Personality and mental health: Genetic or environmental?

Nick Martin:
Great pleasure Clare.

Clare Blake:
How much of our personality is from our parenting environment and how much is genetic?

Nick Martin:
Well, first of all, we can only speak in population averages. We can never say for a particular individual how much of their personality is genetic and how much environmental. But at a population level, we can say it’s about 40 or 50 per cent of the differences between people are due to genetic factors.

Clare Blake:
That’s a high number.

Nick Martin:
Well, it’s intermediate. I mean, for IQ it’s about 80 per cent. For social attitudes it’s a bit lower, sort of around 30–40 per cent. For a psychiatric symptoms of depression, for example, it’s about 30 to 40 per cent. So personality is in the mid range of genetic influences on behavior.

Clare Blake:
So we can have a go at our parents more about how smart we are other than how nice we are? Let them off the hook in a few things.

Nick Martin: 
Perhaps, yeah.

Clare Blake: 
So what elements of our personalities are influenced by our genes?

Nick Martin:
Well, that depends upon how you describe personality. And there’s a whole branch of psychology that has attempted to provide a taxonomy of personality over the years and there’re lots of different ways of classifying it. But the two dominant dimensions – and it’s best to think of it in terms of dimensions – are extroversion versus introversion, and neuroticism versus stable. And that actually describes a lot of the variation in personality, but there are other dimensions as well. There’s a dimension called tough-minded to tender-minded, another called socially desirable versus basically, ‘don’t care’. Another one called harm avoidance, which is actually closely related to neuroticism. Reward dependence – I think some of us probably recognise that behavior. And novelty seeking, which is related to impulsiveness. So these are all different dimensionalities that have been proposed and studied in relation to genetics. And they all come out with that figure somewhere between 40 and 50 per cent.

Clare Blake:
It must be incredibly difficult to measure.

Nick Martin:
Well, behavior is difficult to measure, yes. But that’s the whole branch of psychology. That’s what so appealed to me about it. I mean, it’s easy for a physicist to say, ‘Oh, this is all rubbish because you can’t use a thermometer or an oscilloscope or something’. I mean, quite clearly there is such a thing as personality and differences between it, because everybody talks about it. So the challenge for psychologists is to try and measure that. And we do that using questionnaires – people are highly familiar with those. What they may not know is how much work goes into the development of those and the criteria by which one would regard them as useful or not – how reliable is this measurement from day to day within the same person, from year to year? And the constructs I’ve spoken about – neuroticism and extroversion – are incredibly stable over the lifetime. Once people have reached about 25, their personality seems to be pretty much set in concrete. There’s actually also quite a lot of consistency before 25, as people are growing up, you can see these traits developing and any parent can see that in their own children. The other important aspect of measuring personality is whether they predict anything interesting or useful. That’s called the validity; a good test of the extroversion-introversion questionnaires. You find that used car salesmen tend to be more extroverted and librarians tend to be more introverted. So that’s a reasonable reality check. And there’s huge amounts of work that have been done like that, that give one some confidence that these scales are useful.

Clare Blake: 
You’ve studied twins for literally decades, and that’s really important and very helpful in assessing the nature versus nurture.

Nick Martin: 
Absolutely. That’s really where the subject has come from. It’s the comparison of the similarity of identical and non-identical twins. The method is really very simple, conceptually. Identical twins share all their genes in common. So any differences between them must be due to environmental factors. Non-identical twins are just like, actually, ordinary brothers and sisters – they share about half their genes in common. So to the extent that they are less similar than identical twins, you can attribute that to genetic factors and [with] a little bit of simple algebra, you can actually calculate the relative proportion of genes and environment they’re contributing, which is where the numbers I gave you earlier come from.

Clare Blake: 
When you get those results, what are you hoping to do with them?

Nick Martin: 
Well, knowing from the twin studies that there’s a substantial heritability – in [the case of depression] about 37 per cent – we can then ask, ‘Well, what are those genes?’ Fortunately, beginning in 2005, we have this technique called genome wide association scan, which depends upon typing very large numbers of cases with the condition, disease – depression in this case – and controls with gene chips, which are little chips about the size of a postage stamp on which you can type up to a million gene variants, called single nucleotide polymorphisms, or SNPs for short. And then you do that for tens of thousands of cases and tens of thousands of controls. And you do simple statistical tests to say, ‘Well, is this variant more common in cases than controls?’ As I said, that’s called a genome wide association scan, or GWAS for short.

