 Good afternoon, ladies and gentlemen. Welcome to the second lecture of First Day of Nobel 35. To introduce our second lecture is the Professor of Geology in this college, my longtime colleague, Dr. Keith Carlson. Thank you, Richard. It's my duty and pleasure to introduce our second speaker, Dean Hamer. As with our other panelists, you could say that Dr. Hamer has credentials to burn. These include academic degrees from Trinity College and Harvard University, a lengthy list of articles, a wide variety of journals. Dr. Hamer is the chief of the section on gene structure and regulation in the laboratory of biochemistry at the National Cancer Institute in Bethesda, Maryland. There he pursues his current research interest for genetics of human behavior. As we've seen this morning, we're amazed and dazzled by the successes and the accomplishments of the science of genetics. How exquisitely satisfying it must be to determine that a trait such as cystic fibrosis or, we saw this morning, its spectrum of other diseases, how satisfying it must be to determine that there is a genetic origin to that trait and to locate the gene on a particular chromosome to determine its base pair code, the protein produced, and finally the physiologic consequences of that protein. The technology and procedures of this are ingenious. The results of such research are definitive and perhaps of immediate important application. One might say to scientists who accomplish this most flattering of words for a scientist, you do elegant research. Dean Hamer has chosen to work on the role of inheritance in human behavior. These traits are more elusive and hard to define. Because the brain and its chemistry are complex, behavioral traits are complicated and unlikely to be controlled by single genes. Furthermore, these traits have large developmental and environmental, that is, nurture components. So the problems of behavioral genetics are more difficult, less tractable. Answers are often less definitive and one could say messier. A scientist would hardly be cheered by the words you do messy research. And the disillusioning idea that some components of our behavior are beyond our rational control has never been good news. It's natural that any loss of free will should concern us. Further, the media and the public may well misinterpret the results of the research as genetic determinism or mistakenly apply them to racial or ethnic groups. So some might feel that the results of the genetics of behavior are somewhat scary. So if the genetics of behavior are messy and scary, why would anyone want to do that? The answer, of course, is our fascination with our own behavior. It's what makes humans human. So we ask, why does Bill Clinton constantly live on the edge? And why does Ken the star pursue him so relentlessly? And indeed, what makes Governor Ventura behave like that? But more importantly, why do each of us behave the way we do? Dean Hamer is a scientist who's willing to do the research on genetics of behavior and has the patience to explain or interpret the results to the media and to the public. He's done this superbly at conferences such as this in his delightful and best-selling book, Living with Our Jeans, co-authored with Peter Copeland. We're pleased to have Dean Hamer here today to speak to us on the topic, genes for human behavior. Thank you very much. It's a pleasure and an honor to be here. And I'll tell you in advance that, yes, we probably can understand the behavior of Bill Clinton. And in fact, I'll talk about some of the genes that Bill Clinton may have. We probably can understand star to some extent as well. But as to Ventura, I would have you remember the preceding lecture where the discussion was of organisms that came from outer space. I actually should start with a bit of a confession, which is that I've been involved in human genetics for only a short period of time, only the last six years of my career. I spent most of my time working on very obscure and ivory tower sorts of topics in molecular biology. My preceding project was one on the regulation of metal metabolism by metallothionine and saccharomyces cerevisiae, which I will not be talking about this afternoon. And sometimes my colleagues and my friends ask me, whatever prompted you to change from such a respectable area of science to studying human behavior, and especially human sexual behavior, which is how we really started in this field. And I'm always reminded of something that was told me by my professor in graduate school on the occasion of the first time that I ever had to stand before an audience like this and talk about research, a much smaller audience, of course, which was just before my thesis defense. And he told me, Dean, you should always remember when you give a talk of this sort that about 1 third of the audience will be listening to you and understanding you. About 1 third of the audience will be listening to you, but not understanding you. And the other third will be pretending to listen to you and having random sexual fantasies. So that, assuming this also holds in Minnesota, it's simply a way of increasing my audience share. Well, as you heard this morning, we are very soon going to know the entire sequence of the human genome. We will know the exact chemical composition of all three billion bases that make up our genetic blueprint. And I want to address two questions that go beyond just the genome sequence. First of all, to ask, what are all those genes doing? Why do humans have to have so many genes, whether it's 80,000 or 140,000 is irrelevant. It's a lot. And what I will argue is that the answer, at least in part, is that many of those genes are involved in our behavior, that they are involved in the way that we think about things, the way that we feel about things, and the things that we do. I'll argue that many of the genes that make up a human are involved in precisely what make us human. For example, the ability to write poetry and to enjoy it, or the ability to build a building like this one or listen to lectures in it. Perhaps even our interest in a higher power in religion and God. And the evidence for that comes really from two lines. One is studying the genome and finding out that many of the genes are expressed in the brain. As again, you heard this morning, the first experiments were done on brain RNA. And the other line of evidence comes from a science called behavioral genetics that I'll be speaking about in depth. The other question that I want to tackle is once we have this information, what shall we do with it? In many cases, looking at disease genes, for example, the answer is obvious. We should use them to improve people's health. But when it comes to something as individualistic, as behavior, the answers are often more complicated. And I'll argue that there is great potential for good, that understanding the genetic basis of behavior can have good effects, both on individuals' mental health and also on our understanding of ourselves and of our society. And then at the same time, there are dangers in this research and there are things that we need to look out for and things that we need to start considering now before the research has actually reached an advanced stage. So let me start out with a general introduction to behavior genetics. It's a science that goes back about 100 years to the turn of the last century, but which has been controversial for all of its life. And I'm gonna emphasize three points. And now if I could have the slides, please. Oh, there we go. The first is that most broadly defined behaviors are to some extent influenced by genes. The second point is that genes predispose behavior, they don't predeterminate. In other words, they may influence a behavior without absolutely determining it. And the final point is that the way that genes affect behavior is quite simply by making proteins and structures and chemicals that are involved in the way that the brain works. Well, how do we know that many different behaviors are influenced by genes? The basic experiments of behavior genetics were both developed by Sir Francis Galton back in the last century. And he described two ways of looking at the effect of genes. One is to look at twins. Twins, for example, identical twins have exactly the same genes, but if they've been separated at birth, often have different environments. Adoptes, by contrast, have different genes, but the same