UCL Health of the Public


Transcript for 'The genetic revolution'

Xand van Tulleken

Hello and welcome to season four of Public Health Disrupted with me, Xand Van Tulleken. And at this point, Rochelle would normally say, and me, Rochelle Burgess, but unfortunately she has been called away for a family emergency at the last moment. So I'm going to say instead, Rochelle isn't joining us. She's normally my cohost. I'm a doctor, a writer and a TV presenter. Rochelle is a community health psychologist, and she is also an associate professor at the UCL Institute for Global Health.

This podcast is about public health. More importantly it's about the systems that need disrupting to make public health better.

So join us each month as we challenge the status quo of the public health field, asking what needs to change, why and how to get there. In this episodes, our experts are diving into the exciting world of genetics, exploring the latest technological breakthroughs and innovations. They'll discuss not only the value of these advancements, but also the ethical considerations surrounding them, particularly in how they can enhance the health of the public.

Our first guest today is Adam Rutherford. Adam is a lecturer in genetics and society at UCL where he teaches about genetics, evolution, race science and science communication in the biosciences department. He's also a Sunday Times bestselling author, a broadcaster, a columnist with over two decades of experience in writing about biology and the origin of life. And I should say he's one of the select group of people that my brother calls whenever he doesn't understand something which is reasonably often.

So Adam, Thank you so much. And we're also delighted to welcome George Davies-Smith, who is a clinical epidemiologist and director of the MRC Integrative Epidemiology Unit. His work looks at how we understand the cause of diseases and how to prevent them. George is known for his groundbreaking work in mendelian randomization, a method that uses genetic variation to identify the causes of disease that can then be changed to reduce the risk of the disease. George is passionate as shown during his leadership of studies like the Avon Longitudinal Study of Parents and their Children, So George and Adam, thank you so much for being here.

Adam, can we start with you? Because when I trained in medicine, the world of genetics was sort of exploding. And for my entire career, both as a broadcaster and a doctor, there has been this extraordinary promise, a great deal of which has been delivered, I suppose. Can you give us a quick tour of genetic epidemiology, epigenetics, as it's kind of relevant to our life?

Adam Rutherford

The history is really interesting and I can talk about this for absolutely hours. In fact, I do as a lecturer at UCL. So I was at medical school in 93 and the human genome projects officially started in 92. And I did medicine for a year and then I went to do undergraduate genetics and I did a PhD in genetics and I finished my PhD in 2001.

Now, the first announcement of the completion of the human genome projects, and I'm using air quotes for the people who can't see me, was in July 2000. And this was when it was a big deal. And President Clinton was on stage with Tony Blair was on the monitor behind him, and the two leaders of the private and the public consortium for the human genome project were there on stage. And this was a big deal, a landmark in the history of science.

And Clinton said a couple of things, really grandstanding things about how this was going to change everything in science. He said that today we have learned the language of God. Now, I remember the BBC report. It was a mutual friend of ours. Vivian Parry did this report and covered the announcement. And I remember her saying that Experts predict that something like all human diseases would be understood within the next couple of years and many of them will be cured. Now I was looking for various genes in the draft sequences. Now I don't need to tell the audience how complex the human genome is. It is probably the most complex data set that exists in the known universe.

At that time that it was announced that we will cure all diseases, I was looking at a sequence of about, I think it was about 180 letters of genetic code on chromosome 14, and we didn't know which direction it was pointing in. So I remember sitting in the lab and watching this announcement going, yeah, I think you might be overselling this just a little bit because what I'm looking at doesn't make any sense whatsoever.

This was my first real experience of sort of grandstanding in science, the hype machine that I think is worth sort of trying to understand and scrutinize without sort of completely throwing it out. I do think that the Human Genome Project was one of the most important aspects of the biological sciences, perhaps ever, but not for the reasons that it got sold as.