Clare Blake: 
GWAS if you’re googling it.

Nick Martin: 
That’s right. And that’s just what revolutionised this field in the last 10 years or so. I mean, we know more about the genetics of complex traits. I’m not talking just about depression, but all the other common complex traits: Heart disease, psychiatric diseases, neurologic diseases, autoimmune diseases, you name it, arthritic diseases, and so on. All of these have made huge strides forward as a result of this technology. So the aim in every case is to expand our knowledge of the biochemical basis underlying these diseases with the hope that this will lead to the development of new drugs and there are already plenty of examples of that happening right now. But the other aspect is that [with] these genome wide association scan results, you can create what’s called a polygenic risk score. And you just simply add up all of the SNPs that are influencing a trait, and you can add up the extent to which they’re doing that and get for an individual what’s called a total polygenic risk score for that disease. And you can look at the distribution of this in the population, and it will have a nice normal distribution. And you can show that those people at the top end of the distribution, say the top five or 10 per cent, are at greatly increased risk of the disease than those at the bottom of that distribution.

Clare Blake:
At a clinical level, that would be extraordinary information.

Nick Martin: 
Well, exactly. And so you asked me, ‘Well, what’s the use of this and how’s it going to affect doctors?’ In fact, I think this is probably going to have – at least in the shorter term – an even bigger impact than the effects on pharmaceutical development, which I think will come, but that always takes longer. It’s just a very long process, whereas this, quite frankly it’s available now. And in fact, it’s already happening. I’ve just, along with many other people, just had my genome genotyped at 23andMe for a couple hundred dollars and you can get your individual results from that. You can download them. You can upload them to another website called impute.me, which is free. Within 24 hours or so, it’s fed you back your polygenic risk score on about 500 different traits and diseases…

Clare Blake and Nick Martin:
If you want to know.

Nick Martin:
 If you want to know. And you need to be prepared for that.

Clare Blake: 
Let’s talk about depression and genetics. One in seven of us will have major depression in our lifetime. So how much of that risk is environment and how much is genetics? It the same as everything else?

Nick Martin: 
It’s about 37 per cent. There’ve been huge numbers of studies done. [There was] a recent meta-analysis that actually meta-analysed all these studies on hundreds of thousands of people. And the figure comes out that depression is about 37 per cent. That’s substantial, but it’s actually at the bottom end of the range for psychiatric conditions. For schizophrenia, for example, it’s about 80 per cent. For ADHD, it’s about 80 per cent. Bipolar, it’s about 60 per cent. So depression is slightly less genetically influenced than those other conditions I mentioned, but still a third of the variation between people is a lot. That really is the starting point of my research, which is to actually go further from the twin studies and say, ‘Well, what are those genes?’ Because if we can find those genes, maybe that will give us new clues on how to treat depression.

Clare Blake:
Because at the moment, finding out that you’re genetically predisposed, does that then become a risk factor?

Nick Martin:
That’s a very good question. And I don’t know of any studies that have been done directly addressing that. In fact, our ability to do that study has only been available for about the last year, because it’s only in the last year that we’ve actually been able to find over 100 of the genes for depression and we’re, for the first time, able to start calculating a person’s risk. I should say right away that it’s still a very, very inexact tool. It accounts for only about two or three per cent of the variation in the population. We’re going to need very much bigger studies than we’ve done so far to get that figure up to the point where it’s a more useful instrument. But in principle, we can do it.

Clare Blake: 
On studies, you’re currently running the Australian arm of the world’s largest genetic study of depression. And I guess you have a lot of hopes for this study.

Nick Martin: 
Yes. The Australian Genetics of Depression Study, we’ve recruited over 20,000 cases of depression from around Australia, both by news broadcasts like this, but also by directly going through the department of human services and identifying people who have been prescribed antidepressants at least four times and sending them letters asking them to take part in our study. I should emphasise that all of that is done with the strictest privacy guidelines. We don’t get their names and addresses, we rely on them to contact us. And from all those people who volunteer, those 20,000, we asked them to give us a saliva sample, to spit in a tube and mail that back to us. And we extract DNA from that and then get it genotyped. And we’ve had that done for over 15,000 people so far. That is contributing to the worldwide total of people taking part in this depression study, which is now over 200,000. Believe it or not, we’re actually going to need about 10 million cases to find about 80 per cent of the genetic variation in depression. But I believe we’ll actually get there within the next 10 years or so, we’ll get to that point. And then we really will have a very powerful instrument that will be able to tell people their risk of depression, if they would like to know that.