environment. They come from biologically unrelated families, yet are brought up with the same parents. So the basic idea is that if a behavior is genetically influenced, then it will be similar in twins, even if they've been brought up apart, but different and adoptees. That approach has been applied now to virtually hundreds of different types of behavior measures. Here's just one example, which is the personality trait of extroversion. This is a measure of how much you like other people. Do you like to go out and go to parties in the evening, or would you rather stay at home alone reading a book? And what you see is that if you compare identical twins to one another, they are quite similar. They correlate about 40% or so. If one of them is very extroverted, the co-twin usually is as well. If one of them is at stay at home, the co-twin usually will be as well. If you look at fraternal twins who are just like ordinary brothers and sisters and share half of their genes, they are somewhat correlated, about 20%, but less than identical twins. And if you look at adoptees who are biologically unrelated, but have been brought up in the same household, going to the same school and the same church and so on, they're essentially completely unlike one another. It's from this sort of data that one can conclude that many behaviors are influenced by genes. In this case, the level of influence or the heritability as it's sometimes called is about 40% or so. This has been done for a lot of traits. Here are some that I gathered together when I was talking to a group of judges who are interested in the intersection between genes and genetics and the law. And you see that things like petty criminality are linked to genes. Obviously, there's no gene telling people to go out and rob a 7-11, but there's something about people's personality that makes or is associated with doing that that is genetic. Various types of addiction like alcoholism and drug addiction are strongly influenced by genes. Mental functioning, for example, schizophrenia is certainly influenced by genes and even protective factors like a person's cooperativeness and even their sense of religiosity may be influenced by genes. Simply from looking at twins, we know that a lot of different types of behaviors can be swayed by genes. The second point is that although genes are important, they certainly aren't the whole story. For most behaviors that are looked at, heritability is at most about 50%. There's no behavioral trait I can think of that is completely determined by genes. Why is there this plasticity? There are many reasons. First of all, the environment is very important. Where you're brought up, how you're brought up. Those are all key factors. Unique environment, things that happen to you and you alone within your family are also important. And then there's also something called gene-environment interaction. And at this particular conference, I couldn't resist showing this slide of gene-environment interaction, which is petty criminality and Swedish adoptees. This is a study which was done in Sweden and where the investigators looked at two things. First of all, whether or not the biological parents of the adoptees had a criminal record and they took this as a marker of the genetic influence. And second of all, what the home environment into which the kids were adopted was like. Did they have any problems? Did one of the parents become alcoholic? Was there a bitter divorce, for example? The remarkable thing is that if you look at the children when they're 30 or 40 years old and you count up how many times they were arrested, which is well known in Sweden because of the great record keeping there, it turns out that if you have neither of these factors, the rate is very low. Having a troubled environment adds just a little bit. Having parents who had gotten in trouble with the law adds a bit more. But the only time you see a really big effect, a high rate of conviction, is if you have both genes and environment. So in this case, you can't say it's the genes that made them criminals and you can't say it's the environment that made them criminals. It was a combination of both factors acting together. This is probably the rule rather than the exception in behavioral genetics, where genes are setting up a person to respond to a particular type of environment. Well, what are these genes? What are these mysterious factors? For 100 years, we've only been able to say that there is something genetic that is involved, but we've had no idea really what it is. And this is where the future of behavior genetics lies in understanding what the genes are, what products they make. After all, this is what is going to lead to practical applications, such as, for example, drugs, diagnosis, and so on. Well, in thinking about that, we realize it's a complicated problem because behavior is such a complex phenotype. It is the reaction of the entire organism to its surroundings. And it involves not only the genes of the organism, but also the environment and also the development of the brain, where humans are not born with a fully wired brain, we're born actually with a very much unwired brain, and a lot of our development occurs after the time of birth. So the science then really is in trying to relate people's genotypes to their phenotypes. Genotypes means the DNA of a person, the precise sequence of their genome, and phenotype means the things you do, the way you are, and so on. And often in behavior genetics, this is a long and twisted route. It's not a simple one-to-one correspondence. There may be many intervening steps, and so as you heard in the introduction, it is still a messy science, but an interesting one. Again, just to emphasize all the complications, when we talk about genes and behavior, we are not talking about any single gene for any behavior. There is no criminality gene, for example, although there may be many different genes that are involved. We are always concerned about the effect of the environment. None of these behaviors are completely controlled by genes. And I'll also talk about something called pleotropy, which is what happens when one gene does many different things. And again, this is quite common in the brain. I'll give you an example with Sarah Tonin. One last point I should make about behavior is the whole measurement issue. It's one thing to say a person is five foot, 11 high, or weighs 165 pounds. You can be quite precise about that, saying that a person is extroverted or that a person is an alcoholic is a much more subjective matter. And so we always have to worry about our measurements. Given all these caveats, scientists have developed a number of different ways to try to link genes to behavior. For example, in animal model systems, one can actually knock out genes or alter the genes of an organism and see what effect that has. Other people have conducted linkage studies where they look at families and try to figure out regions of chromosomes that go together with a particular trait. The approach that we've taken in humans is to look at particular candidate genes, genes that we already know are important in the brain and try to see what effect they have on particular aspects of behavior. And to illustrate that, I'm going to talk about a gene that's involved in a brain chemical called serotonin. Serotonin is a very important chemical. It's a small chemical in the brain. It's synthesized by a group of neurons in the raffae nuclei. And these neurons extend to almost every part of the brain, the forebrain, the thinking part, the emotional part of the brain, the memory part of the brain. Serotonin is spread out all over the brain performing a number of important different functions. Psychiatrists and psychologists have been interested in serotonin for a long time. They've known about it for a long time. They've done a lot of experiments using drugs that affect serotonin or other experimental manipulations. And from those experiments, they've concluded that serotonin does apparently a lot of different things. For example, it's been invoked in depression and anxiety and suicidality, negative affect, negative emotionality. It's also been involved both in good types of social behavior like social cooperation and what we might call more aggressive types of social behavior like violence and impulsiveness. And also in sexual behavior and libido, as I'll discuss. So I have to