This week’s guest is Dr. Jacob Michaelson, who talks about his work in autism research and the SPARK Study, a landmark autism research project that seeks to build a research community of tens of thousands of individuals with autism and their families. It’s a great discussion on an important topic with strong ties to traditional public health disciplines such as epidemiology and biostatistics.

To find out more about Dr. Michaelson and the SPARK study, visit michaelson.lab.uiowa.edu/spark-study

Emma Meador:

Hello, everyone. Welcome back to From the Front Row, brought to you by the University of Iowa College of Public Health. My name is Emma Meador, and I am joined today by Alexis Clark. If this is your first time with us, welcome. We’re a student run podcast that talks about major issues in public health and how they’re relevant to anyone, both in and out of the field of public health. Today, we are joined by Dr. Jacob Michaelson, who serves as the division director for computational molecular psychiatry. He also is an associate professor of psychiatry, biomedical engineering in communication, sciences, and disorders. Dr. Michelson also has numerous publications, primarily in the areas of genetics of neurodevelopmental conditions, gender difference in neurodevelopmental conditions, genetics of language ability, and machine learning to improve the interpretation of non-coding genetic variance. Today, he is here with us to discuss the SPARK autism research project and how the University of Iowa plays a vital role in this.

Alexis Clark:

Welcome to the show, Dr. Michaelson.

Jacob Michaelson:

Thank you. Thanks for having me, you guys.

Alexis Clark:

So to start us off, can you just give us an overview of your personal career path and how you landed at the University of Iowa?

Jacob Michaelson:

Yeah. So, well, first of all, I guess I’ll start right at where I am right now. And I’m in a department of psychiatry, and it’s an unusual place for me to be because I’m not a psychiatrist, I’m a PhD scientist. And actually, if you want to get really specific about it, my PhD is in computer science. So what in the heck is a computer scientist doing in a department of psychiatry? Well, I guess the short answer to that is that I do genomic research because many psychiatric conditions are heritable, meaning that they have a strong genetic component. And to be able to navigate and do the kinds of analysis that are required in genomics today, you have to have a pretty strong computational background. So that’s where I am right now. And I love being in a department of psychiatry because you’re very close to the actual human beings that you’re studying and you’re trying to understand better.

Jacob Michaelson:

And I love the colleagues that I have in this department, the clinicians and the practitioners. And so, yeah. So how did I get here? I’ve always been really interested in a wide variety of sciences and especially if they don’t on the surface seem like they really go together. And so that started off, I was really interested in biology, but I was also really interested in engineering. I have kind of an engineer’s mind, I guess. I like to build things and I like to take things apart. And so I studied biological engineering, I got a bachelor’s and master’s at Utah State University. And that was one of the few programs at the time that would allow you to do a lot of really heavy math and also computer science and programming and stuff like that, but also get a background and a foundation in biology.

Jacob Michaelson:

And so I took that and I decided I’d done a lot of wet bench lab work, but also computational work too. And I reached a fork in the road where I decided that I actually really enjoy the computational part because if an experiment goes wrong in your code, you just fix it and you run it again. Whereas if you’re running a lab or running experiments, it could be thousands of dollars. And at a lot of time in setting up those experiments. And I knew that I really like to iterate very quickly. And you can do that when you do computational research. So I did my PhD in computational genomics and systems biology in Dresden, Germany. So at the Dresden University of Technology. And I really enjoyed doing that overseas and learning how science works in Europe. And we had a very international team and I really enjoyed those years that I spent there and made some good connections.

Jacob Michaelson:

And then I did a postdoc at UCSD in psychiatric genetics and in the lab of Jonathan Sebat, who has done a lot of autism and other kinds of research in psychiatric conditions. And that was a great experience for me too. I made many more connections and we did some of the first whole genome sequencing studies in autism. And then that got me… I actually coauthored a paper with the person who was the department chair here at the time, Jimmy Potash, he’s now at Johns Hopkins. And he created a new division in the department. And it was a division of molecular psychiatry, which is a new thing. And he was recruiting for a faculty position where he wanted somebody with expertise in bioinformatics and other computational approaches and preferably somebody who had a little bit of experience in psychiatry. And so it ended up being a perfect fit for me and just the mission, and there was a lot of synergy with other faculty members and there was a really unique environment here at Iowa that made that decision to come here, a no brainer for me. So that’s how I ended up here in the department of psychiatry.