Clare Blake:
Talking about those participants, they are particularly responsive in this area of study, aren’t they?

Nick Martin:
They are, they are. And I’m not sure how to put this. Unfortunately they tend to… be concentrated in the more highly educated members of the public. We’re not really getting a great representation across the whole social spectrum of… The people who take part in our studies are much more likely to volunteer and be attuned. So we’re not really getting a full cross section of society.

Clare Blake:
That’s unfortunate, but they are incredibly motivated.

Nick Martin: 
Well, they’re very motivated. And so that’s why I shouldn’t say, ‘unfortunately’, because we’re extremely grateful that they have volunteered and we want a lot more of them to volunteer please, because we need huge numbers. I don’t actually think that that bias in the social characteristics of our sample is actually critical to the quest for finding genes for depression.

Clare Blake:
If you’d like to be a part of that, I’ll give you our website for details later on. Not only do diseases possibly run in families, but groups of diseases.

Nick Martin:
That’s right. That’s what we call genetic correlation, where you find the same genes causing condition A, also cause condition B. And a very good example of that is anorexia nervosa, which turns out has a very high genetic correlation of about 0.5 with obsessive compulsive disorder. So we’re studying both of these things at the moment. We recruited over 3000 anorexia patients, which contributed to a worldwide total of about 17,000, from which we found the first eight genes for anorexia. Within both that study and the depression study, we’ve also got items for obsessive compulsive disorder, so we’re actually directly studying that correlation.

Clare Blake: 
So you could have anorexia and OCD, or you could have anorexia and a sibling could have the OCD?

Nick Martin: 
That’s right. Well, in both cases, these things run in families. I mean, a very interesting question is… because anorexia is overwhelmingly a female condition. I mean, not totally. About 10 per cent of cases are male, but it’s over 90 per cent are female. So the question is, ‘What are the brothers of anorexic females like?’ And the answer is they’ve got a pretty high rate of OCD.

Clare Blake:
Because so many females had anorexia, do you think that’s part of the reason why in the past that they thought it was an environmental thing?

Nick Martin: 
Well, I think the default, until recently, was to assume – and it’s a hangover of the Freudian psychology from the 30s – to imagine that it was all environmental. And this is not a harmless assumption because it causes many individual patients, and even more so their families, to torture themselves with guilt: ‘Where did we go wrong? What have we done wrong? What could we have done differently?’ You asked at the beginning, ‘Why are we doing this and why does it matter?’ Just merely understanding that genetic factors are so important in these diseases in many ways is very liberating in alleviating that guilt. And I’ve had so many people ring me up or just make the comment to me, ‘Oh, it’s great to know this because we wondered what we’d done wrong, or why our daughter has got this condition.’ And knowing this really puts a completely different light on it, removes the blame.

Clare Blake: 
It’s extraordinary and it is a family disease because everybody suffers.

Nick Martin:
Well, that’s right. Of course that’s the other aspect of mental disease, that it’s not just the patient who suffers, it’s the other family members who have to live with that. At least if we can spread the message that they shouldn’t feel guilty at an environmental level. They might be guilty in the sense that they transmitted those genes, but there’s nothing you can do about it.

Clare Blake:
Well, I think family Christmases would be a bit different anyway, a bit less blame throwing.

Nick Martin:
Well exactly. And I think that’s important. So for that alone, I think this work is worth doing, but our intent is more than that. The hope is that through this phase of actually trying to find these genes and the neurochemical pathways, that we’ll be able to design better drugs than are available at the moment. The other aspect of our depression study is that we’re looking not only for the genes that cause depression, but for the genes that influence how people respond to particular medications. So it’s a commonplace that antidepressants only work for some people. And we’ve quantified that quite dramatically in our own study. I mean, you prescribe to a very large sample. We’ve got 15,000 patients [participating in our study] and the first antidepressant they take will only work for about a third of them. A third of them [will have a response that is] so-so; and [in] a third of them [the medication will not work at all, or they will even have] adverse reactions. So it’s quite clear that we can do a lot better.

Clare Blake: 
It’s a very crucial time because people who are about to go onto antidepressants need the help quickly. And it’s very much trial and error. If you could change that to, ‘This is your medication, this will work on you,’ that would, I imagine, save lives.