confess when I as an ex-molecular biologist looked at this list of phenotypes, I said, psychiatrist, what do they know? Obviously, this can't be right. No chemical could do all of these different things. And so we decided to take a genetic approach to this problem and see what effect genetically manipulating this chemical would have upon people's behavior. Well, we are very lucky in that there is a gene that controls how much serotonin is available throughout the entire brain. It's a gene that codes a protein called the serotonin transporter. And this is a protein that sits on the membrane of serotonin-producing cells and cleans up the excess serotonin from the synapse. And as a result, the amount of serotonin transporter that you make controls how much free serotonin is available to do all these things that serotonin does in the brain. It's a very fortuitous setup in that we humans only have one serotonin transporter gene so that by looking at that one gene, we can actually look at serotonin levels throughout the entire brain. Well, it turns out that the serotonin transporter gene is different in different people. There is a variable sequence, a bit of DNA that differs from one person to the next. And this particular variation is in what might have been called junk DNA a little bit ago. It's in a region upstream of the gene, in the region that actually controls how rapidly the gene is read out or copied into messenger RNA. This variation consists of a repeated sequence. It's about 16 bases long and it has, excuse me, about 40 bases long and it has 16 repeats. It turns out that sometimes people's DNA is missing one of these repeats. Rather as if you were singing, Mary had a little lamb over and over again and you forgot one of the stanzas. Some people have 15 copies and other people have 16 copies and by looking at a person's DNA taking a little smidge of blood or some cheek cells, we can distinguish between those two different forms. Well, it turns out that that little tiny difference in DNA is very important. It determines how well your serotonin transporter gene works. This slide just compares the efficiency of the long form of the gene which is shown in dark and the short form of the gene shown in the hollow bars. It shows that the long form of the gene works about three-fold or two-fold better than the short form of the gene does. This is an assay in some cultured cells but if you actually look at people's brains or I should say the brains of autopsy samples, you'll find that people with the long form of the gene make about three-fold more transporter than the people with the short form of the gene. So this was actually a great opportunity for us to test the effects of genetic variation on people's behavior because we had this genetic variation that altered the levels of this very important protein. Well, what behavior should we look at? Knowing that this variation was very common, about 50% of people have the short form and 50% have the long form, it was obvious that this could not be a gene for a mental illness or some rare trait. It had to be something common and we decided to look at a personality trait called neuroticism. Neuroticism is a sort of catch-all trait. It doesn't mean that you're neurotic necessarily but it relates to being anxious, hostile, depressed, self-conscious, impulsive and vulnerable. This is what we might call negative affect. It's seeing the world through dark colored glasses. It's getting up on the right side of bed. This aspect of behavior and personality has been recognized by psychologists ever since the time of the ancient Greeks, actually. Different psychologists call it different things. Some people call it anxiety and talk about compulsion, shy, suspicious, emotionality. Other people call it harm avoidance and talk about worry, pessimism and so on. These are actually all related measures and they all measure the same underlying trait which is a certain emotional sensitivity. We thought serotonin might be involved in this aspect of normal personality because of the evidence of it being linked to depression, anxiety and suicidality which are really sort of extreme forms of neuroticism. So we did an extremely simple experiment. We went out, we collected several hundred people. They're actually all siblings. And then we gave them a questionnaire that asked people about neuroticism and we also took a sample of their DNA. These are the sort of questions that you ask. I am easily frightened. I rarely feel fearful, anxious. I often worry about things, et cetera. And you simply check off a little check mark showing how much you feel about these things. When we tallied up all those results, we got the data that's shown on the next slide. We're here, we are looking at two different groups of subjects from different sources and we're looking at five personality traits. Neuroticism, the one that I just mentioned, as well as extroversion, openness, agreeable and conscientiousness. This is an inventory which is supposed to separate people's personality into these five major traits. And then we compare people with the short version of the gene or the long version of the gene and ask, are they different for any of these personality traits? When we did that, we found there was a difference and just as we might have predicted, it was for neuroticism. People with the short form of the gene scored about four units higher than people in the long version of the gene. We repeated that on a new set of individuals and got the same result. We've now repeated this experiment on additional populations in the United States, in Israel and in Japan, and in every one of those cases have seen the same trend for people with the short form of the gene to show higher levels of neuroticism and anxiety. So what does that mean exactly? Well, it means that the gene is involved in this trait just as one might expect. But how much is it involved? How determinative is it? If you have this gene, does it mean that you are consigned to a life of sadness and pessimism? Well, it's more complicated than that actually. This slide shows the level of neuroticism that people have as a function of whether they have either the long gene or the short gene. As I told you, having the short gene on average increases neuroticism by about a half a standard deviation, which is quite significant. But there's also a tremendous amount of overlap between the curves. There are people with the short gene who are perfectly happy and optimistic and people with the long gene who are sad, depressed and worried about things. So the gene is having a quantitative effect. It's tipping the scales, but it certainly is not an absolute effect. Why is there so much overlap? Well, probably in part because there's at least 10 different genes that are involved. So if we knew the other nine genes and we're beginning to learn some of them, we could separate the curves a bit further. But beyond that, there's the environment, which of course you can't measure by any genetic test, which is also important. So that there will always be some overlap between these curves, no matter how much of a person's genetic information we understand. We were emboldened by this. It was in a way an expected result since we knew what serotonin was doing or we thought we knew and since we knew what this genetic variation was doing, but also an exciting result because it was one of the first times it was really possible to link a specific genetic variation to a specific behavior in humans. And this emboldened us to start looking at some of the other aspects of personality and behavior that serotonin had been involved in. For example, we've looked at seasonality, which is a measure of how much your behavior and your feelings change when the seasons change. I suspect that you living up here in Minnesota, perhaps many of you, noticed that when you're going to school in September that you start feeling kind of gloomy and you may be thinking, oh, this is because I'm going back to school, but you may well find when you get older and you're working continuously that in fact it's something about the change in seasons that affects the way you feel about things and personality. For some people it can also be reflected in changes in eating habits, eating more carbohydrates, sleeping habits, sleeping a longer period of time, sort of going into a mini-hypernation, if you will. Well, it turns out this is also