Jo Carlowe Hello, welcome to the Papers Podcast series for the Association for Child and Adolescent Mental Health, or ACAMH for short. I’m Jo Carlowe, a Freelance Journalist with a specialism in Psychology. In this series, we speak to authors of papers published in one of ACAMH’s three journals. These are the Journal of Child Psychology and Psychiatry, commonly known as JCPP, the Child and Adolescent Mental Health, known as CAMH, and JCPP Advances. Today, I’m interviewing Dr. Rosa Cheesman, Postdoctoral Fellow at the PROMENTA Research Centre, Department of Psychology at the University of Oslo in Norway. Rosa is the First Author of the paper, “Why We Need Families in Genomic Research on Developmental Psychopathology,” recently published in JCPP Advances. This will be the focus of today’s podcast. If you’re a fan of our Papers Podcast series, please subscribe on your preferred streaming platform, let us know how we did, with a rating or review, and do share with friends and colleagues. Rosa, thank you for joining me, can you start with an introduction about who you are and what you do? Dr. Rosa Cheesman Yeah, I’m Rosa. I’m a Postdoc based in Norway. I’m originally from Wales. And I’m really interested in figuring out how genetic and environmental factors work together to shape differences between children and their mental health. And my usual approach to this is trying to bring together these, kind of, different sides of genetics, so we’ve got the traditional twin and family side and then also, more recent, exciting developments based on DNA data or genomic data. Jo Carlowe Great, thank you. In your research, you combine twin and family data with new genomic tools to understand gene-environment interplay in childhood psychopathology, but I was interested to learn that you yourself were a participant in a twin study. Rosa, can you tell us a little bit about this and whether it influenced your career choice? Dr. Rosa Cheesman Yeah, that’s right. So, I have an identical twin called Margie, and she’s actually a Researcher as well. But since we were born, we have been participants in the Twins Early Development Study, which was set up by Robert Plomin in the mid-90s, and it follows 20,000 twins across the UK. It’s a really great study. Yeah, I think I was always a keen participant and then, when I got a bit older and started studying for my degree and doing a little bit of genetics, then it really hit me, “Oh, that’s what they’re actually trying to do in this study.” I eventually figured out that twins make such a great natural experiment for disentangling genes and environments. And then, yeah, I approached Robert Plomin, after I finished my degree, and they luckily needed a Research Assistant and I did that and then started later on doing a PhD, working on the actual data that my twin and I had contributed to, so that’s quite a nice feeling. Jo Carlowe Hmmm, yes, pretty amazing. And let’s turn to the paper now, so “Why We Need Families in Genomic Research on Developmental Psychopathology,” which was recently published in JCPP Advances. Can you give us a brief overview of what you looked at and why? Dr. Rosa Cheesman When I started my PhD research, the zeitgeist in genetics had really changed. So, I’ve already mentioned this, kind of, history of twin and family studies and that’s really the foundation of genetics. But then since the, what some people call, DNA Revolution, like, that came after the Human Genome Project, we had this wave of studies based on actual DNA differences between people. Like, we could suddenly start to find actual pieces of DNA, not genes, that cause risk for mental health problems and we could start to allocate measures of people’s individual risk for mental health problems. So, we couldn’t do any of that with twin studies and there was a lot of excitement about this new DNA technology. So, even at the beginning of my research on the Twin Study, I was using DNA methods, but I was deleting observations for one twin from each pair. As the field has been developing, we’ve become more and more aware that this family data is extremely valuable and we need to bring back the family data, but also insights from twin studies if we’re going to push the field forward and address challenges in genomics. So, yeah, the rationale behind the paper was to take stock of where we are at with genomic studies of childhood mental health and then try and figure out what’s valuable about family data and how can we continue to make the most of that. Jo Carlowe Is there a, kind of, key message in the study? Dr. Rosa Cheesman The key message really is that without taking the best of both worlds, like, the innovations with DNA-based technologies and then also the family structure, without combining those, we’re going to miss out a lot of information on the social context of genetic influences. And we really need to bring these worlds together if we’re going to understand the mechanisms how genetic and environmental forces work together. Jo Carlowe In the paper, you review two related challenges at the cutting edge of genomic research and developmental psychopathology. So these being missing heritability and, secondly, distinguishing between genetic, environmental and confounding effects. In lay terms, can you help us understand these challenges? Dr. Rosa Cheesman I think maybe I’ll focus on the second one. If we take this issue of disentangling genetic and environmental effects, this is such a longstanding problem, and it doesn’t just go away when we start studying DNA. One of the coolest tools in these new DNA studies is what I mentioned before, the personal, individual measures of genetic risk, which we call polygenic risk scores or polygenic indexes. But for most purposes, the whole idea behind that is that they’re measures of people’s personal or individual genetic risk for a certain mental health trait or disorder. But a key issue is that actually, it’s not just personal, but it’s circumstantial. And a key reason for that is that if you imagine DNA that I’ve inherited, that increases my risk for depression, any piece of DNA like that will be present in my mum or my dad, or both of them, and that DNA will also be likely to exert effects through them, shaping the way that they behave and the way that they create the environment for me. So, what we’re really finding is that these polygenic risk scores for individuals don’t just capture individuals’ genetic risk, but they capture their environmental risk. Jo Carlowe Hmmm, yeah, and that’s really helpful. And what about missing heritability, which is the other challenge? Dr. Rosa Cheesman Yeah, so the missing heritability issue is about how we – we have this benchmark of how much genetic variation influences differences in mental health from twin studies. For example, we know that for childhood depression and anxiety, genetic variation explains about 40% of differences between children in their symptoms. But then when we look at the more recent DNA-based methods, we can only