Emma Meador:

Awesome. Well, thank you so much for sharing. I think it’s so interesting how all of your different backgrounds were able to come together and you’re able to find your real passion and what you love. So that’s awesome. And then can you tell us a little bit about SPARK? What is the mission of this research initiative?

Jacob Michaelson:

Yeah, well, SPARK is the largest research community for autism. It has over 250,000 participants nationwide, and there are 31 clinical recruitment centers, and we are the Iowa site for SPARK. And I’m actually the principal investigator in Iowa for SPARK, which again is an interesting situation to be in given that I’m not a clinician and not a provider, but we do do a lot of autism research and we actually use the data. That’s the cool thing. If you look at SPARK and across those 31 sites, it’s not every site that actually uses the genomic data. A lot of these sites or the PIs are more on the clinical side of things than on the research side.

Jacob Michaelson:

And so the mission of SPARK or the goal is to improve the lives and the prospects of autistic individuals and their families through a better understanding of the basic biology of autism. And that also links up with just a better understanding of how the human brain works in general. And so I like to tell people, this isn’t an official line from SPARK, but I like to tell people that SPARK, its task is to make a map of autism because autism is incredibly heterogeneous. There’s a lot of diversity within autism. And so SPARK’s job is to help to chart a map of autism. And it’s through a combination of genetic research. So we’re making genetic maps of autism, but also, and probably more importantly, where we’re doing more to canvas that the lived experience of the wide variety of people who live with autism. And so in doing that, we learn more about the strengths and the challenges of autistic individuals.

Jacob Michaelson:

And so that’s really what SPARK is. It’s given that we now know, and we didn’t always know this, but we now know and appreciate the huge amount of diversity that exists within autism, SPARK is the effort by credible scientists and physicians to just map the basic biology of autism.

Alexis Clark:

So typically those institutions that are partnered through SPARK, you said, have a clinician running their SPARK wing or initiative, I guess, for lack of better words. What challenges does not being a clinician and leading this initiative in Iowa, what have they proven to be?

Jacob Michaelson:

Yeah. That’s a really good question because there are challenges because I don’t spend every day in the clinic, however, that is offset by, we have an incredibly formidable network of clinicians that I work with in the department that I have interactions with all the time. And then across the state, actually we have I’m on the board of directors for the autism society of Iowa. And that’s another forum that I have a lot of contacts in the community, I’m in the community a lot. And it’s those connections that I credit for our success in and recruit folks with autism into our study. And so it’s a little bit challenging because I don’t see people with autism every day clinically, but I think in the grand scheme of things, that’s really not that much of a detriment to our work.

Speaker 3:

Yeah. That makes sense. So with UIHC and the CDD being right next door, do they play in any partnerships with SPARK or your research studies?

Jacob Michaelson:

Yeah, absolutely. The providers at UIHC and specifically those in child psychiatry and at the CDD are some of our most valued partners. And I just want to say that in our recruitment efforts, when we keep track of the data, we see that when a provider endorses SPARK and provides information about the study to a patient, that person is far more likely to enroll than through any of our other recruitment methods. And so that’s why it’s really important to have clinicians in, on the whole process. And so beyond that, though, I would say that Dr. Lane Strathearn, who is the director of the CDD, and he together with a director of the Iowa Neuroscience Institute, Ted Abel, they recently were awarded a large center grant from NIH for research on intellectual and developmental disabilities, and that includes autism. And I’m one of the co-investigators on that grant. And so this is really exciting and, and it was recently featured in the newspaper, and it provides a further opportunity for research powered by the SPARK community.

Alexis Clark:

Absolutely. That is very exciting to hear. Congratulations to everybody involved with that.