The number of genetic diseases that have been cured as a result of a sequence in the genome, which basically was announced in 2000, and the finally final complete annotated one was published about six months ago, right? The number of diseases cured since 2000 when that announcement was made is effectively zero. I mean, it's not what was being sold by the media. However, I also believe that there wasn't a way that biomedical sciences, including genetics, which is the one that underlies all other aspects of biology, there wasn't a way in which it could progress without the Human Genome Project having delivered, incidentally, as a public project, on time and under budget.

Our understanding of the major diseases has been utterly revolutionized. The understanding of evolutionary biology and the origin of life and um uh things like human evolution you know just things that we love talking about in the media like the fact that Neanderthals were our ancestors and not our cousins that is a result a direct result of projects such as the human genome project.

Xand Van Tulleken

Can I ask about a couple of those diseases? I had tutors at medical school working on cystic fibrosis, working on muscular dystrophy, who seemed kind of agonizingly close at the time and it feels like we are inching ever closer. From someone who's a bit closer to this research than me, we would begin to sort of go, yes they are treatable and in some cases curable. Is that the sense you have or are you really now trying to manage the hype down a bit?

Adam Rutherford

No, I think these are good and interesting examples, because one of the parts of the narrative of the history of understanding genetics or inheritance, because it predates the whole field of genetics, we're culturally wedded to the idea that there is a deterministic view of biology. The idea that you are fated to be in a certain way.

Now, when genetics emerges as a science at the beginning of the 20th century, this is sort of reinforced. The whole history of genetics has pointed us in one direction, which is that there are genes, so units of inheritance, that have a sort of deterministic effect on biology. Mendel was working on pea plant, so it was a gene for crinkly skin, or a gene for purple flowers, or a gene for the height of a pea plant. If you have this version of the gene, this is the outcome, whether it's blue eyes or a receding hairline or particular diseases like cystic fibrosis or Duchenne muscular dystrophy or whatever.

In the 80s, we got really good at identifying the disease genes that actually cause diseases like cystic fibrosis and Duchenne. And the reason they were the first ones identified is because they follow this very clear inheritance pattern, which is like the way that the peas, the Mendel's pea plants work. So they're Mendelian in their inheritance pattern.

Now, by the time you get to the human genome at the beginning of the 21st century, what we really discover then is, well, one, those are the outliers, and two, everything is much, much more complicated than the story we've been telling ourselves for 100 years you ask the question about if we've made progress in understanding those and treating those diseases. The answer is absolutely yes. We've made astonishing progress and in the time since me and you were at medical school here at UCL to now, life expectancy for cystic fibrosis has gone up from the fourth decade to the fifth or sixth, right?

Duchenne muscular dystrophy is one of the diseases which has the highest chance of being treated by gene therapy in the current era. So in a sense, these are great examples of how huge amounts of progress have been made as a result of genetics, but not at the same rate or in the same way that the public discourse was sold back in the late 80s, 90s, and then up to the point in the Human Genome Project where we're saying, we've nailed this, now we understand genetics.

Xand van Tulleken

George, can I move from the kind of clinical examples maybe that Adam was talking about into population health and talk about how this knowledge of genetics has begun to affect the way we think about our total populations.

George Davey Smith

Yes, so population health science, epidemiology in particular, is fundamentally concerned with identifying causes of disease and things that you can then intervene on to improve population health. And it obviously is most focused on the diseases which have a very high burden of morbidity and mortality in populations, which are generally multifactorial complex diseases like type 2 diabetes. currently heart disease, like many cancers, for which there is no major cause that's been identified.

Xand van Tulleken

And George, why are there so many genes in type two diabetes? Like why isn't it just caused by one gene, the way that cystic fibrosis is?

George Davey Smith

Well, to put that very simply is that the main modifiable factor that relates to type 2 diabetes is your body mass index. And within that, it's the particular distribution of fat within the body. And a huge number of genetic variants relate to body mass index and body fat. So once you get large enough sample sizes, all of those vast number of variants will relate to type 2 diabetes if you have enough numbers. And the reason why there are large effect mutations of low prevalence that relate to hugely different levels of body mass index. But those of course will not reach high prevalence because they have severe disadvantages and will have had historically very severe disadvantages in terms of survival. So the variants, the common variants which are being examined are ones that are either nearly neutral with respect to survival and the number of offspring you produce because of survival past reproductive or to and then past reproductive age etc.

