 So thank you for that. Hands up if you can still hear me. Excellent, excellent. Okay, so I was shown just before this talk two sheets which appear to have somewhat slightly different content leading to confusion about what I'm actually talking about. So I'm just going to talk about what I'm going to talk about and I hope you're going to enjoy it. And if you don't enjoy it then I hope you learn something. Okay, so what I'm going to cover in reality, the themes I'm going to cover I'm going to give you, I'm guessing very few of you are experts on genetics. Very few of you are experts in psychiatry. So I'm going to give you a very general hopefully light orientation to the topics of genetics and mental illness. And I'm going to illustrate this talk largely by actually by talking about schizophrenia. But in fact the principles and the content of what I'm saying applies across the board and I will make reference to other disorders as I go. So I'm going to give you the orientation. I'm going to tell you what we've actually found in general terms. But more importantly I'm going to discuss what that's telling us that might become useful clinically because I'm a psychiatrist as well as a researcher and I'm not just in this for a laugh. And then because medicine and its area can be a rather dry topic I'm going to lapse briefly into the more esoteric topic of evolutionary psychiatry which preoccupies people with a very different bent in life than I have but people seem to be interested in it. So I'm going to briefly touch on that if time allows. Okay, so this is one of two slides that will give you a brief orientation to schizophrenia. And I apologize the only way I can see my slides if I look that way but I'd rather look at you. My glasses are the wrong strength for actually seeing these slides so I'll do my best. So schizophrenia is what's classically called one of the major psychotic disorders in psychiatry. And typically it starts in an obvious manifestation in people in their late teens or their early adulthood. Now the sort of symptoms that people experience that are the kind of the most dramatic ones in many ways are hallucinations which are sensory perceptions in the absence of there actually being anything to perceive and delusions which is more of a kind of subjective thing but their false beliefs and I'll give you some examples of those with reference to the pictures in the side in a moment. What most people who are not involved in psychiatry or who don't have relatives with schizophrenia don't appreciate is in many ways those are not the most damaging symptoms. They're the scary terrifying alarming the symptoms that you see in movies and so on if you ever watch movies about psychiatry but the really really impairing symptoms are what we call negative symptoms and basically a lot of people of schizophrenia have impairment in their ability to socially interact with other people. They have a low or an unresponsive very flat mood in fact that can be very debilitating and they often neglect personal care neglect eating and they really can become the most disadvantaged people in our entire society. Another feature that's not appreciated even by quite a lot of psychiatrists is that people with schizophrenia often have intellectual or cognitive impairment so they can't think as fast or as clearly as perhaps an average person or their IQ if you believe in IQs but measurements of intelligence are reduced in some but by no means all and often these impairments actually are evident in some shape or form all the way from birth and this is why those of us who research schizophrenia and what I'll refer to sometimes as Allied conditions really conceptualize this disorder in a lot of people as being something that happened early in life probably whilst a person is developing in the uterus and I read this fact or this statement a year or so ago when I was preparing a different talk according to the World Health Organization acute schizophrenia is the most disabling condition of all medical conditions including having all your limbs amputated this is rated as producing a greater degree of personal suffering how they do these ratings that could be challenged so onto these pictures on the side it can be very difficult giving this kind of dry description of what schizophrenia is to really appreciate the degree of suffering involved and this chap called Brian Charney who was an artist was a man with schizophrenia and actually his treatment was quite successful but for various reasons that I'm not aware of he decided to stop his treatment and whilst he was descending into relapse into his psychosis he kept notes of what was happening to him and he did a series of increasingly bizarre paintings of himself and so in the top painting you can see an average looking person perhaps slightly more handsome than average, bearded gentlemen shortly after stopping his medicine his self-portrait becomes a bit bizarre I don't know if you can see them there's a giant mouth in his forehead and that he describes as depicting one of the cardinal symptoms which is having you feeling that your thoughts are being screamed out loud and that other people can hear them and that must be pretty terrifying there might be I was hoping to if you'd see these little waves coming in through the window that's supposed to depict him hearing people shouting from the streets talking and commenting on what he's thinking about and that must be pretty terrifying and it's a generally kind of blue colour because he was very depressed at the time things got more bizarre and so a few