Jacob Michaelson:

Thank you.

Alexis Clark:

So how prevalent is autism in the State of Iowa?

Jacob Michaelson:

Yeah, that’s a good question. So there are a number of estimates, but a recent survey from a couple of years ago, pegs the prevalence at probably around two to 3% in Iowa. And it depends on how you look at it, from some perspectives that’s actually quite common in something around 3% and other people might think, well, maybe that’s not that many, but it’s actually quite a common neurodevelopmental condition.

Emma Meador:

Yeah. Thank you for sharing that. That’s very interesting. I didn’t realize in my perspective, I think that’s very common and that’s a lot more than what I had anticipated. So yeah. Thank you for sharing that. And then what is one thing you have learned about autism that you wish more people knew or was better known?

Jacob Michaelson:

Yeah, that’s a good question. So I guess I would say that there are a lot of things that I wish people understood a little bit better about autism, but from my perspective as a genetics professional, is that a lot of people don’t realize that autism is strongly and really overwhelmingly genetic. And so the heritability for autism is about 80%, which is really on the high end. I mean, a lot of other conditions are less… Conditions, I guess, that are well known to have a strong genetic component, cancer and heart disease and stuff like that, they have heritability that’s not even close to that high. So it’s very genetic, I guess, is the big thing.

Jacob Michaelson:

The other thing that’s connected to that though, that I wish people understood a little bit more is that the gene variants that predispose to autism they’re distributed throughout the general population. So they’re not gene variants that only exist in people with autism, they’re there all throughout the general population. So from a certain perspective, most people have at least a dash of autism in their DNA. And it’s just that for some people, it reaches a level that it becomes clinically meaningful and they need additional assistance. They need additional help. And that feeds into, I guess, what I would say is the last point about understanding that it’s genetic, is that, well, I would optimistically say there’s been a lot of dialogue in recent years, especially over the past couple of decades about how to talk about autism, because for some people, and I would say most autistic people, autism is a big part of their identity. And so you can imagine that if autism is a big part of your identity, it comes across as a very negative thing.

Jacob Michaelson:

If other people are talking about autism as something that needs to be cured, or it’s a disease, or even more on the gradient of becoming more appropriate than calling it a disease, calling it a disorder or calling it a disability or these other things. And there’s a lot of nuance in how we need to talk about autism. But recently I’ve been thinking about it and I think probably at least for me, one of the best ways to talk about autism is to acknowledge that actually it’s a normal form of human variation that brings with it, and I think this is the key part because in saying this, I don’t want to minimize the challenges that autistic people face because they certainly face challenges, but I think if we can say it’s a normal form of human variation that brings with it very specific medical, mental health, and daily living challenges, then I think we can get to a more appropriate understanding of what autism is. And I would go even beyond that and state that in many cases, it also brings with it certain strengths too. And I think that’s a perspective that we probably need to see more of in the research that we do.

Jacob Michaelson:

So that’s what I would hope people would understand that it’s strongly genetic. That those genetic variations are out there everywhere in the general population doing different things and being concentrated to different degrees and different individuals and leading to different needs. But ultimately it’s a normal form of human variation that is going to require some specific medical and mental health and other support services. But I think a pretty fair way to understand autism and its strengths and also its challenges.

Emma Meador:

I love how you called it a normal form of human variation. I think a lot of people don’t really look at autism like that. And I think it’s a great way to describe it. And then I do have a question going off of how genetic autism is. Is there any epigenetics that play into this or any environmental factors or anything that increases the chance of someone having those markers turn on and develop autism or a more severe form of autism?

Jacob Michaelson:

Yeah, that’s a great question. And somebody that I think you should invite on your podcast to answer that question is Lane Strathearn, who is an expert in the epigenetics of autism. And he and I worked together a lot and it’s interesting because I’m the geneticist and he’s the epigeneticist. And so that sets up this people want to pit those two things against each other. And the reality is autism is not a hundred percent genetic. It’s really strongly genetic. And I guess, even when it’s strongly genetic, it’s still challenging to disentangle exactly what genes and what variants are playing a role, because it’s one thing to know that something’s genetic, it’s entirely another thing to know exactly what the genes are and exactly what the [inaudible 00:16:07] mechanisms are and how that all works. And we’ve still got a long way to go on that.