There's things which have very large effects, will tend to have adverse consequences and don't move beyond low prevalences, which are recreated through mutation, but then they don't move to high prevalences. Or very occasionally, things which are advantageous and then rather quickly move through populations. And some of those are currently polymorphic around the world and clearly in some places at least had were very strongly selected for. Stories become more complex the more we understand them, and also genetic stories become more dependent upon the environment the more we understand them.

Xand van Tulleken

I love your axiom that stories become more complex the more we understand them. I feel like that may be universally true across multiple fields.

Adam Rutherford

Cystic fibrosis, is also a good example within this story about stories becoming more complex the more that we understand them. Because even though, I think the numbers are that, I think something like 70% of cystic fibrosis in white Western populations is caused by a single mutation, single deletion in the CFTR gene. So that is a gene of huge significance in the etiology of the disease itself.

But we talk about penetrance in genetics. So the idea of how the weighting of the underlying genetic architecture in how the disease actually manifests. And people in the cystic fibrosis community, you will see that there is an enormous range in how severe the disease actually manifests. And the more recent studies which have looked at the whole genome rather than just the one gene which is the underlying architecture of the disease. I've shown that there's at least, I think from memory, there's at least five or six genes that have an enormous effect on the severity of the disease itself.

Xand van Tulleken

I want to come back to population health and making health policy, um, because it seems so relevant to what we've just been talking about. Can I ask you then a bit about consumer genetics? There are a number of different companies you can send a cheek swab to, and they will tell you all kinds of things about yourself from how caffeine sensitive you are through to your risk of certain diseases and so on. It feels to me like what you're hinting at is that these services may not be nearly as valuable and impressive as the marketing makes them out to be. Is there a role for kind of private consumer genetics at the moment in terms of health or is it just curiosity and ancestry and these kind of things? Adam, can I start with you?

Adam Rutherford

I have quite a hostile relationship with the consumer genetics markets, mostly for scientific reasons, that you've sort of alluded to. I think they're a manifestation often of what we're talking about, which is that the story is more compelling than the complexity of what science can actually say. So there's a number of reasons for that, one of which is that the majority of the way that these data is compiled. So when you take a cheek swab or you spit in a tube with some saliva and you send it off to a company, and what they do is they extract your DNA, your genome, and they compare specific bits of your genome, individual letters with their own database or a public database about which we know certain things about what those individual letters are associated with diseases or with ancestry or so on. Now, that sounds like it should be very informative, but in fact, it has limited informative power, partially because the way that these databases are, the way that this knowledge is acquired is reliant on populations. And so they have limited statistical power for individuals whilst being very powerful for large groups of people. You only begin to see the significance of these variants in the genome if you look at thousands, tens of thousands millions of people. And I think you just have to approach these things with enormous amounts of caveat, emptor and scepticism. There is some utility in them. And ancestry testing is very effective. Well, genetic testing is very effective for identifying immediate relations like brothers, sisters, parents, first cousins and so on. But relatedness genetically drops off a cliff after about third cousins.

Xand van Tulleken

But George, what strikes me about these conversations is that, of course, they will bleed into the thinking of people making policy. Surely this has implications for the NHS and for the government. How are things like general wide association studies or the use of AI and genetics or how are you seeing it relate to health policy?

George Davey Smith

I think the point with the genome-wide association data is, I think, very rapidly, it's so cheap. I think everyone should have an NHS chip, which has as additional variants all the rare variants that relate to drug reactions. There are many variants that relate to drug reactions. The drug reactions are uncommon. But if you've just got those, if you just have those data on your computer and it's just there, you would use it and it would be useful. There needs to be medical education in ways that allows the next generations of doctors people to use it. I mean, why start, if you actually have the data of the 4% prevalence variant which massively increases probability of muscle aches for people on statins, why wouldn't you look it up before putting someone on a statin and put them on a PCSK9 inhibitor if they have that variant?