weeks later I believe he's painting himself as these two hollowed out eggs and what he's describing in the text that accompanies of it is that because everyone can read his mind he's hollowed out he has no secrets, he has no personal life there's mouths on that painting which is the people commenting on him and smiling and grinning there's crows flying away which he describes as being reminiscent intentionally of a scene from Vincent van Gogh's last painting which is supposed to denote both his increasing suicidal ideation but also his thoughts being dragged by some kind of machinery out of his head and then there's more paintings and ultimately he committed suicide so it's a tragic story but if you're interested in it there's a website at the bottom that gives all the pictures and all the way he's describing how he's feeling now some statistics as kind of alluded to in the last slide when people talk about schizophrenia it's often with a total sense of despair but in fact many people have a good response to treatment but for 20% of people who have it they remain chronically in a condition not identical to what I just described but in some shape or form like that but even the people who do pretty well find it almost impossible to get jobs and in part that's secondary handicaps people hear that they've had schizophrenia they don't like it, they're afraid of it, whatever they have a very high suicide rate but even those that don't kill themselves live on average 10 to 20 years less than someone who doesn't have schizophrenia so it's a physical health problem as well as a mental health problem there has been effectively no new treatments in the last 60 years and the treatments that they exist were discovered entirely by chance and nothing new has really emerged and that's largely because we don't know its origins other than we do know that it's genetic or there's a substantial genetic contribution so this is a slide about my own origins I did, as was mentioned earlier on I did a training in psychiatry in Paisley in a non-academic institution it was frontline psychiatry I saw loads of patients I read a book from Edinburgh that's the thing I read on the right hand side that was supposed to teach me about the origins and the treatments and everything that people knew about schizophrenia and what was evident is that people didn't really know a great deal about what caused schizophrenia so there were some speculations there was the subject of today's presentation genetics was thought to be involved there was a tidy increase in risk of schizophrenia people born in the winter months there was a lot of chit chat about stress causing schizophrenia I think this is possibly a lot of this was coincidental if you remember the previous slide people with schizophrenia often is evident in early adulthood and the evidence for stress was quite often happens when people join the army or go to university or take up their first job so that could just be developmentally coincidental people with schizophrenia more commonly were found in cities and there was arguments raging are cities toxic themselves or do people with liability to schizophrenia drift into the anonymity that city life affords there was some really pretty terrible thing from the perspective of the parents of schizophrenia parents got blamed for the origins of schizophrenia either because they were supposed to communicate in strange ways or they were supposed to be cold all sorts of things that I don't consider as relevant but these were theories at the time and then there was this thing dopamine which is basically a brain chemical so there was some biology, some psychology, some sociology and that was mainstream textbook psychiatry you had stuff that was operating outside the mainstream which I didn't learn much about so on one side we have Sigmund Freud who had some theories about the origins of schizophrenia on the other side we have Glasgow Zoo and RD Lane many of you will probably have read his works I myself haven't and we had all this type of mumbo jumbo really going on about what might be causing schizophrenia and the problem is people with schizophrenia do not live in many respects ordinary lives they are exposed to different environments they are usually poor they are often taking drugs, they are often malnourished you name it, anything that is disadvantageous they are often experiencing it which means you've got the opportunity for schizophrenia itself to be causing some of these things not these things causing schizophrenia but what seemed fairly certain because you're born with your genes is that having schizophrenia didn't cause your genes and that's why I got interested in studying genetics so how do we know that schizophrenia and these other disorders are partly genetic? well basically what we've got plotted here on this axis here is the higher the red goes up the bigger your risk of schizophrenia and along this axis here what we've got is how closely you are related to someone with schizophrenia so the rate in the general population is about 1% then if you have a parent or a sibling or a child, a first degree relative we call it it's about 10% your risk of schizophrenia if you have a twin that's not genetically identical your risk is about that 5-10% if you've got a genetically identical twin or what we call an MZ twin your risk jumps to about 40% so the degree of risk follows your genetic relationship to someone else who's affected and that's a hallmark of genetics and the heritability here so the involvement of genes is about 60-80% there's imprecision in the