Jacob Michaelson:

I would say that the epigeneticist also have a pretty big challenge too, because they’re working with a smaller sliver of the overall risk for autism. And so they face many of the same kinds of challenges as they may understand that certain things are environmental risks, but then understanding exactly what the mechanism is, is incredibly challenged. So lots of studies have been done. Lots of even epidemiological studies have been done on environmental risk factors for autism. A very popular one is looking at the maternal fetal environment. So during pregnancy things that can happen or exposures that can happen, that can tweak risks one way or another for autism. But again, those are things that I’m not an expert in, and I’ve got a big enough challenge just focusing on the genetic aspect of it to understand. But yes, there are environmental contributions too. Probably just as I would say one of the more interesting ones that is both a genetic and an environmental factor that determines the course of risk for autism is your sex because autism is a very male biased condition.

Jacob Michaelson:

Many more males are diagnosed with autism than females. And we’ve even been able to show that if you take genetic risk factors that occur on the autozone, so not the sex chromosomes, the expression of those genetic risk factors can be very different depending on whether that happens from the perspective of a female developmental trajectory or a male developmental trajectory. So even if you had the same risk factors, they would unfold in very different ways, depending on whether it’s in the context of a male or female development. And so obviously biological sex is something that is genetic because of the sex chromosomes, but also it’s an environmental thing because of just being placed within the context of those developmental trajectories.

Alexis Clark:

Yeah. Thank you so much. I don’t have a very large background in epidemiology, so I always appreciate when Emma asks those epi specific questions. So jumping back to SPARK, what are some ways people can get involved in it?

Jacob Michaelson:

Yeah. Great question. So the first thing that people can do without enrolling, because we would love it if people would enroll, but you can just check out the website, which is SPARKforautism.org. And then specifically you can, if people are interested, if you have a professional diagnosis of autism or your child does, then you can enroll online @SPARKforautism.org/uiowa, and the whole enrollment process, your whole participation can happen entirely online. You can do it all from your home. So you would go to that website, you create an account and sort some questions. And within a week or two, a tube shows up in a package and you spit in that tube. And then you put it back in the box, you send it in the mail and that arrives at a lab that sequences your DNA. And I should also mention that there’s compensation that’s available for people who participate.

Jacob Michaelson:

So scientists like me and my team, then try to understand the link between the DNA variation that is discovered by sequencing the DNA that’s in that’s saliva that sent in and the variation and the lived experience of autism. That’s our main job is to understand how that variation in DNA relates to differences in the experience that people have. So for instance, we recently published a paper that found the first gene linked to an eating disorder in autism, and that eating disorders called ARFID, which is a form of a very restricted diet, and restricted eating patterns. It’s very common in autism. And we’ll soon submit a paper on sleep problems in autism. And we also have other ongoing research in creativity and autism in gender identity and gender diversity in autism and on language ability and autism. And so these studies were only possible with the generous participation of thousands of SPARK participants.

Jacob Michaelson:

So it’s important that potential participants know that their impact goes well beyond the initial enrollment. This is not a traditional study where you just enroll once and then you never think about it again, there are a lot of opportunities that come up down the road, and certainly people don’t have to participate in those, but those opportunities to participate in those follow-up research studies are made available to them. So SPARK would love to be, and this is one of its goals to be the resource for autism research for the next 20 years. We want to keep people engaged over the long term.

Jacob Michaelson:

And so what that means is that as the textbooks about autism, as they’re rewritten, because we’re learning more about autism all the time and we’re learning that our perspective on it needs to be updated. It’s those who have participated in research like SPARK, who will be represented in that updated understanding, which is why we want as many people as possible to participate. We want the conclusions that are drawn and the research that comes out of all of these efforts to be representative. And that can only happen if we get broad participation from the community. And so that’s one of our major goals with SPARK. So I guess that was a little bit meandering from your original question of how people can get involved in SPARK, but really it’s just taking that first step of going to the website SPARKforautism.org/uiowa. And the people can start the process there.

