmacher later today about the connections between those and where that goes. So we've already seen a lot of headlines of 10 new genes for finding diabetes found, but really it's 10 new SNPs that articulate areas of the genome where something important is going on. So I mean, the precision with which we're talking about this is it, I don't know if I'm comfortable that we really have a firm handle on it. Maybe it's just that I don't have a firm handle on it, but is there a sense that it's more than just me? So one of the other things that was in Dr. Manolio's talk earlier today, and I was wondering how you guys are struggling with this or thinking about it, was knowing when the findings that are in the headlines of the science articles or nature articles or whichever really are real. And it's the sort of the issue of significance. You know, when is a finding significant enough that it's worth reporting on? How are you guys handling that? Anybody have any observations on that or not? So I mean, I'm finding that, and the corollary that I think is going to become important about this is that most of you have stopped writing about individual genes being found. A gene gets found for X disease. It's like, okay, whatever. And now what we're having is, and it was usually for rare diseases, so it was not what's going to be striking your family for the most part. So now we're going to have this shmia of GWAS reports coming out with lots of genes for common diseases. Are we going to run into the sort of the same editor fatigue that people aren't going to want to hear about it anymore because, well, we've just heard 10 stories or 15 stories about common disease genes for heart disease or mental illness or diabetes or whatever. Any sense about how we're going to overcome Joseph, that issue? Yes. Yes, editor fatigue? Yes, quite clear as I think about this, why this should be different in some substantial way from any other problem of explaining science to the public and any other result. I mean, it's not like landing on Mars where you have a fixed event that you can move forward from. It's a process. All the questions that you're asking are relevant, but they're relevant to every set of issues that we cover. But I think, and this is the way I've been thinking about, is that what's different about this story, this GWAS story, is that it's a trend story. And of course, editors love trends, just like they love dogs and babies. But the trend and the trend that's changing here is the sort of the capacity to do the kinds of studies that hasn't been possible before. I mean, it's not just the completion of the genome, which was in 03. Oh my God, that's ancient news for most news editors. But it's also the completion of the HapMap which created this tool that's leading to this explosion. So there is really waypoints, Joe. You can only do one trend story. And then you have to get back to the new hope or no hope. But no, but... Thank you, Victor Kong. You're right, exactly. But the tension is, after you've done the trend story, and then there's 50 small... Incremental advances. ...mileposts on the trend, which one do you do, and when do you do it again? Well, whenever we send you a press release, of course. And by your lunch, absolutely, absolutely. What? No, Francis? It does seem, though, that these will not be all generic stories where you simply do a global replace in your press release and put in the name of the new disease and everything else is the same. As I'll go into talking about diabetes, the things that you discover are themselves, at times at least, kind of interesting. You didn't expect you were going to encounter a zinc transporter that's only expressed in the islet cells of the pancreas when you did this scan of diabetes, for instance. So there's always details in there. Or maybe the phenotype is interesting. Alayne will hint that we're aware of a paper on human height. And that probably will get people's attention a bit more than a disease that most people are not as familiar with in their personal experience. We're all familiar with height in our personal experience. The obesity story that came out two weeks ago seemed to get a fair amount of press. Again, here's a validated gene as opposed to previous claims that clearly plays a reasonable role in obesity. The homozygotes for the risk allele having something like three pounds heavier than the ones who were not. And everybody's interested in that, especially in our culture. So can't you do something with that instead of saying, oh, it's another genome-wide association study of another disease. Make what's interesting out of it the details as opposed to the generic approach. As long as the details are interesting, of course. That zinc thing or that zinc transporter thing you said was just... I was actually heading out the door. So it's clear that what would be interesting to the science types might not always be as similar excitement to the general population. I think that's fair. So that's a sort of a cautionary tale for moderation of expectations, you scientist types. Ma'am, I am talking exactly then. No. Yes ma'am. It's Joanna from ABC again. Yes. I'm slightly different. I think that most people in the room in that we're television and also instead of print and then we are somewhat self-limiting in the amount of time we devote to stories. So frequently a piece on world news will be a minute and 30 seconds where you have to try and communicate the entire story in that short amount of time. And we don't do very many genetic stories at all. Because they're so visual. Exactly. First of all, what do you show? And second of all, how do you even talk about it? If you're talking about a SNP, you have to back up. Most people only have a vague concept of what a gene is and how it relates to a protein and what a SNP is. And so you've used up all your time and more just laying the groundwork to be able to get to the significance of the findings. So we end up not doing anything at all. So that's why we need you to come up with a great metaphor. I'm not sure it's even going to be possible in a metaphor to do it in a minute and 30. Yes, Chris. Hi, Chris Wangek now with NIH. I think the greatest difficulty for me in writing about these things, and I still do outside of NIH is the implication. I don't know if it's a certain moral obligation when we talk about genome. What's it, what's it mean? We're going to hear about disease and genes with diabetes. Diabetes is now the leading killer in Mexico. And 15 years ago it hardly existed. The gene pool hasn't changed and there's such implications with lifestyle, same with obesity as well. You report these things and genes more and more becoming a cop out that we can't do, we can't change our body because of these genes. And that's the hardest thing for me to relay. This understanding I know is supposed to be a balanced nature versus nurture, nature versus molestation, whatever you want to call it. It's, I never understand the balance to report. And what's the long-term implication? We're going to replace these genes