measurement there is an environmental contribution it's not entirely genetic and if you do the same types of studies to other psychiatric disorders you find they vary in the relative importance of genes so even anxiety has quite a measurable genetic component all the way through to autism at the bottom which is the most genetic of the disorders but schizophrenia and bipolar disorder which is a major mood disorder they're both predominantly genetic and you've got other disorders in between so big deal, these things are heritable and that's what the story was when I was a young trainee you could see that things were heritable but we didn't know how to study DNA so there wasn't really much you could do with that now what we think we're trying to do and it is a process and it's a long process is we're trying... basically we know that genes make proteins if we can find genes that are related to schizophrenia we know the proteins if we know the proteins we can try and work out what's happening in the cellular components of our body if we can work out what's going on in the cellular components of our body the hope is we can work out what's happening in the body as a whole organism and by understanding that we should be able to then instead of just guess new treatments design new treatments based upon a deep understanding of what the cause of schizophrenia and these other conditions is so that's the kind of enthusiasm this was the enthusiastic model I set out with in something like 1990 or so and spent a decade maybe more finding nothing and the reason for that will perhaps become clear but the old model of genetics that perhaps if any of you studied biology a long time ago genetics was essentially largely a deterministic concept so the basic deal is you'd find a big family like that people shaded in white are people with a disease this is classic genetics there's one bad gene one faulty gene in that family anyone who inherits the faulty gene gets the disease no argument anyone who doesn't get that gene doesn't get the disease no argument this is a purely deterministic model of disease and this applies to things like Huntington's disease for example however after searching for 10 years or more myself and many more colleagues around the globe we never found anything psychiatric disorders and in fact pretty much all common disorders be it cardiovascular, diabetes other common things that people suffer from don't adhere to that model rather they adhere to this rather complicated slide which I'll go through slowly disease risk conferred by genetics is probabilistic and there are many causes so on this slide what we've got is not many risk factors for the disease along the bottom increasing number of risk factors until you've got a lot of risk factors for the disease if you follow the dotted line that's how often in the population people have not many people have a small number of risk factors not many people have a high number of risk factors most people by definition have quite a lot of risk factors if you then follow the solid line that's the probability that you'll actually become ill so you can have a lot of the dotted line you can have a lot of risk factors but you don't become ill until you get quite a lot of risk factors and then your body can no longer compensate the body is actually pretty well designed it might not look it in my case but the body is well designed it can tolerate a lot of problems and still keep functioning and still keep functioning well it's like straw once the straw in the back of the camel scenario basically you can keep chucking straws in the camel it'll be fine but ultimately the whole system will collapse no one straw is causing the camel to collapse no one risk factor is causing the disease and to find this type of thing the tools simply didn't exist until maybe about 10 years ago or so so there was emergence of this technology and I won't give you the details because I'm sure that you're absolutely not interested in it but it's simply a sort of jazzy name called genome-wide association studies now in order for you to understand DNA a cartoon I think the DNA double helix is very well known amongst everyone no matter what their background I have say two copies of a gene one I got from my mother one I got from my father essentially they're identical but every so often these letters denote the genetic code there's a difference between those two copies and that's called a polymorphism it just means it's a variation that occurs and we all have millions of these things and some of them lead to disease but because there's millions of them it's pretty difficult to find out which ones lead to disease we now have the technology in you sir, you sir, you madam, me to measure millions of these things from your DNA and to measure it in huge populations and this allows us pretty simply to find out which of these things are more common in people with a disorder like schizophrenia than in people who don't have schizophrenia and that's called a genome-wide association study for those of you who don't know the technology the genome is simply your total body's content of DNA when those tools became available pretty rapidly I led this study which was the first time a common genetic finding was implicated in any psychiatric disorder now the importance isn't the finding the importance was we had to study 20,000 people to make that one finding we had to get DNA of 20,000 people and clearly, actually in the first 15 years of my work my colleagues collected 1,500 people so to make this discovery required a serious culture