Emma Meador:

I think that is absolutely fascinating. And it sounds like you guys have been able to find so many very interesting results that are very specific to autism that I didn’t realize were. So that’s very cool research you guys are doing and going off of that. What do you think the most pressing issue is that you want to solve in your career that the public may not be informed about?

Jacob Michaelson:

Yeah, that’s a good question. And it’s a difficult question to answer too, because I think I wouldn’t come to my job every day if I didn’t think that everything that we were working on was important. And certainly I could go on for a long time about very interesting scientific questions that are very important. But when you say pressing issue, one thing comes to mind that is pressing and that a lot of people are working together to try and address these problems. So, and what I’m getting at is that we have some striking blind spots in science. So for instance, most of the data generated in biomedical research and especially in genetic research, disproportionately benefits people of European descent. And this is the major area of concern specifically in genetic risk prediction because our models work well when they’re applied to white people, but they’re not nearly as predictive when applied to people of other ancestral backgrounds.

Jacob Michaelson:

And so part of this is that there are fewer barriers and by barriers, I mean, historical barriers, cultural barriers, linguistic barriers, geographic barriers, there are fewer of those barriers to get white people to participate in research. Scientists simply don’t have to work as hard to recruit white people into research. And so it’s natural that if your sample was overrepresented for that group, that your models are going to be more predictive in that group. And so the predictive models end up being tuned largely to genetic variation that exists in white people. The other issue, and I think this is related to the first that I brought up, is that there needs to be greater represent representation of the Black community among scientists themselves, right? So there are a lot of great Black scientists, but we need more.

Jacob Michaelson:

And specifically, I think in genetic research need more Black scientists, and I don’t see many of our representation problems — so, and by that, I mean, representation in study participants — I don’t see those problems being solved until there are more Black scientists running those studies, because I don’t know that those issues can be addressed if it’s white scientists trying to get more representation from these underrepresented communities. We need more Black scientists. And when I say Black scientists, I mean you could equally extend this to Latin X communities and a lot of other underrepresented communities. I’m just using the Black community, because I think that’s a really good and very immediate example of these issues. And so how do we do that? I mean, first of all, I don’t think that there is any quick fix for this challenge, for science.

Jacob Michaelson:

Yeah, because I just think about, again, going back to Black scientists running these studies, they are going to be able to reach out to their communities with more credibility and authority than is possible for white scientist and muster in that context. And so to get to that point where we can see more representation among the scientists who are running the studies, I feel like that needs to start in high school at the latest and probably earlier than that. There needs to be more work in evangelizing science as a viable career path to Black students. And I think that we, as scientists, need to do our part to make sure that we clear that path so that those enticements and campaigns, and evangelizing that we do about science as a career, that those aren’t just empty promises and that they meet stiff headwinds on the way into that career.

Jacob Michaelson:

So the reality is that my sphere of influence here in Iowa is such that there are limits on what I can do on those fronts in terms of the representation and the study participants and among scientists too. But one more immediate thing that we can work on. And by the way, I should say that we are intent on doing our part on those other fronts too. But one of the things that we can have a more immediate impact on is improving the methodology and the models and the analytic methods. So that data from nonwhite participants isn’t simply set aside because it doesn’t work with the method that we’re used to, which unfortunately is the that’s par for the course right now. It’s simply easier to set that data aside and think about something else maybe to do with it, but to focus on that large majority group, because we know that our methods are going to work there.

Jacob Michaelson:

And I think that’s unacceptable. I think we owe it to everybody who participates in research to actually use that data that they sacrificed to make available to qualified scientists. So I think there is a lot that we can do on the methodological front so that we’re not simply stymied in the face of an ethnically diverse sample. And so that’s one of the things that we work on more immediately. But yeah, I guess just to sum up, that’s obviously not the only issue that I find important, but in terms of something that is pressing, I think that is a pressing issue because it translates directly into continued disparities and the societal benefits of biomedical research.