affected by this gene. People with a short form of the gene are more seasonal on average than are people with the long form of the gene. Again, a lot of overlap, but also a clear effect of the gene. What about social personality traits? What about social cooperation and some other traits that had been invoked with serotonin? We looked at a personality trait called agreeableness, which covers areas like whether you trust in other people, whether you are straightforward and honest to other people, compliance, which is how much you obey orders, all measures of social cooperation and how you get along with other people. It turns out that this trait also is related to the serotonin transporter. The same gene that makes people high for neuroticism makes them low for agreeableness. In fact, when you calculate out all the numbers, the interesting point is that although neuroticism is only, excuse me, although the gene is only responsible for a few percent of the variation in either of these traits alone, it's actually responsible for about 10% of the co-variation. To put that in simple terms, if you wonder why people who are worried and anxious also tend to be rather disagreeable, this gene is a good chunk of the reason why. We've begun to look at the effect of the genes on more concrete behaviors, not just the way people feel or the way that they check off a questionnaire about themselves, but actually the things that people do. One behavior we've been very interested in at the National Cancer Institute is cigarette smoking. Cigarette smoking is, of course, a complex behavior. It involves trying your first cigarette going on to become a regular smoker, often becoming addicted, also involves quitting or not quitting, relapsing, and so on. But there was quite a lot of information suggesting that this complex behavior might be affected by serotonin. The main evidence, really, is that people that are anxious and depressed often have a very tough time in quitting smoking. In fact, a lot of smokers, including myself, until I finally quit, would smoke when we were feeling blue about things or feeling worried about things. Nicotine is actually a great drug for that purpose, except for the side effects. Well, we looked at the effect of this gene on smoking behavior and it's also involved in that. In fact, in this case, there's an interaction between the gene and neuroticism so that people that have both the short form of the gene and a high level of neuroticism have the roughest time in quitting smoking and are most likely to be smokers. For example, when you look at people's ability to quit smoking, there's a much, much greater correlation in people with a short form of the gene that's reversed than in people with a long form of the gene. We're also interested in the possible effects of serotonin on sexual behavior. There were experiments in animals suggesting that serotonin was involved in sexual frequency. There's actually some evidence that this might be in the case in humans, too. That evidence comes from the drug known as Prozac, which probably many of you have heard of. Prozac is a specific serotonin reuptake inhibitor and it actually does the same thing that this gene does. It varies the amount of serotonin that's taken up at the synapse. Serotonin reuptake inhibitors like Prozac are very effective antidepressants and, of course, that's what they're marketed as, so that fit in with what we know about the gene and neuroticism. But they have a side effect, at least in some people, and that side effect is to reduce sexual behavior, to reduce libido, and then sometimes to actually reduce the ability to have an organism. Orgasm. That was probably Freudian. That was not Freudian. Reduces the ability to have orgasm. Anyway, we decided to look and see if the gene had the same side effect as the drug has. And it turns out that it does. Here I've just split the world into people that either have the short form of the gene, the long form of the gene, or a heterozygous. Then I've split the whole population into those that have sex less than once a week and those that have sex more than once a week, which is sort of a natural dividing line in our data. So as you can see, genotype has an effect. For example, people with a long form of the gene are mostly having sex less than once a week. People with a short form of the gene are more often than not having sex more than once a week. Well, I expect several of the high school students now have suddenly woken up and are trying desperately to remember, short gene, long gene, what does it all mean, how to stick that together. It's very simple to remember. If you're happy, optimistic, and everything seems fine, that's great, but you're not having sex that often. You're all worried, anxious, et cetera. You're having sex more frequently. Actually, this effect has nothing to do with personality. It's mediated independent of personality, probably because serotonin acts in so many different parts of the brain. It happens to have this effect in the part of the brain that's involved in the libido and those other effects on personality in different parts of the brain. It is actually an interesting finding, I think, in that one evolutionary puzzle was, why should mother nature have given 50% of people, this gene that makes you feel miserable, depressed, worried, pessimistic about the future? And of course, the answer is that the gene really doesn't care how you feel about things. As long as you're doing what's on this slide, the gene will be passed along, and so will continue to expand. Well, that's the main gene I wanted to talk about, but let me just give you a brief rundown on a few other genes and brain systems people have been studying, including ourselves. For example, another brain chemical of interest is dopamine. Dopamine is the exact opposite of serotonin. It's the brain's feel-good molecule. This is what makes you feel good after a great meal or sexual experience or smoking a cigar, all three of those and et cetera. We were interested in the role of this gene in novelty-seeking, a personality dimension that relates to how much people like doing new things compared to how much people are satisfied with the tried and true. And the gene that we looked at is one called the dopamine D4 receptor gene. It codes for one of the many receptors that receives the signal from dopamine and was of interest to us because it also has a repeated sequence that's variable from one person to the next. And it turns out this variable sequence is actually in the coding part of the protein. It affects the type of receptor that you make and that affects how well the receptor interacts with dopamine. What we found is that people with long forms of this gene have higher levels of novelty-seeking than the people with the short form of the gene. That's been replicated in some populations, although not in many other populations, and there's been some speculation that might be an accidental finding. More recently, though, that same long form of the gene has been linked to a tension deficit disorder, which of course is, in a way, a sort of pathological form of high novelty-seeking. And that's been replicated quite consistently in many different populations by many different investigators. Just since we have been talking about sex a little bit, I should say that this gene also has a sexual side effect, not on the frequency of sex, but on the number of different sexual partners. We ask everybody in our studies how many different males and females they have had sex with. It turns out the gene has an effect in a rather interesting way. If we look at heterosexual males, straight guys, and ask how many females have you had sex with, the gene actually doesn't have any effect long and short of the same. But if you ask these same heterosexual males, have you ever had sex with a male? And some of them have, for example, in college, after too many beers last night. It turns out the gene has a really big effect, and people with a long form of the gene are much more likely to have had sex with another male than people with a short form of the gene. Quite the opposite is true for gay men for whom male partners are the usual partner. The gene has not too much effect, but when you ask how many females they've slept with, the gene has quite a large effect. So again, a sort of sign effect of the gene, it's not that this gene is whispering into one's