change and this is reverberating throughout genetics and medicine more widely people need to get together and so what's happened since that first study back in 2008 is that loose collaborations formed and this looks like some kind of evil empire you might see in a backdrop of a James Bond film but basically we've formalized structures in 2014 there was about 35 countries involved with lots of research groups coming together in a formal way sharing resources by 2019 I think it's about 50 countries I can't have lost track about 1,000 scientists and as we've pulled the resources we begin to find more and more and more now the details of this next slide are not important there's one box that's very important but basically what's here is I've got all the chromosomes lined up from number chromosome 1 to 22 there's a red line if any dots are above that red line it means we've found a genetic variation that's associated with schizophrenia and these days in the most under paper that's currently been written up we've studied 150,000 participants we've found over 300 genetic things that are related to schizophrenia the key thing about the nature of what we're finding is that of these 330 things we've found every single one is common I'll come back to how common they are in a moment each increases your risk it's the straw and the camel's back each one is increasing your risk by less than one tenth of one percent it's a tiny increase in risk one of these is doing almost nothing and there's much more to be found the other disorders depression, bipolar disorder attention deficit disorder autism they're going through similar processes they're finding similar things but they're less advanced than schizophrenia which is why my focus is on schizophrenia because I can illustrate more but the other disorders are going through similar processes now so far I've talked to you about these common genetic variants that have minimal effects but occasionally there are these things we call rare copy number variants it doesn't really matter what we call them there's a picture of a chromosome there sometimes people get a hole in their chromosome a big chunk gets knocked out of it so they've got a chunk of their chromosome missing they quite often lead to medical conditions in this case this person has got a... the name of the syndrome doesn't really matter what matters is that instead of a minuscule increase in risk of schizophrenia the person with this hole in a chromosome has a 25 fold increase in risk of schizophrenia so this is not like putting the straw in the back of a camel this is like putting a giant tree on the back of a camel camels are pretty strong so they can cope with the trees so they still need a number of straws but they just can't cope with as many straws and since those discoveries the same technology that I refer to allows us to look for holes in chromosomes systematically and a bunch of these have been found now there's about 15 of them there's a few key properties that I'm going to look to now and important properties I'm going to look to later first of all they're rare each of them occurs in less than one in a thousand people they have these big effects on disease but they're still not deterministic you can have these there'll be people in this room with at least one of them the combined total frequency in the general population is about, you know, one, two percent so there'll be people in this room with them you can have these things and live a perfectly high achieving life not develop a psychiatric disorder so they're not deterministic but they do increase your risk a lot so this is kind of what we've found we've found lots of, this is how common something is from common to rare this is how big the effect on disease risk is up there and we've found a whole lot of things that are common and have tiny effects on risk we've found a whole bunch of things that are rare and have relatively large effects on risk the shape of this curve is important and I'll come back to it when we're talking about evolution later on however, so far this is just a stamp collecting exercise we're finding things if this is not going to do anything useful we may as well not have bothered so, my main motivation for getting into this was to try and understand biology so what have we been able to find out? well, we've got a lot of these things so we can begin to look right so we've got 300, say we get 330, 340 genetic findings related to schizophrenia there are ways where we can say ok which bodily tissues are those genes active in and there are so many hypotheses about the origins of schizophrenia that that's a very legitimate question because there are even people who think that schizophrenia has no biological basis so what you can do is you can say well are these genes active in fat tissue the immune system, the cardiovascular system the gut, the liver, etc, etc these colours here correspond to the colours here and if you see a dot above that line there then that means the genes are active in that tissue they're more active in that tissue than any other tissue and the tissue in schizophrenia that the genes are active in is the central nervous system it's that kind of, I don't know what you call that colour, orange-ish there's pretty much no evidence for genetic activity in any of the other tissues so the main risk organ is the brain now if I'm talking to much younger psychiatrist in training I regard this as the, if you'll excuse the language no shitsherlock slide on the other hand speaking to my contemporaries when I was training there are plenty of