I think it would be valuable and also it would be valuable because one would actually properly train the next generation of doctors to understand those data and to be able to actually include them in decision making and not to have an inflated view of the role of genetics at the individual level, but understand how in some circumstances it could be valuable. But the wonderful thing about genetics, is actually through the randomisation of genes within families. They are truly randomised from parents to offspring, as close to randomisation as you get in natural systems. So that means that the genetic variants don't become associated with all the confounders which have led to extraordinarily misleading findings in epidemiological studies. It's extremely difficult to measure exposures that are socially patterned, but in particular exposures that are then influenced by your health status. Most diseases when you get ill, they'll have some effect on your behaviours, but on your circulating cholesterol level, they will affect your weight, anything you get. When you go to the doctor, the doctor will tell you to drink less alcohol.

So, you know, so they'll have that they get this reverse causation where the sort of outcome or things leading up to the outcome influences the exposure. And there's been some of the some of the very strongest epidemiological associations when I started out in epidemiology were things like HDL cholesterol that related to lower risk of coronary heart disease. And it was that was believed to such an extent that it was being called good cholesterol. This huge investment of pharma money in randomized control trials, literally billions spent in randomized control trials, raising HDL cholesterol, and there was no benefit. But no observational analysis could explain away this very strong inverse association which if they had been available it would have said don't invest in HDL cholesterol raising, was the genetic data, where differences in common genetic variants when sort of added together, whereby you could then put people into groups who did not differ by anything due to reverse causation, anything due to confounding, those groups did not have different risks of coronary heart disease. Whereas if you did that, you group the people by genetic variants related to LDL cholesterol, sure enough, you show this very strong positive association, which you can then say is a likely a causal association. And, you know, for me, the most depressing example but one which has direct policy relevance is alcohol, where for many years it was said that, you know, moderate drinking might lower cardiovascular disease risk and there might be beneficial effects of red wine, etc. Which obviously influences the sorts of messaging that's given about alcohol. You could do the Mendelian randomisation studies show very clearly that there is no beneficial effect at any level of alcohol consumption. It's adverse all the way. For cardiovascular disease, it's adverse, particularly through blood pressure, which it has a very strong effect on. I think all health, giving information about health risks should be about allowing informed choices, allowing people to make choices on the basis of what's true.

Xand van Tulleken

George, those are all kind of beautiful examples of the complexity and the way in which the stories that we tell can be kind of misleading, especially when we're thinking about genetics in a less complex way. And yet there are real policy implications for things like, I guess I'm thinking of the UK bio resource or the decipher program in Cambridge with looking at children's DNA going forwards.

Again, over the next generation, this will yield incredibly important results. Where do you see, I don't know, in 10 or 20 years time once those programs have followed people for longer periods of time, the policy implications?

George Davey Smith

Yogi Berra said it's hard to make predictions, especially when they're about the future. The policy implications come from what genetics tells us about, about preventing things, alleviating things through making environmental changes. Because there's this whole sort of notion of gene editing is being tried out supposedly in the New Zealand and the UK for cholesterol. Interestingly, the ongoing trials are in people with familial hype very high cholesterol, who are having the PCSK9 gene edited to lead to having very low LDL cholesterol levels or low LDL cholesterol levels. It's a test case in a very unfavourable economic situation, I think, because there are drugs which can produce the same effect, small interfering RNAs, injections every six months can lower your cholesterol the same way. And even if I was in the range of it being thought I should, you know, go for such, I wouldn't go for something which would be permanent, like, gene editing.