ear, have sex with him, have sex with her, et cetera, but rather the gene makes people interested in new stimuli, new phenomena. And of course, for a heterosexual man having sex with another male is what's new and unusual and just the opposite for a gay male. One other trait that we've been interested in, and in fact, the place that we started our research, is sexual orientation itself. That is, whether people are not, are stably attracted, either to members of the opposite sex, which of course is heterosexuality, or members of the same sex, which is homosexuality, or something in between. The majority of people, of course, are heterosexual oriented, but there's a substantial fraction of individuals that are homosexual oriented or bisexual, so there's variation in the trait. We were interested in whether or not there's any genetic basis for that. We did a lot of family studies following up on twin studies done by others, and found an interesting pattern in families, the pattern shown here, where when you look at gay males in a family, they're most often clustered on the mother side of the family. For example, in this family, the gay males, maternal uncle and maternal great uncle were both gay. That was interesting to us because it suggested a possible genetic basis, which would be X chromosome linkage. Remember that males only have one X chromosome, their other sex chromosome is a Y, which they must get from their father, and therefore genes that are on the X chromosome tend to be passed down on the mother side of the family, which is what we saw for male homosexuality. So to test that, we did what's called a linkage experiment, which is where you look in families at the co-segregation between the trait and large chunks of chromosomes, and found the very interesting and provocative finding that there was linkage between this trait and a region called XQ28 on the X chromosome, which simply meant that in the 40 or so families that we studied, that most of them where there were two gay brothers, the gay brothers both had the same chunk of the X chromosome. Subsequently, there have been three additional studies for a total of four. Two of them also have found linkage to XQ28, although one study in Canada did not. And if we simply average together all of the data, which I think is the fairest test, it looks like there's a significant, although certainly not overwhelming linkage between this region of the X chromosome and male sexual orientation, which simply tells you that there is probably a gene somewhere in that region that is somehow involved in male sexual orientation. We know that it's not simply a gay gene because the same gene also appears in heterosexuals in its opposite form, but it does seem to be involved somehow in people's choice of either same sex partners or opposite sex partners. I throw that up just to sort of leave the scientific part of the talk on this type of note and to point out that although four or five years ago when we did these experiments, this was as good as it got, it's no longer an adequate description. And when the genome sequence is complete, it will include the entire sequence of the XQ28 region because they're not gonna exclude that just because there's a gay gene there. They're gonna sequence that along with everything else and we'll actually have the opportunity to figure out exactly what that gene is and what it does. Whether it's something involved directly in sexual behavior, something involved in the hypothermic regions that have been invoked in sexual behavior or something that acts more indirectly is not known, but that's the type of information or that's one type of information which will come out of the genome project. We could turn off the slides now, maybe even just turn up the lights a little bit so people can wake up. So this all brings up the question, where is this research going? And I know the next speaker will address this in more detail, but I'd like to just begin you to start thinking about that question by bringing up a few practical applications of this research, which we may or may not see in the near future and get you starting to think about what the implications are. I suspect that the first place where you're gonna see this research used is in diagnosis. If any of you have gone to a mental health counselor at the college or at your high school, you know that they listen to you and they may say, well, I think that you have anxiety or you have clinical depression, but they don't really diagnose you in the sense of explaining why that particular trait is present. I think what's gonna happen in a very short period of time is that when you go to a mental health professional, they'll listen to your story, but they'll also take a little DNA sample or perhaps even look at your DNA chip card if that comes to pass and they'll look and say, well, it sounds like you have depression and I think it's the serotonergic type because you have this variation in your serotonin receptor or it's another type because you have this variation in your dopamine receptor or they may even say, look, all your receptors are fine. Let's talk more about your environment, what's going on at home, what's going on in your job and so on. I think this is gonna be a very good thing. I think that it will bring psychiatry and psychology to the same sort of level that medicine is now. After all, if you go in with a severe cold, the doctor cultures your bacteria to see what antibiotic will work. If you go in with a great malaise, you're tested for HIV virus. You don't just think about it or talk about it, you get a test for it. This will bring psychiatry and psychology at least partly to the same place. There is however a danger even in this relatively benign technology and the danger is, well, if you get the information, what's to prevent your insurance company from getting the information? What's to prevent a potential employer from getting the information? What's to prevent the medical school that you've applied from getting the information and saying, we really don't wanna spend $500,000 educating you because you have a 67% chance of schizophrenia or a 20% chance of alcoholism. To my mind, this would be a real misuse of this type of information. And I think it highlights the need that we have in this country to have stronger rules and regulations and legislation protecting our privacy, all of our medical privacy, our genetic privacy in particular. Of course, this would be largely obliterated if we were like Sweden and had socialized medicine so that people could actually get coverage if they're gonna have a disease instead of not getting coverage if you have a disease, which is the American system, but that of course is kind of a different talk than this one. The other thing I should say about diagnosis is that it's great when a person actually has a problem. When a person is having a mental health problem and comes in for help, it can be very useful. But as a predictive tool, this would be inappropriate. For example, suppose we found a gene that was involved in cigarette smoking. Some good-hearted person might say, well, let's go to the high schools and figure out who has this gene and really focus our educational efforts on them. This would be a very bad idea in part because some of those people who have the gene are never gonna be smokers at any rate and it's really unfair to them. And worse because some of the people that don't have the gene will become smokers. And in fact, if they're not educated, if they think that they're protected, the problem may be even worse. The second area where we're gonna see practical applications is in new drug development. This is a large impetus behind much of the genomics research that's being done. The idea that it would be possible to discover new drugs or tailor old ones to particular individuals. Again, I think that this will be a terrific development. Right now in mental health, we have only very primitive drugs for most of the serious problems like schizophrenia and bipolar disease. They were discovered by accident. They have terrible side effects. I think that when we understand more about the biochemistry of these diseases, that we'll be able to develop a completely new repertoire of drugs. Once again though, there's always a danger. And the danger is the one of using drugs not to treat diseases, but simply to alter people's behavior and what might be seen as a