people who are not willing to assume that the main risk organ existed never mind that it was the brain we can go a little bit further than that so the brain has multiple cell types the neurons, most people have had neurons they're the kind of cells that are the process the information in the central nervous system they're the kind of the main where the main action is traditionally thought to be but there's lots of other cell types that have got important roles and we need to know what are the cells in the brain that are most likely to be aberrant in schizophrenia so by the same principle we can see which of these cells the genes are active in and again it's colour coded and above that blue line is only the big green one and that's the cell corresponding to the neuron we know the main activity is in the central nervous system we know it's in neurons now neurons are quite big structures and there's this there's this tiny bit that I've cartooned in here called the synapse now for those of you with no biological background the synapse is the bit where two nerve cells speak to each other one kind of releases chemicals that is picked up by another one and that's how they communicate with each other and so that's the point of contact between two different neurons which I've got a bad cartoon over there so when we look at where in the neurons the genes are concentrated in it's in this particular structure of the synapse now the details I think are probably not that exciting to this audience and in fact but we know that the signal is concentrated in this type of neuron and we know that it's concentrated in the cell that's receiving the signal rather than generating the signal now that's pretty much as far as our biology has gone but that's gone a long way from this kind of vague stuff that I was being trained about it's still not deep enough to really understand exactly what's going wrong with these proteins exactly how it could then be targeted if at all for treatment so there's still a lot of work to be gone about I think we've come along quite a lot in a relatively short period of time and some of these molecules suggest immediately new ways of targeting them with treatments I believe that's optimistic but there are some people pursuing some of these molecules now trying to see if there's new ways of treating it but I think we need a deeper biological understanding than that I'm going to move slightly on to the... because one of the things that's been very surprising to psychiatrists is the structure of psychiatric diagnosis and what genetics is telling us about that so for those of you who are unaware you know, when you go and see a GP or a physician or someone else and you've got a bad back's a bad example but maybe a pain in the chest or something like that the GP will come up with a hypothesis as to why you've got the pain we'll usually organise some tests which may or may not confirm the hypothesis you go and see a psychiatrist you've got a bunch of symptoms the psychiatrist says those symptoms fit schizophrenia those symptoms fit obsessive compulsive disorder those symptoms fit depression that's where it ends there's no biologically... biological way of testing whether or not the doctor's hypothesis is correct so diagnosis in psychiatry is a hypothesis it can be reliable but that doesn't mean to say it's biologically valid and given that a lot of the treatments we offer are biological we need to begin to make biologically valid diagnoses not just on the basis of symptoms but on something else so does genetics have anything to contribute towards this story? in order for me to explain how it's contributing to that story I have to talk a bit more about what I described earlier on is these highly polygenic disorders and this is where the high frequency of these risk factors and their small effects becomes critical so bear with me it looks terribly complicated it's in fact very simple each of those squares with colours in it represents a different person now imagine there are only a hundred risk genes for schizophrenia we know I've already told you there's more than that but let's just imagine there's only a hundred each of these tiny boxes represents a possible risk gene it's a 10 by 10 grid so there's a hundred risk genes so you picture that let's imagine each has a frequency of the population so 10% of the population carries one copy or more if you've got one copy of the risk gene you've got a blue square if you've got two copies of the risk gene you've got a red square if you've got no copies of the risk gene you've got a white square even with only a hundred risk variants there's enough ways of filling in these squares for every single person in the world to be different so no two patients is identical in terms of their risk genetics what's more you can do the maths even with only a hundred risk genes that means the average person in the population this isn't the average person who has schizophrenia this is the average person in the population has 20 of them and there is no one statistically who has none now you can multiply that up by at least a factor of 10 there's at least a thousand risk genes not a hundred so the average person probably has about 200 of these things now what we can now do I mentioned that we could measure your DNA using the same technology I could grab your DNA and count how many you've got and how many you've got and I can see who is relatively more or less at risk where are you on this risk curve and when you do that it looks like this so there's about 10,000 people representing this graph and this