Xand van Tulleken

Very interesting. So, okay, so there is promise and hype, but still caution about exactly how policy might be made around this. The cases you're raising are very interesting ethically, and I think genetics has this long complicated history of digging very deep into what makes people who they are and the stories we tell around it therefore have ethical implications as well as the research itself and the idea of quite fundamentally perhaps changing people. Adam, you've written a huge amount about this and your bookshelf behind you I know is kind of laden with Eugenics books. Can you talk a little bit about what we could learn from the history of genetics and the thinking around eugenics and what we need to be cautious about going forwards?

Adam Rutherford

I've sort of evolved into becoming a historian of science, mostly because I can understand some of the things that George and my colleagues actually does. You become so aware of how history does repeat and the same conversations that we have now have been happening throughout history. Now, in looking at the history of genetics, which is where I sort of got into the field, you very rapidly come across two key political ideas which are at the absolute foundation of the fields of biology and more specifically genetics. And the first is scientific racism, right? The foundations of biology are effectively in service of European expansion, not in parallel with, but actually the classification of humans is designed by the first effective biologist whose influence is still very present in our work today in order to justify the subjugation of people mostly in Africa, but also around the world, right? So the roots of the field of biology are associated with the political ideology. And then the same thing happens, but with a slightly different nuance in the 19th century, which is that after Darwin and the discovery of natural selection, evolution by natural selection, there is the emergence of the, well, I describe it as the science fiction of an older idea, which is basically population control through an historically, It's through arranged marriage or infanticide. And what it is it emerges not just in the UK as an idea, but in the building that I'm sitting in now and effectively associated with the genetics department at UCL. And the thing that we were talking about earlier about the rediscovery of Mendel's brilliant work, which was done in the 1860s in Moravia, is that it was translated into English in the first decade of the 20th century. And it coincided perfectly, just a cosmic happenstance of coinciding perfectly with the growing concerns about immigration, about eugenics, which was enormously popular at the time. And so particularly in America, less so in the UK, but the American eugenicists, really latched onto this idea of Mendelian inheritance with such a fervour that the ramifications, the repercussions, the echoes exist in our present day. So for example, trivial example, Xand, what is the first human characteristic that we use to teach Mendelian genetics in people?

Xand van Tulleken

I would say eye colour, right?

Adam Rutherford

And everyone knows, because you learn this at the age of 16, that you draw one of those nice tables with brown eyes dominant over blue eyes, and you need two copies of the blue-eyed version of the gene in order to have blue eyes, right? Now, it's not really true. I mean, eye colour, like so many other characteristics, is influenced by dozens of genes. Eye colour isn't a binary thing. It isn't brown or blue, people have different eye colours within the same iris, between irises, and so on. And because the key mistake that people make in genetics is to look at it as deterministic rather than probabilistic, we culturally embedded this idea that if you have two blue-eyed genes, then you'll have blue eyes. Now, where does that come from?

It comes from the eugenics movement. It comes specifically from the leader of the eugenics movement in America, a guy called Charles Davenport. And so many of these ideas come specifically from that lab. I get my students to list like 20 genetic fallacies of monogenic determinism, and they're things like, uh, tongue rolling.

We've known that's not true since 1940 when studies were done on twins where they were discordant. So one could do it and one couldn't. Things like widow's peak, cleft chin, attached earlobes, the hitchhiker's thumb. They all come from this era the first 20 years of the 20th century, where particularly American eugenicists were really obsessed with finding the gene for a characteristic. And there was a political motivation for this because they were eugenicists who figured that if you could draw a pedigree of every characteristic, both desirable and undesirable, then you have a pivot on which you can enact eugenics policy, which is what they did in America. Tens of thousands, maybe hundreds of thousands of people were sterilized against their will or without their knowledge in the 21st century in the majority of American states because they had such an enthusiastic eugenics policy. Now we now know, and I think we knew at the time that this was pseudoscience because many other geneticists knew that the inheritance patterns of simple traits, supposedly simple traits like eye colour, didn't hold true. But also they adopted this model, this Mendelian model for sort of really poorly defined pseudo psychiatric diagnoses like feeble mindedness, but also mad stuff like sexual proclivity or even seafaringness. This is the foundation, a sort of cultural foundation of genetics, the legacy of which is in the language we continue to see in the press. Scientists discovered the gene for something where that something is a complex behaviour. The idea that this was going to be a simple relationship is scientific lunacy. You know, the real core of what George and I share is that the problems here are to do with scientific literacy, genetic literacy, and understanding how genes work.