Nick Martin: 
That’s exactly our intention. At the moment, as you say, there’s just a lot of hit and miss in prescribing. It’s just guesswork. Two thirds of the time it doesn’t really work. And so they then have to washout, switch to another. This is highly distressing for the patient. Of course, it’s expensive too. The hope of our project is that we’ll be able to find genes that predict which people will respond best to which medication. Increasingly in the future everyone will already have all the genotyping done. And you can just simply do a quick look up and say, ‘Oh, well, actually you should use this drug because that one won’t work for you’. And we should be able to do that in a predictive way.

Clare Blake: 
How far away [from that are we]?

Nick Martin: 
I don’t think that’s more than about 10 years away. I think already there is a bit of this going on. I mean, for certain rarer conditions, like prescribing warfarin as a blood thinner for potential stroke victims. I mean, we know that for some people, there’s a highly adverse reaction to warfarin and there’s a simple genetic test that people can take. I’m not actually sure whether they don’t already take it before it’s prescribed. And there are quite a few other drugs for which there are already tests that can be done to determine whether it’s a good idea to prescribe that or not.

Clare Blake: 
Do we have to wait for the cost of doing a genome to come down? It’s quite expensive now.

Nick Martin: 
No, well you see, this is not true because you, no doubt, have been talking to the people who talk about [genenetic] sequencing, where it still is very expensive. It’s $1500 or $2000, or whatever. And if you add in all the bioinformatic processing, that’s highly expensive. The fact is you can get 98 per cent of that information with a gene chip that costs $20. And this is something that the sequencers haven’t really quite got their heads around yet. And in fact, at that cost we can afford to do, well, we could afford to do the whole population, if they wanted to.

Clare Blake: 
That’s right. Whether they’re prepared to have their genes on file.

Nick Martin: 
But I mean, in the future, one can imagine that people will just… See, once you’ve had that chip done, that then predicts everything.

Clare Blake: 
Well, I know blood pressure is the same. It’s very hit and miss, and that could be genetically…

Nick Martin: 
Exactly. Already we’ve got the ability to predict blood pressure, cardiovascular disease from a gene chip. When I say predict, these are all probability statements. That’s just telling you your risk. But in that sense, it’s no different from going to the doctor and having a cholesterol test or your blood pressure or whatever. And then you’ll look up a little nomogram that will give you your individual risks. So this is just adding one more figure to that, that will refine it. And that’s already powerful enough for coronary artery disease and for breast cancer and a few other diseases too. But our aim is to make that list very much longer.

Clare Blake: 
I’m going to say there’s a lot of hope here for a lot of incredibly vulnerable people. And a scientist remarked recently, ‘The long drought without much progress in understanding the genetics of depression has recently ended.’ Would you agree with that?

Nick Martin: 
Yeah, I think that’s right. Because, well first of all, as I said, we’ve known for a long time from twin studies that there are genes affecting depression. But in terms of finding what they are, it’s taken us a very long time. And when we published our first genome-wide association study on depression in 2009, and we had 10,000 cases: Nothing. And then two years later we published another [study, which] was I think 20,000 cases: Absolutely nothing. By that stage, we’d realised that we needed to have really huge samples. And so then my colleague Naomi Wray, who’s at UQ – used to be here at QIMR actually – published last year the first really successful study with 44 genes. And that was based on a sample size of 130,000 cases and about double that number of controls. And then that was followed within a year by another study, adding more people. So the sample is now double that, it’s 260,000 [people] and… over a million controls, and [from that we found] 102 genes. And then we’ve just added to that our Australian Genetics of Depression Study, which is quite small relative to that. But we’ve got about another 10 genes that have just come from what we’re adding.

Clare Blake: 
It really does sound like the drought is over. And at the end of the drought, there’s a flourish of spring that hopefully will be useful and practical.

Nick Martin: 
Well, that’s right. But I mean, we now need people with the appropriate skills, molecular skills, to actually pick up these hits, as we call them, from the genome-wide association studies and do the functional work to actually turn these into something that drug companies may be interested in.

Clare Blake: 
I should say, all the information in this podcast is general in nature and not personalised medical advice. You should always seek your own medical advice. If you do want to be a part of a very, very, very large study, which will help a lot of people now and in the future…

Nick Martin:
It’s the Australian Genetics of Depression Study and our website is geneticsofdepression.org.au. geneticsofdepression – one word, no dots, no capitals.

Clare Blake: 
You said that a few times. Thank you Nick.

Nick Martin: 
Pleasure.

Clare Blake:
Maybe you or someone you know needs help with some of the things we’ve talked about today. You can call Beyond Blue, 1300 22 46 36 or Lifeline on 13 11 14.