Alexis Clark:

So you did address how in the science community, you all can address this problem, but looking at the public health community as a whole, whether that’s in research or bio statistics or public health in general, do you have one piece of advice that you would give the incoming workforce?

Jacob Michaelson:

Well, I think it’s critical that… I mean, as I’ve been reflecting on the past several years, the past 10 or so years of my career, one thing that has changed a lot is that I’ve gotten more involved in the stakeholder communities and with the stakeholder communities that we do research on. And so I think it’s critical to know your stakeholder community, and it takes steps to interact with them and to be involved. And I think that’s the only way to appreciate the humanizing aspects of the science that we’re doing. And I think in particular, in epidemiological type studies where you’re dealing with vast numbers with just enormous numbers, it’s very easy for those numbers or for the human beings that represent those numbers to evaporate away in course of doing your statistical modeling and stuff like that. And so I think in formulating the questions that are to be asked and how they’re to be asked, and just more generally what the priorities are, I think it’s important for all scientists, regardless of what your background is or what your profession is. It’s important to be involved in communicating with stakeholder communities.

Emma Meador:

Well said. I think that is great advice. And then the last question we have for you is, what is one thing you thought you knew, but were later wrong about?

Jacob Michaelson:

That’s also a good question. So I think it’s related to that last one, because we did, and this is something that I only recently saw the light a little bit was that it’s tempting to think of stakeholder communities, especially if they’re different from you, it’s tempting to think of them in a caricature form or to reduce the complexity of the viewpoint of that. And what more specifically, I mean, by that, is that it’s tempting to say that community X believes Y. And so it gives the impression that there’s perfect consensus in that community about some issue, whether it’s pro or con, and we learned that in a very striking way when we recently did research on the attitudes of the LGBTQ community about genetic research at the intersection between mental health and sexual and gender identity.

Jacob Michaelson:

And we went into that with a fair bit of trepidation, just assuming that the stakeholder community for a lot of good reasons, would be very against this research. That was our hypothesis just going into it. But it was fascinating to see… And in hindsight we should have known this, but people are complex and we all wear a large number of different identities. And picking any single of those identities, that’s not all of who we are. And so I think it’s just important that we not reduce the complexity of those groups that we do research with, because we found that actually there were a lot of people who were stakeholders, who were very in favor of this research, and it was only a minority that were against. And there were other predictors that we hadn’t contemplated that were far more predictive of a person’s attitude on this research than whether or not they were a stakeholder in this community.

Jacob Michaelson:

And that was very interesting to me. And that’s a lesson that has stuck with me, and I now apply to every other group that we work with, whether it’s the autistic community or every other community. That if you think that there is a consensus among a community, then your sample is not representative enough because the fact is there’s almost never perfect consensus, and there is a variety of viewpoints. And if you’re getting a false feeling of consensus, then it’s probably time to think about who you’re missing in your sample.

Alexis Clark:

That’s an excellent point. Thank you so much, Dr. Michelson, for coming on today with Emma and I. We greatly appreciate it. I have learned a great amount and I’m sure everyone out there listening has learned a great amount.

Jacob Michaelson:

Great. Well, thank you so much for inviting me. It’s been really enjoyable.

Alexis Clark:

Yes. And we will post in the description of today’s episode, the information about how you can learn more about SPARK and get involved with that. So if you are interested in that, check out the description of today’s episode. That’s it for episode this week. Big thanks to Dr. Michelson for coming on with us today. This episode was hosted, written, edited, and produced by both Emma Meador and Alexis Clark. You can learn more about the university of Iowa College of Public Health on Facebook. Our podcast is available on Spotify, apple, podcasts, and SoundCloud. If you enjoyed this episode and you’d like to help support the podcast, please share it with your colleagues. Our team can be reached at cph-gradambassador@uiowa.edu. This episode was brought to you by the University of Iowa College of Public Health. Stay happy, stay healthy and keep learning.