George Davey Smith

I think there's a case can be made that the causality of mendelian genetics to eugenics, the causality might run the other way that it's it was just in the US, it was one branch of mendelianism in the US that made it look that way because in the UK which was seen to be that sort of quite important in the intellectual origins, if you can call it intellectual, the origins of eugenics, there was the famous, often oversimplified divide between the Bateson and the Mendelians and Pearson and the biometricians. The most virulent eugenicists, many of the most virulent eugenicists like Pearson were strongly and totally anti-Mendelian. I mean, they weren't just arguing in a different way for quite a stage they were totally opposed or highly opposed to Mendelian ways of thinking.

In the UK, some of the best known eugenicists, if you just go and read Nature in the early part of the century, someone like Ernest McBride, was a Lamarckian. This always brings me to a real bugbear of mine, which is the current promotion of the notion that there is a serious amount of transgenerational epigenetic inheritance in mammals, which there isn't, and there isn't because of Mendel's laws and because of the reprogramming of most of the epigenetic material at gamete formation and conception and early development.

If you actually think it through, the actual implications of if there really was transgenerational epigenetic inheritance and the market inheritance in the way McBride says it is there just truly is a biological continuous imprint of bad things that happened to particular groups and you think of the particular groups to which bad things happen, you can see it would then be said to be damaged by that generations down. I think Mendelian genetics is incredibly liberationary, gives everyone a fresh start with a sort of roll of the dice. I mean, sadly, the social inheritance, of course, is what leads to the transmission of disadvantage, but thinking that it was literally embodied.

Adam Rutherford

My old tutor, Steve Jones, who you will all know, he was used to open his introduction to genetics first year lectures by saying, what is asking us what is the single most heritable trait that exists? And not a student ever gets it right. And it's money. It's just wealth, right?

It's not quite as straightforward as the stories that we sometimes tell. Now, Francis Galton, the guy who comes up with the, with the word eugenics, who's Darwin's cousin and really develops the concept from a, in a sort of, I think, pseudo scientific way, but from a statistical way in the second half of the 19th century bequeaths a significant fund to UCL in 1904 to set up the Eugenics laboratory effectively. And he was included in that, is the first professor of Eugenics, a position which still exists, although not in that name.

Really significant biologists of the 20th century hold this post. Carl Pearson, I think, is arguably one of the most important scientists who's ever existed. About, I don't know, probably about 40% of statistical techniques that we use today come from Pearson. The second professor of eugenics at UCL was Ronald Fisher, R.A. Fisher, and I'd say about 40% of the statistical techniques that we use today came from Fisher.

Now these guys were active eugenicists. Pearson's resistance to the Mendelian view that the Americans were promoting was not because he was opposed to eugenics. It was because he was a virulent eugenicist and a virulent racist and anti-Semite who felt that the crap science that they were doing in Cold Spring Harbor where the American eugenicists was based was detrimental to the global eugenics movement.

There is a difference between the US and the UK, which I think can be characterized quite simplistically like this. A lot of the eugenics movement and sort of enforcement of eugenics in America came from scientists, particularly the Cold Spring Harbour lot. In the UK, it was much more a grassroots movement organized by the upper middle classes. And I would say and have argued in various places that, for example Fisher and Pearson's influence on the global eugenics movement was minimal. We never had a eugenics policy in this country. Their rejection of the ultra mendelian version, which was the framework for American eugenics, and I think that America is the second most enthusiastic embracer of eugenics, they were completely distant from that. And the first most enthusiastic embracer of eugenics, at least in the West is Nazi Germany. And the main scientific, financial and legal influence of the Nazi eugenics machine, which concludes in the Holocaust, was the Americans. They were literally funding, directing and providing the legal framework for the establishment of what became the Nuremberg Laws in 1933 and 35 in Germany. So I am not, I'm not excusing UCL's role in this history. I just think it's important that there's good scholarship in this. A few years ago, it started in 2018, UCL had its first official eugenics inquiry, which concluded just before COVID started in 2020. And some of the outcome of that was the redistribution of the money that Galton had left, which had been well-maintained for a century.