is if you count up the number of risk alleles these people don't have very many these guys have got quite a lot and the more they've got the bigger the risk of illness I don't want anyone to go away thinking that we're now doing genetic diagnosis of psychiatric disorders because we're not without giving you the figures this is woefully this is no good for diagnosis you'd get vastly more people wrong than you would write if you ever tried to apply this methodology what you can do is say okay we can measure people's risk for schizophrenia and where they are on that spectrum are there any other disorders that look a bit like schizophrenia so are there other disorders that share a lot of these risk genes with schizophrenia when you do that and this I have to say none of my psychiatric colleagues would leave this when we produced something that looked like this about 10 years ago now on this graph you've got the higher the bar is the more similar the disorder is to schizophrenia in terms of genetic risk along this axis all we've got is different psychiatric labels every single psychiatric disorder that we've mentioned that we've measured is more similar to schizophrenia shares genetic similarity to a greater or lesser extent with schizophrenia the most similar disorder is bipolar disorder which is the kind of major mood disorder with mania and depression the depression share is about 50% correlated with schizophrenia even anorexia obsessive compulsive disorder ADHD you name it they share risk with schizophrenia if you take those CNVs they're the holes in the chromosomes I was talking about earlier on the ones with the big risks on schizophrenia every single one that we've found so far that increases risk of schizophrenia also increases risk of intellectual disability most of them increase risk of autism as well so the bottom line is that when you look at it from the biological genetic perspective there is no neat separation between the psychiatric diagnosis that doesn't mean to say that there is no validity at all in psychiatric diagnosis people with schizophrenia are more similar to each other than people with schizophrenia are to someone with autism so there's degrees of relationship so there is some validity but it's not a clear cut biological validity so this leads us to wonder can we instead of relying entirely on a diagnostic label can we begin to cluster people by the degree of genetic similarity that's illustrated in the next slide so imagine you've got a disorder let's call it bipolar disorder manic depression and old money let's pretend there are only kind of two forms of that with different biologies so although they've got same symptoms same outward appearance internally there's only two biologies biology blue, biology red biology blue really needs treatment A biology B really needs treatment B but we've got no means of telling the difference between A and B based upon the clinical picture so what we decide to try and do is if we take people with bipolar disorder can we divide them into people who've got high schizophrenia liability because bipolar disorder shares risk with schizophrenia and low schizophrenia liability and as it happens to a beginning approximation we can so if we take the population of people with bipolar there are basically two classes of bipolar which are more severe and less severe and they're called bipolar one and bipolar two exactly what underlies them is important it's only the bipolar one that shares similarity with schizophrenia, bipolar two doesn't so you've already got a biological basis for dividing those two types of bipolar disorder we can take bipolar disorder further and divide that into people with bipolar disorder who have psychotic symptoms that's hallucinations and delusions very prominently, very frequently they're quite similar to people with schizophrenia whereas those that don't are much less similar to people with schizophrenia so again there's a point there of beginning to divide and carve up what we call stratify bipolar disorder so this is the start of being able to divide up disorders the aim is that we can then target treatments more effectively we can understand the subgroups better I'll give you one ADHD slide here I was going to give two slides but I'll skip one of them so we also, I've said to you that we cannot diagnose we cannot say who's going to have schizophrenia or ADHD or depression based upon the genetics but looking to see whether or not genetics distinguishes amongst people with a disorder is a different question so challenge one, ADHD is its onset usually in childhood for a reasonable number of people it begins to peter out in terms of the main symptoms by the time they reach adolescence and early adulthood but for others it remains persistent and for those people, ADHD some people view as almost a trivial thing people with ADHD get vastly inflated rates of being incarcerated drug addiction, loss of life expectancy, blah blah blah ADHD is a serious disorder and particularly in those where it doesn't peter out so this is where I'm going to skip so we looked at population of people and this line basically, all these various lines here this is age on this axis and months and this is the degree of symptomatology on this axis and we're really interested in the top two lines you get people who are the age of 47 months that's about 4 years old you get two groups with high levels of symptoms but one of the groups begins to decrease by 100 months or 86 months the other group remains chronic up to 200 months and we haven't measured beyond that the group with