capture a small amount of the inheritability that we’ve got from twin studies. So, it’s called ‘missing’ because we know it’s there from twin studies, but we just can’t seem to pick it up when we’re measuring actual DNA. So, that’s a big problem because if a lot of our new evidence is not playing with very much signal and it’s a lot more noise, then we’ve got big trouble in interpreting what the findings really mean. Jo Carlowe Hmmm, that’s super helpful, thank you. Rosa, your paper focuses on three approaches to combat these issues, can you talk us through them? Dr. Rosa Cheesman In the article, I didn’t want to stop at saying, “Oh, there are loads of challenges and we need to inject the family context back into genomics.” I wanted to also try and provide a framework for how people can go about that. So, I described three different approaches that I can take in the field to bring the family context back in. So, one of them is to integrate family data in with our DNA data, so that we have this family structure. The second one is to triangulate family and twin studies with DNA based genomic studies, which means that we directly compare results across different ways of doing things and acknowledging that there are different strengths and weaknesses. So, if we get results that, kind of, match up well across different methods, that’s much stronger evidence. And then thirdly, I wanted to suggest that even if people don’t have family data, they can still use knowledge, concepts and findings from more traditional twin and family studies when trying to design and interpret genomic studies. Jo Carlowe Thank you, and in the paper, you say “Developmental Psychologists are particularly well-placed to capitalise on these strategies,” can you elaborate why they are so well-placed and how might they be encouraged to get more involved? Dr. Rosa Cheesman Yeah, sure, so if we take a specific example of – relating to this difficulty of separating direct individual genetic effects from the family effect. So, first of all, Developmental Psychologists are really well-placed to find ways to integrate data from families and genomics, like, we get the inspiration from classical psychology designs. So, for example, one study that I did previously was really inspired by the classic adoption design in psychology where I was interested in this idea that polygenic risk scores capture the environmental as well as genetic effects. So, to test that, I compared how much the polygenic risk scores predicted educational attainment in people who were adopted versus people who were raised by their genetically-related parents. And I saw that the effect of the polygenic score for education, it was still significant, but it shrank by half when looking at people who were adopted. So, in the space of education, we could really combine this classic adoption design with DNA data to be able to evaluate the interpretation of the results. I think also, Developmental Psychologists have a lot to contribute in genomics now in terms of digging deeper and trying to understand how technologies like polygenic scores, how well they predict things over time and how they’re really mediated through specific traits and thoughts, feelings and behaviours and the environments. Jo Carlowe And any thoughts about how they might be encouraged to get more involved? Dr. Rosa Cheesman Yeah, that’s a good question. I think that we have a lot of people who have training in statistical genetics and twin studies and psychology. But, yeah, I think that a key thing is to try and widen access to training in all those things, so that people with different expertise can talk to each other better and develop more of a shared language. ‘Cause I think what’s quite interesting is that there are starting to be some terms that, kind of, mean the same thing, but they’re – the terms are different between different subfields. So, for example, we have this term that came up in about 2018 in statistical genetics world called ‘genetic nurture’ and that captures the notion that parents’ genes influence how they parent their children. But actually, that concept really not new because if you look back to developmental psychology and twin study literature, that comes under the umbrella of ‘gene environment correlation’, which is this idea that your genes influence the decisions you make and how you create a niche around you. That also overlaps with other concepts from, like, behavioural ecology and animal behaviour. So, yeah, I think it will be really cool if we can move forward in childhood mental health studies by using genetic methods and also taking all this knowledge from other disciplines. Jo Carlowe Hmmm, yeah, that’s such an important message. Rosa, is there anything else in the paper that you would like to highlight? Dr. Rosa Cheesman One thing I’d like to highlight is that even when we have this family context and we also have the DNA measures, so we can really separate out what’s the individual’s genetic effect and what’s the effect of their parents, when we do that, that doesn’t mean that genetic effects are somehow purely biological. So, I think a really key point is that genetic influences on mental health always work through the environmental in the same way as, like, your genetic influences on lung cancer go through smoking behaviour. We need to keep in mind, like, the psychology and the traits in between the genes and the outcome. Jo Carlowe Are you planning any follow-up research or is there anything else in the pipeline that you would like to share with us? Dr. Rosa Cheesman Well, I’ve been really happy to publish an article recently in the JCPP, apart from this review article. It was an empirical study of genetic risk for ADHD and how that impact on children’s education in a way that’s very dependent on the school context. So, that was an interesting example, I think, of how we can use genetic data on families and then combine that with the wider social context to understand really how our schools work and get a better understanding of gene-environment interaction. Jo Carlowe Finally, Rosa, what is your take home message for our listeners? Dr. Rosa Cheesman Yeah, my take home message for the listeners is that when we care about these new DNA-based technologies that claim to, kind of, have a prophecy of your personal genetic risk for something, we have to keep in mind that the family is important there. We need to interpret those predictions, in the context of the person’s family, because it’s very difficult to separate genetic and environmental influences. And that doesn’t mean that it’s impossible, but we have to always keep in mind the context. Jo Carlowe Brilliant, Rosa, thank you ever so much. For more details on Dr. Rosa Cheesman, please visit the ACAMH website, www.acamh.org, and Twitter @ACAMH. ACAMH is spelt A C-A-M-H, and don’t forget to follow us on your preferred streaming platform, let us know if you enjoyed the podcast, with a rating or review, and do share with friends and colleagues.