performance enhancing way or even in a performance controlling way. This may seem a bit like fiction, but we actually do this to some extent in our country already. And here I'm talking about attention deficit disorder. A disorder which is a real one, but which is suddenly grown from a rate of about 0.1% 20 years ago to over 20% in some school districts at the present time. And I personally question whether these are real diagnosis or if it's a matter that school teachers don't like boys to be boys, which consists of running around wildly hitting things and not paying attention to the teacher and would rather control them through the use of medications. I think we really have to be careful that we use medications to treat diseases and not to try to alter behavior in what we consider a socially acceptable way. Well, of course, the final frontier of this research is to be say, why bother with diagnosis? Why bother with drugs? Let's just change the genes. Your child has that pesky gene, gay gene? Well, bleach it away. Give them the straight version instead. Your child is going to be depressed or anxious. Why not change the gene and make them into a happy individual? Well, here we are very slippery grounds indeed. I think on two different fronts. First is the practical front. As I've tried to emphasize to you, many of these behavioral genes are pleiotropic. They do a lot of different things. So you might think that you were taking away the bad depression gene from your child and in fact, leaving yourself with an individual who could never have pleasure in sex would then be very depressed because of that and probably bring some sort of legal suit against you for wrongful birth or whatever. But second, there's the whole moral aspect of changing humans. At the very first of these conferences, there was a speech, there were speeches by a number of noted geneticists, Tatum, for example, some of the people who were involved in setting the stage for what we know about genetics today. But one of them was by an individual named Shockley. Shockley has become mostly notorious for his view that racial differences in IQ test scores have a genetic basis. Something which behavioral geneticists have never endorsed because behavioral genetics is the study of individual differences rather than of group differences. But what was truly amazing to me when I read the entire text of his speech is that this was only a small part of it and that the major thrust was a eugenic thrust. The idea that somehow the fact that we now have higher education so that women go to college instead of sitting at home having babies starting immediately after their first puberty, that this is going to result in the genetic degradation of people and eventually our species will go out of existence because so-called natural selection is not allowed to work. We have to be very careful that we don't slip into the line of ever-altering genes for an individual's purpose to this sort of eugenic nonsense. Why do I say that this is nonsense? There are really three issues. First of all, eugenics simply doesn't work because all of these genes are so complex, they have so many different effects because the traits are so very complex. If you simply try to get rid of the bad genes by getting rid of the bad people or not letting them reproduce which is the basic idea of eugenics, it simply doesn't work on a practical level. The second point is that eugenic ideas, like all social engineering ideas, always leave open the question, who is to decide? Who gets to decide what is good and bad for our society? And who gets to decide and make the inevitable choice of who has to suffer for that? And lastly at all, there is the important idea of autonomy, the idea that each of us has a certain value as an individual with all our good parts as well as our bad parts and the idea that each of us does have some control over our own fate and certainly over our own reproductive fate. And I think it would be a very sad idea if that control were taken out of the hands of individuals and given instead to the state. So those are what I see as some of the fairly immediate prospects for this type of research, how it will affect ordinary people. But I'll close by saying that perhaps the most important application of all of this is not what your doctor is going to do or your insurance company is going to do or what the Genetics Institute is going to do. Perhaps the most implication is for how we understand ourselves and how we understand our loved ones and how we understand our children. Some people think that if behavior has genetic influence then somehow we become automatons. That somehow we can be excused for our bad behaviors and yet at the same time not thanks for our good behaviors because they are simply a product of our genes and nothing else. I think that that's an intrinsically wrong view. I think that the fact that we are genetically influenced should let us realize that yes we do have limitations, yes we do have particular strengths and to take advantage of that knowledge to improve ourselves. If you have a strength for example if you're a high novelty seeker use that to good. Use that to explore new worlds and look out for drugs and alcohol because it's easy to become addicted. If you have a propensity for high levels of neuroticism try to use that to become a more sensitive person a more inward-looking person rather than falling into depression. I think that if we understand that this is just one more way of knowing ourselves that we can appreciate and use the information in our own lives. With that I'll close and thank you very much. For a question if it would help to get to the next question. Our shares will be in the aisles to accept your questions as you pass them forward and I'll start this afternoon by asking whether anyone has on the panel has a question or comment for Dr. Hamer. It's a question Dean. We heard this morning and also implied in some of the things you said this afternoon of the inordinate complexity of the genetic control if that's the right word for it of what goes on in the brain. And yet your findings seem to point to at least one gene having a very large effect and an astonishingly broad range of effects. Have you just got lucky or could you please comment on the apparent contradiction between the simplicity of your findings and the possible subtlety of the systems you're studying. Well the serotonin transporter gene accounts for about two or 3% of variation in most behaviors we've looked at. So if you figure most of those behaviors are about 50% heritable you might say oh there might be 10 to 20 genes. And the question is, is the serotonin transporter gonna be as good as it gets as strong as there is? Or is it gonna be sort of a middling example? Or is it gonna be a weak one and there's some much stronger genes out there that we don't know about yet? My guess is that serotonin transporter will be in the upper range of how strong these genes are. And I say that first of all because of course hundreds of labs around the world have spent huge amounts of time and money doing linkage studies of various traits. And the only thing that can be concluded from those is that there are no major strong genes. There's no one gene that is 50% of what's going on for any of the traits that have been looked at at most or a few percent. And second of all I think this is a pretty strong gene because there's only the one of them and because drugs like serotonin reuptake inhibitors have such profound effects on behavior, different aspects of behavior. So my guess is that this will be average to high average and that other genes may have even smaller effects. But I'm glad that you say that it's such a strong gene. To me, two or three percent is not all that bad. Well I mean it's a question with the cups half full or half empty but I mean if you look at the experience of the plant breeders for example before even we got into molecular genetics it was clear that as the member John Jinx once saying, ah he said well number of genes is directly proportional to industry of investigators. Right. Dr. Fox killer. Dean, I would like to ask you to go back to your comments about Shockley or you might have talked about the bell curve or more generally the long history of the misuse of heritability arguments particularly most notoriously in questions about race and IQ. And because it seems to me the way you presented