the high burden of ADHD genetic risk alleles are the ones that remain high so we can to a statistical level at the moment not a clinical level but to a statistical level we can start to distinguish those people who might need more interventions very early because the other ones to a greater or less extent will take care of themselves now evolutionary psychology why people are preoccupied with this I don't know but nevertheless there is a huge discipline of evolutionary psychology in my view it's entirely speculative but we've now got some hard data to talk about it so there's a couple of... this applies to most psychiatric disorders again I'll illustrate with schizophrenia there's no clinical aim here as far as I can tell you what I'm going to tell you will never be of any clinical value whatsoever but it is a curiosity the people are interested in that it is a curiosity so people with schizophrenia have on average one third of the children that people without schizophrenia have got now in real... in Darwinian terms if you like this is natural selection if you don't have children then your genes don't get passed on and that disorder those genes should vanish from the population this is what's called negative selection and actually only having a third the number of children is actually a very intensive form of selection it really should if all things being equal vanish from the population there are very few if any disorders where the number of children is reduced by quite so much that where the disorder remains as common as schizophrenia is so in order to try and explain this there's been a whole load of creative thinking going on so some people... what people like to do I think because it's a nice intellectual exercise they can sit around having a glass of wine and talk crap with their mates and speculate back in Greek times or something like that having schizophrenia and genetics would be advantageous I've given you a few examples here one of the most popular ones is based upon the I forgive the language mad genius idea that the people who don't fully manifest schizophrenia but the people who are related to those with schizophrenia they have got more creativity and this is a benefit to themselves or society and that explains its persistence we've even seen someone say that being an artist I don't know if we've got artists but I guess we'll probably have some creatives in here being creative or being an artist is the male... the human equivalent of a peacock's tail this is a highly sexist type of remark but the basic idea is that male artists will be very desired amongst females and so they'll mate with them having a big peacock's tail I'm not saying I buy it I'm just telling you that's the story someone else proposed the Odyssean personality this is such a major paper that I was unable to access it on the internet but nevertheless in a manuscript that I wrote last year I was specifically told I had to cite the Odyssean personality I think the general idea is that people who have high schizophrenia liability might be cunning like Odysseus maybe a bit more involved in out-of-the-box thinking magical thinking more cold-hearted, more detached willing to take a risk that kind of thing but without reading the paper I can't be sure but anyway they speculated that that would be advantageous in a world plagued by terror, strife and war and then I don't know if anyone's read this book by Sebastian Faulks he deals with a theory proposed by this gentleman Tim Crow that the basic evolution of schizophrenia parallels that of language and that really in order for the brain to formulate language that had to become structured in such a way that you were liable to psychosis I've never bought that, but there we are so they're the exciting ones that people can chat about and argue about till the cows come home and as far as I can see there's very little way of showing that it's true there's a really boring theory that actually the thing, these genetic effects can remain common just by random chance although having schizophrenia itself has a strong negative pressure on you most people who carry these risk alleles, i.e. all of us are not subject to this and that just allows things to float around kind of by chance and things by chance can drift to become common so that's the boring theory now looking at the evidence so the anecdotal theory, this is a chat back in 1931 it cannot simply be chance that among geniuses the healthy constitute only a small minority now this kind of works where people would read biographies of famous composers or painters and so on and conclude that they were mentally unwell but of course that sort of thing is open to huge bias because people who are very creative and who happen to do very dramatic or outlandish things tend to get biographies written about them so the people who are very creative but just kind of sit nicely about it and don't trumpet it will have less written about them these guys however, do we have any accountants or auditors in the audience? so these chaps here somewhere in Sweden decided they would define people as being creatives if they were PhD level scientists or they were occupied in the visual and non-visual arts and what they found out in a pretty systematic study so you could argue about the definition of creative nevertheless amongst this group of people there was pretty convincing evidence in fact that these people were more often to be found in the relatives of people with major mental illness as a contrastor they chose accountants and auditors who they were classified as less creative