And that was redistributed to various new members of staff, excellence fellows, and me. So my salary is literally paid for by Francis Galton. And that is a thing that I will never not find very amusing. The second thing is the decision to remove the name of Pearson, Fischer, and Galton from the UCL campus. I don't think you should remove the names of people unless you contextualize why you've done it. And that's the course I teach here.

Xand van Tulleken

I do, I love this thing about Francis Galton playing your salary. We should say, I mean, you've written so many books, Adam, but a brief history of everyone who ever lived, how to argue with a racist and possibly my favorite, where are you really from?

Adam Rutherford

thank you for saying that.

Xand van Tulleken

So we ask every guest what piece of art or music or poetry or literature or something you've encountered. It could be a scientific paper, could be another human. What has disrupted your perspective and helped you do your work?

George Davey Smith

Well, I can say the only thing that affected my life the most was hearing Sister Ray when I was 15, the Velvet Underground, and it's still my favourite song to this day. It does everything that in those days you never heard songs do. I mean it was nearly 18 minutes long and you didn't hear songs that went on for 18 minutes. It had verses and choruses but then it broke out into sort of Ornette Coleman style, sort of free noise in the middle of it but it kept on with a... it also had this continuity just of having this for 17 and a half minutes. Mo Tucker just played this is a very rather simple drum pattern. It remains the most extraordinary piece of music.

Xand van Tulleken

So good, and once you've heard something that's deviated that much from what's gone before it, you kind of can't look back. That's fantastic. Oh, I love that

Adam Rutherford

I'm going to follow up with a music one based on what George was just saying, which is not quite the same thing, but Taylor Swift is a big part of our family life. I probably wouldn't have been into Taylor Swift as much as I am if I didn't have daughters of a suitable age. But at a gig once one of the questions from the audience was knowing that we like a bit of Taylor Swift in our house, but also I'm a sort of Darwinologist and somewhat obsessed with Charles Darwin. The question was, you have to pick one and the other one is erased from existence. Taylor Swift or Charles Darwin?

I think the audience assumed that I was obviously gonna choose Charles Darwin. But the truth is, and it would always be Taylor Swift because Darwin described one of the best ideas that anyone's ever had, but it was just waiting to be discovered because that's all we do in science, right? This stuff is out there. It's about designing the experiments to reveal the thing that is just.

part of the natural world. And we know this from the Darwin story anyway, because Alfred Russel Wallace came up with approximately the same idea at approximately the same time. But I am telling you, you rerun the tape of the universe again, and no one writes, Shake it off. I think that Taylor Swift is the natural inheritor of Bruce Springsteen's crown of narrative storytelling in music.

Xand van Tulleken

That's fantastic. Thank you so much both for being so generous with your valuable time. We really are very grateful. It's totally fantastic.

You've been listening to Public Health Disrupted. This episode was presented by me Xand van Tulleken, produced by UCL Health of the Public and edited by Annabelle Buckland at Decibelle Creative.  Our thanks again to today's guests, Adam Rutherford and George Davey Smith. If you'd like to hear more of these fascinating discussions from UCL Health of the Public, make sure you're subscribed to this podcast so you don't miss future episodes. And come and discover more online and keep up with the school's latest news, events and research. Just Google UCL Health of the Public. This podcast is brought to you by UCL Minds, bringing together UCL knowledge, insights and expertise through events digital content and activities that are available to everyone.