it you presented it as if the problems were political that we shouldn't do this kind of study groups identified by racial characteristics. But in fact the criticism of what was wrong with Shockley or the bell curve or the long tradition that is associated with it is that the scientific abuse of concepts of heritability. And in fact heritability is one of the most notoriously misused concepts in scientific history that there is and the critiques going back to Lancelot Hogman are well known that heritability there's no such there's no such thing as heritability in any absolute sense it depends on the populations that are selected but especially the measures of heritability depend on factoring out genes and environment in ways that for human interesting human behaviors cannot be done. So why in your own account do you still make use of heritability arguments? Heritability is a proportion and it's specific for the environment that it's measured in it's the proportion of variation in a trait that is because of genes. I actually had a slide about heritability and specifically did not show it because it is a relative measure. It is a very useful relative measure as long as you understand its limitations as long as you understand that it only applies to the particular population that you looked at. It's main use is to show that a trait has some genetic influence if a trait has zero heritability geneticists can pretty much forget about it. I agree that it has certainly been misused and in many cases the misuses are not by the geneticists themselves but by others but I think geneticists have done a poor job in explaining the limitations of heritability. The other major misconception that people have about heritability is that if something is heritable at all it's not changeable and of course this is completely incorrect. Many heritable traits are extremely changeable. Many traits that are not heritable, for example the language we speak are quite unchangeable. So I think that heritability does have a useful purpose in studies but it's very important to understand the limitations. Let me also just say about the whole bell curve issue and that. It's so infuriating for people that are involved in this type of science to see any of the results appropriated and misused in the way that the bell curve did. The simple fact of the matter is that all of the measures of IQ that have been done have been done on relatively homogenous groups usually in Sweden or someplace like that is a matter of fact and don't account for the group differences and don't try to measure the group differences. I think that with regard to IQ the most striking finding that I've seen recently was an experiment done by Steele at Stanford where he actually took students, both white students and African American students and administered them a type of IQ test basically but half of the students he told this is an IQ test and half of the students he said this is just to normalize something else. He gave them no idea that it was an IQ test. Well as it turns out the blacks and whites performed equally well under the neutral condition so they didn't know it was an IQ test. When they were simply told that this is an ability test the black scores went down relatively. So it's not just a matter of the environment in terms of how many books you have at home or where you went to school. It's the whole environment of our country. This whole environment set up by books by the bell curve actually has in effect a very subtle one but apparently an important one. The difference in that study was almost 15 IQ points which is the white black difference. Dr. Bachman. Dean you used as your examples single gene examples but could you tell us what about the progress in say where you think that perhaps three or four genes may together influence something and which would then not be apparent in the kind of analysis that you gave us but we know there are multiple genes that are going to affect a particular property or trait and these are more complicated to analyze. Could you tell us where we are in that level of analysis? Right. Many behavior geneticists now are very interested in the idea of multiple genes controlling behavior and are trying to establish both the experimental and the computational methods to be able to quantitate that. For example some people working in my area have seen interactions between the dopamine transporter and two different dopamine receptor genes which is very logical because that forms a natural circuit basically and when you look at that you'll find that for example a change in a particular receptor is effective only when you have a particular version of the dopamine transporter. So it actually makes sense and may define a pathway. The complications at this point are that of course if you look at 100,000 different genes exponentially then you can see a lot of interactions that may not be real because there's so many possibilities and we have to worry about that and the second thing complication I think is that it's hard to distinguish between real biochemical pathways like three different dopamine genes where they're really interacting in the sort of biochemical sense versus what you might call psychological interactions. Like for example people that are highly neurotic, highly anxious about things may not show very high levels of novelty seeking because they don't wanna go bungee jumping because they're afraid of getting hurt even though they really have novelty seeking. So it's a sort of tease out those phenotypic interactions from real biochemical interactions. Can I comment on that? As one of the people who sort of tried to fiddle with models of epistasis, that's interaction between genes and so on. One of the things that you predict from certain kinds of epistasis is that the resemblance between non-identical twins would drop relatively sharply as the number of interacting genes goes up. The fact that we tend not to see a major discrepancy in the ratio of the MC to DZ correlations implies that if there are loads of genes operating and there could well be, their effects broadly speaking are relatively additive, I think. Or they're interacting in a way that is not the kind of interactions that you tend to think of when you think of interactions which is you need this gene but you also need this one and you need this one as well because if it's clusters of genes that are sort of required, you don't get the kinds of twin results, for example. I think you're describing. Could I add a question, Dean, you used the word control with respect to these genes you're talking about a little bit ago and I at least don't feel comfortable with the notion that the kind of genes you're talking about are ones that are controlling the behavior as opposed to influencing a pre-deposing towards one another behavior. I mean, we've got, one of the things I work on is sexual behavior and flies and you'll hear about that. There's a mutant called couch potato which as you'd expect just sits there and it shows no sexual behavior but that's really got nothing to do with sex. It just didn't interest it in moving and so I wonder, you know, in terms of these genes and other genes that might influence these behaviors, whether you really meant control or a more stepped back sort of. Well, of course for the partner of a couch potato it does affect sexual life. Well, I think you bring up two issues. First of all, I shouldn't use the word control because none of these genes truly control things. I mean, control is like a light switch either it's on or off. Obviously these are like dimmers or influence and so on. So I think that's a very good point. The second point is the question of are they acting sort of directly versus sort of indirectly. Couch potato would be an indirect example. Fruitless that you'll talk about would be a direct example. That's very hard to address by genetics and unless you have some real biochemistry and you can get in there, it's very hard to know. I think that in the case of the serotonin transporter gene that so much pharmacology has been done, that animal experiments have done which I didn't mention both in chimpanzees and in mice and knockouts have been done. And the fact that you can stimulate the receptors with agonist or take them out with agonist indicates that it's a fairly direct effect in that case. But as soon as you talk about experiments where you're looking