now don't shoot me, I didn't come up with this and I think if you spoke to Al Capone he would find out the benefits of creative accountancy quite early nevertheless these people were underrepresented in the families of people with schizophrenia and they had all sorts of elaborate things so there's a story case to be answered we're back to our nice genetics here so this is the you measure genetic liability to schizophrenia you get this curve this was done in Iceland they defined creative individuals as belonging to the national artistic societies of actors, dancers, etc and if that green line is the midpoint the average genetic liability to schizophrenia people who were creative were more likely to be on the high side, score wise so that is consistent with this idea that one of the manifestations of genetic liability to schizophrenia could be increased creativity again they chose as the boring group executives and executives were smack in the middle they didn't have a high or low schizophrenia liability they were just normal folk however selection depends upon all the effects of the risk alleles in the population so for any children who see schizophrenia liability increases risk of psychiatric disorders maybe creativity it also is associated with poor recognition unstable mood various personality measures like neuroticism open this low well being and actually selection is on the these are only some of them selection depends on the net effect here and natural selection really boils down to how many children do you have that survived to go on themselves to have children it depends on reproduction it doesn't rely upon us making a value judgment that it would be great to be creative or it would be great to be this or great to be that it's about numbers of children so we looked at this in a sample of about 150,000 people from the UK we measured genetic liability and we know the number of children and what you've got here on this axis is less this is being slightly more than average schizophrenia liability to more and more and more and more until you get high schizophrenia liability up here is a measure of your number of children that you've got if you're below that dotted line you've got less than the population average if you're above the dotted line you've got more than the population average and basically we found evidence for a difference by sex so for males increased schizophrenia burden wasn't really increasing the number of children in fact if anything once you got to the extreme end it was reducing the number of children for females on the other hand there was an increased number of children by increased burden of schizophrenia liability now that's quite contrary if you remember the peacock story I told you it's the males that have got the big peacock things not the females but if this means anything it would be the women that are wearing the peacock tails but in fact although this looks dramatic in the graph the number of excess children is tiny and it cannot offset the real serious reduction in the number of children that people with schizophrenia got so I asked you to try and remember this slide and I think given the way time is going this will be my last slide we've got all these things that are common and small effect we've got these things that are rare and big effect the way it all works anything above this imaginary curve is very easy for us to find anything below that curve is difficult to find but we need to concern ourselves with that basically if there was anything that was common that had a big effect we would have found it we could be pretty sure we would have found it with what we've done therefore there is nothing that's common that has a big effect this these things are driven to be rare by natural selection these things are kept to have small effects by natural selection so rather than being this positive beneficial effect of all these other weird phenotypes that have been affiliated or speculated to be associated with schizophrenia it looks more like the boring thing natural selection is keeping it down there are no major benefits but on the other hand if you've got a reasonable number of these there's no major hazard either some of you might become creative some of you might become depressed but natural selection is keeping it low so it looks like the true answer is the really boring answer and that is it's just chance drift I'm going to skip this out even though being born in winter was quite interesting and come to the conclusion so you know I hope I haven't been either too aggressive or too defensive but over the years I've had to face a lot of skepticism about there being any purpose to doing psychiatric genetics so I hope at least two of you have been persuaded that despite the shortcomings in our ways of classifying psychiatric disorders and so on mental illness can be approached by genetics and that there's a purpose to it that we haven't got all the answers by nowhere near got all the answers trying to find things that look like they'll become useful from biology, from stratification and even things that are interesting but not useful like evolution also genomics can be applied increasingly by researchers with no primary interest at all in the genetics of A, B, C, D which is the vast majority of people I was very cheered a few years ago when I read a paper the authors of which were the social sciences genetic association consortium I personally would not associate genetics with the social sciences and I'm delighted to see that social sciences have taken it up and with that I'll finish up