with, say, lung cancer, we're smoking. We never said, you know, what is the genetic underpinning of lung cancer? Let's correct it, let's correct it so we can keep on smoking. Maybe that's a fine idea, I don't know. But that was a background for the search for the safe cigarette, which happened all through the 70s. Oh, is that so? Absolutely. So the NCI has spent thousands, millions of dollars on the search, along with the industry, of course, which led to whatever that cigarette was. That was a synthetic fake cigarette in the late 80s. The typical problem for me is how to relay something about the discovery of obesity genes. And I think that's, Chris, you raise a really critical issue, frankly. I mean, we're in the genome institute, so we tend to, you know, everything looks like a gene to us. And of course, lots of things, genes are not predeterminate. I mean, if you have the gene for Huntington's, you're probably going to get Huntington's. That's a bad one, because the penetrance is so high. But the stuff we're talking about now are weak effects that sum up to give you an increased risk. But if you have the genes that will lead you to be obese, but you maintain your caloric intake, you still won't become a beast. Or Amiris, my favorite example is, I may have inherited the genes to be an alcoholic, but if I never drink, then I won't ever become an alcoholic. So there is this environmental interaction that occurs with whatever our genetic makeup is that determines what our trajectory in our life is and what disease that's that we get. But certainly, I would argue that understanding the genetic basis of it is a really important step to being able to intervene in ways in our environment. I mean, medicine is essentially part of our environment to be able to forestall those diseases from occurring and us getting sick. It's critically, I think it's critically important in our stories that we remember that. We certainly try to include that in our press releases and in the backgrounders that we do, although we are very genome-centric. I, you know, fully confess our bias in that regard. Dr. Gamakar. Yeah, I have a couple points from that last conversation. The first is clearly that in understanding the genes involved in disease, often, and most often, is what I was saying, we're not going to change the gene at least in the next many decades. We're going to change the environment. But by understanding how the gene relates to the disease, gives us clues to how to change the environment. Using environment very broadly, including medications, clearly the most famous medication example of that is familial hypercholesterolemia. Now, that's not GWAS site, that's single gene, et cetera. Rare disease. By understanding that decades ago that there was this rare genetic variant that could lead to this huge increase in cholesterol, that explicated basically the causation of atherosclerosis in a way we never understood before. And that told us that G-cholesterol levels really are important in atherosclerosis and that was the birth of statins, which are, you know, among the most widely used drugs that we have, and had a huge public health impact, having nothing to do with genetics, really, in terms of most people that use them don't have variants in hypercholesterolemia. But that is useful to them. Similarly, many of the things that we learn about genes from GWAS studies will tell us some about the basic biology of disease. I'm going to talk later in the afternoon, many of you will know already, the story had an onset macular degeneration. And that was GWAS approach, which came up with several genes that are involved with that, that turn out to be involved in inflammatory pathways. Before that discovery, very few people thought of AMD as the primarily inflammatory disease. But by understanding the involvement of genes there, that is going to be useful for people who have fairly common population, have variants that put them at increased risk for AMD. But also for anybody else who doesn't have those genetic variants, still by having a new understanding of the disease process, genetics will point us to new kinds of interventions and prevention strategies, most of them environmentally, using that term broadly, based. Yeah, thank you very much. I'm going to, we'll whack this up in a minute. There's just two other areas that I want us to, I want you guys to be thinking about, or at least that we're thinking about, or know that we're thinking about. You'll hear from my colleague, Vince Bynum, in a little bit about race and genetics, but we often, we've heard it talked about today, and we've seen it in some of the papers that are in your books, about ethnically homogenous populations. And that's a really complicated subject that needs to be approached thoughtfully. I think it can be done, and it's a powerful tool in genetics, but it's also, it's laden with a lot of sort of social baggage, frankly, and Vince will talk about that, and maybe I'll just not talk about that now and let that be part of it for his session. And the other issue that there's going to be a lot of stories about in the coming months and years as these GWAS studies go forward is looking at phenotypes that are behavioral and linking them to the genomes. These are going to be really complicated, and we're seeing a lot, we're starting to see hints of those kinds of studies already coming out, but its behavior is so multifactorial, and it has so many genes. I don't have any idea if anybody's ever estimated the number of genes that are involved in any kind of neurological organization of the brain. I don't know what the estimate is if somebody else does in the room, but it's clearly going to be a very complicated phenomenon, and we're starting to see some bits of that already, and Elaine will talk, Dr. Ostrander will talk about that in her talk, starting from dog models and extrapolating forward to humans, but one can imagine how loaded all this stuff is going to be, and we even know, I mean historically, if you think back just not that many years ago, there was an NIH director at the Institute of Mental Health, National Institute of Mental Health, who essentially lost his job because he agreed that he defended a study that was being funded at the University of Maryland that was looking at aggressive behavior in primates, and he suggested that this might be a useful thing to study, and you could learn interesting things about it. It seems like an outrageous statement to have made, but it got spun up in a lot of emotion about which populations is he talking about, and who does he mean, and what does he mean by that, and ultimately he was basically driven out of his job. So the sensitivities that are amongst our audiences are quite remarkable, and we can stumble into them dangerously if we're not really thoughtful about the way we approach these studies. So we'll hear about some of the science that's going on in the area of genes and behavior, and this is going to be a very fascinating area, and it's going to be fraught with a lot of conversation, I suspect.