simply at linkage for example or where you don't know even what the biochemical effect of the thing is then obviously we have no way of knowing and it really, what it tells us is we need to integrate the genetics with neurobiology. That's what we are really lacking at this point. Okay, well we've received several questions from the audience. First one is if the attempt to apply heritability and eugenics doesn't work, how can we account for the success in animal husbandry and agronomy? It's your talk Lee. Well, okay let me put it, comment as you were speaking. One of the telling eugenics stories of course were early fruit flyer selection experiments where people selected for things like number of stono, pleural key, that's the kind of number of hairs under the armpit, I don't know if you wanna do that but people did it. And one of the things you found, you had enormous advance on the selection in the direction you wanted, but also it was the fertility dropped like a stone. And the big problem is that when you, there are things about the organization of genetic material along chromosomes which means that selecting for the things you want very often means you want it with a whole bunch of other stuff that you don't want. And so it's actually a high risk occupation. Secondly, I think I want to come back to the, you know it's related to this issue about the bell curve. I don't, I think the day that we think we define anthropology in terms of the things that psychologists measure or the things that are economically significant is the day that we give up being human. Which brings up another question. Oh, excuse me, we, yeah. I think the answer to the question is selective breeding is incredibly successful as long as you're interested in relatively simple traits and perhaps the most spectacular example we have of selective breeding are dogs. So dogs descended from wolves about 10,000 years ago. And today we have dogs like the Irish Wolfhound, seven feet tall, and we have a Chihuahua which can sit in a teacup, and they're genetically identical species. They've been selected for particular traits by virtue of breeders who have, have, have desirabilities about certain type of phenomena. So, but the downside of that of course is exactly what you said. With these specialized traits, you end up getting in many cases in dogs a myriad of genetic diseases and, and predispositions to, to cancer and those kinds of things. But if you're interested in particular kinds of traits, selective breeding can be very, very powerful. There's, there's no question about that. Another question from the audience. Assuming the ability to choose the genes of children as possible, would parents with heritable genetic disease have a legitimate reason for choosing the genes their future child will possess? I think there's no doubt that for very serious life-threatening diseases that, of course, people will want to have the right to choose and they will have the right to choose. And I know that if I were to have a child and that I were worried about myself or my partner having, say, Huntington's disease or cystic fibrosis that I'd want the option to, me personally, to have the option to get rid of that gene. I would not want a state or anybody else telling me that I had to get rid of that gene. I would not want an insurance company telling me that I couldn't get insurance if I didn't get rid of that gene. But I personally would want to have the option to make that decision. I think that when it comes to more complex traits and abilities like sports ability or IQ or the like, that first of all it's not gonna be so easy to do. It's not gonna be one gene. And second of all, that we need to start thinking about that very seriously whether or not we want to have the ability to engineer those traits or not. And I think that that's still an unresolved question. This is also a gene therapy question but it takes a slightly different tack. Do you anticipate that gene therapy will become a corrective tool in the handling and sentencing of recidivist criminals? Will it replace the current practice used on sex offenders? I've actually talked about this quite a lot with judges because I've been involved in a project called genetics in the courtroom where we talk to judges about these precise issues. And let me say first of all that I was worried with talking to judges that they would somehow think that genes explain behavior and that therefore defendants could come in and say I killed them because my genes were bad sort of like the old Twinkie defense. And judges actually are pretty good about that. They said that's complete bullshit. You can't use that defense in my court. We might throw it out immediately. It's gotta be 100% and besides we assume that people have control over their actions. That's the basic assumption of our law and it's gonna continue to be that way. Do I think that it will be used for therapy of criminals? I doubt that pretty much actually. I think that there are so many other potentially effective rehabilitation methods. And I also think that if there were any sort of permanent gene therapy would raise real questions about justice that would probably not be easily absolved. What I do think will happen is that for example the person asked about sex offenders in particular where this will be useful is if for example there were a more effective treatment for sex offenders which they could have the option of using. Cause most sex offenders don't like their own behavior. It concerns them very greatly. And that that would be terrific. You know the treatment that's used for sex offenders now is a sort of chemical castration which first of all has a lot of side effects and second of all isn't very effective because a lot of sexuality is up here not downward testosterone is doing its work. If there were better drugs and actually serotonin reptile inhibitors work for some people then and that were given to them as an option I think that'd be great. While you can't remove genes to see the effect in human research what about temporarily blocking the proteins they code for to see the effect? Is that a direction that we might be able to go in order to distinguish this? In animals, yes. But in humans you have to be kind of careful about going around blocking their genes. This guy would be on my back for sure. No, no, no. Well I think the short answer is that's not the kind of experiment that I could imagine getting past any human subjects committee. I mean we do not have a free range of the things that one can do in the lab. Don't hear that as sounding regretful but I mean in a sense the reason we have human subjects committees is precisely because the scientific and medical communities as well as the lay community are concerned about the kinds of things that we do with people. I mean it's very hard these days even to get DNA from people in studies for all sorts of reasons. The kinds of things that you're doing I suspect are increasingly difficult to get approved aren't they? That's not criticizing what you're doing I'm merely saying nothing that's an empirical fact. All genetics research is being viewed quite carefully now. Let me just tell everybody here that I didn't talk about animal research on behavior at all but there's a lot of work that's done in mice and in drosophila you'll hear about that and even in worms and there are tremendous tricks that can be used to study behavior including there are really cool genetic tricks now where you can turn on or turn off a gene only in one particular tissue and furthermore you can turn it on and turn it off as the organism develops. You just put a little tetracycline into the drinking water and all of a sudden the gene is off and the gene is on. So people that work in experimental animals can really ask some detailed interesting questions about behavior now. Okay, before I ask the last question I'd like to tell everyone that there will be coffee and cider on the Ekman Mall which is behind you out near the chapel and that at 315 we'll have the start of the music before the 330 lecture. So the last question was please comment on Governor Ventura's plan to increase the number of Minnesota children who are above average to 90 plus percent by the year 2001. Another bloody silly idea. I think that's an excellent idea and I leave it to my distinguished colleagues to help them in this effort. Thank you. You must have been in Europe.