The Bioinformatics CRO Podcast

Transcript of Episode 4: Jason Stein

Grant: Welcome to The Bioinformatics CRO Podcast. I’m Grant Belgard and joining me today is Jason Stein. Jason, would you like to introduce yourself?

Jason: Sure. Hey Grant, my name is Jason Stein. I’m an assistant professor at UNC Chapel Hill, in the department of genetics and in the neuroscience center. And I’ve been here for just a little under five years.

Grant: Fantastic. So tenure review is coming up soon?

Jason: It is. 

Grant: Good times.

Jason: I submitted my materials in February. 

Grant: Fantastic. So, What’s your research program about? 

Jason: My research is to try to understand how genetic variation that’s present in human populations, influences brain developments and brain structure, and then leads to risk for neuropsychiatric disorders like autism and schizophrenia.

Jason: So we have several model systems that we use to study this. The major thing that we do is using human neural stem cells. So we have a population of human neural stem cells, each of which is genetically diverse. And then we try to see how genetic variation within our population is associated with differences in the development and the differentiation of these neural progenitor cells.

Grant: What have you found? 

Jason: Oh, so what have we found in that? So we recently put some pre-prints up in the bio archive (“Evaluating brain structure traits as endophenotypes using polygenicity and discoverability”, “Cell-type specific effects of genetic variation on chromatin accessibility during human neuronal differentiation”). 

Grant: And Grace can link those on the transcript. 

Jason: Oh, nice. Yeah. Link them up. Get that–what’s that rating system–the altimetric. Yeah. You got to get that altimetric up.

Jason: So what have we found? So we found that, there’s many sites in the genome where genetic variation influences chromatin accessibility and gene expression. So chromatin accessibility is a measure of the function in the non-coding genome, probably largely due to differences in transcription factor binding.

Jason: So the genome is more open and accessible in certain regions of the non-coding genome, and that allows transcription factors to bind. So genetic variation within those open regions can impact transcription factor binding, and then lead to differences in chromatin accessibility. We found, you know, many thousands of different sites where genetic variation affects chromatin accessibility. And interestingly, they’re very cell type specific. 

Jason: So we did, we did the study in two different cell types: progenitors and their differentiated neuronal progeny. And if you do a chromatin accessibility QTL–so you find genetic variation affecting chromatin accessibility–very few are shared between neurons and progenitors.

Jason: But if you look at expression, like eQTL, there are much more that are shared between, genetic variation, affecting gene expression between two cell types. So we thought that was really interesting. You may have more cell type specificity for chromatin accessibility than you do for gene expression.

Grant: That is weird. Why do you think that is? 

Jason: So, you don’t totally know. So these are just hypotheses. 

Grant: But speculation is good. That’s why you go on a podcast. You can speculate. 

Jason: You can say whatever you want and no one cares, right? 

Grant: No reviewers will reject this. 

Jason: So this is inspired a lot by the work of Daniel Gaffney. He had a paper in Nature, Genetics, and the basic hypothesis for this paper was–and I think was well demonstrated in this paper– that if you have genetic variation, it can impact chromatin accessibility cause it’ll impact like different transcription factors binding to a certain region. But only in the presence of other sort of helpful transcription factors are you actually getting an effect on transcription. 

Jason: So say in one cell type that that other transcription factor is not expressed, well then you can still have an effect on chromatin accessibility, but you wouldn’t have an effect on gene expression. So this stimulus dependence of eQTLs may not be there for caQTLs or maybe less so for caQTLs.

Jason: And that’s kind of a hypothesis that we’re going with. We haven’t really demonstrated it though. Cause you kind of have to do a lot of ChIPseq to demonstrate that, and that’s expensive and very hard as we’re trying to find out. Yeah. 

Grant: So what do you envision as the ultimate practical application of this broad line of work?

Jason: Yeah. Yeah. So I was on a paper with a colleague just down the hallway, Mark Zylka, and this paper was focusing on Angelman’s syndrome and Angelman’s syndrome is like one variant or a variant of large effect mutation, which creates a big change in development and behavioral changes.

Jason: And through Mark’s work for like 10 years, and other people’s work too, they found a molecular mechanism whereby you actually get decreased expression of UBE3A in the paternal chromosome. I’m not the best person to explain this, but I’m on a podcast, so I’m just going to go for it–the basis of Angelman’s syndrome is that usually it’s an imprinted locus in UBE3A in neurons. You usually get expression in just the maternal allele in neurons, but if you have a mutation in the maternal allele in neurons, then you have no expression of UBE3A in neurons.

Jason: Okay. So that’s bad. And that leads to Angelman syndrome. So why is the paternal allele not expressed? Well, his work over the last 10 years has found this molecular mechanism. So they found this long non-coding RNA that seems to silence the paternal allele. So his hypothesis: let’s turn on the paternal allele. So they did this CRISPR screen and found a region of the genome that you can target that decreases that lncRNA, that long non-coding RNA, and increases the expression of the paternal allele. So now you have some expression of UBE3A. 

Jason: Okay, cool. So understanding the molecular mechanism took 10 years, like a long time. So you went from mutation to understanding molecular mechanisms and now with this CRISPR design, they have an actual treatment. And it works in mice and human cells. It hasn’t worked in humans, right? So it’s not like ready to go, but you know, it’s getting there. 

Jason: And so my kind of feeling on this is the same way. If we understand how genetic variation creates risk for psychiatric diseases, then we can begin to say, okay, so we know the genetic variation–and there’s a lot of consortia that are doing this, like the Psychiatric Genomics Consortium–they’re creating risk for psychiatric diseases. Then, if we can understand what the mechanism is–that’s where we see these QTL type of papers come in– then we can sort of start to develop treatments for the diseases. 

Jason: Now it’s different from Mark’s work to the work that we’re doing, because that’s one variation of a very large effect. And the stuff that we’re working on is polygenic effects. Each of which are very small. But still, I have hope that if we can target some of these pathways, maybe multiple of these pathways, that it can lead to some alleviation of symptoms. So that’s kind of like what I envision obviously hasn’t been done yet.

Jason: So just because I envisioned it doesn’t mean that it’s going to happen, but that sort of pathway from finding a genetic variation, understanding the mechanism, developing a treatment is like something I’m hopeful for. 

Grant: Nice. What do you think is the most exciting area of work in biomedical research today? What do you think’s the most promising? 

Jason: Oh that’s a good question. I think this tech development stuff in the biomedical world is really exciting and tech development, not in a computer way, but in a biological way, like creating biological systems to solve scientific problems.

Jason: One of them being gene therapy approaches. If you can actually make a virus that contains gene editing proteins that target your gene of interest and can have some functional effects. That’s awesome. That’s really amazing. I think some other work too, like Jay Shendure. It doesn’t have an immediate practical outcome, but work that they’re doing is trying to make mutations or make recordings of cells as they perform some biological process.

Jason: So for example, they’re taking a cell as an embryo and then making a mutation sort of every time it divides. And then every time that divides, then you can create a linear trajectory of how one cell forms many other progenitor cells, which make many other progenitor cells, which then lead to the formation of an entire organism.

Jason: If you can take that same sort of idea and then move it to measuring each time a cell does something. So for example, fires an action potential, like George Church proposed. He calls it a ticker tape recording of action potentials, or you can record gene expression through development.

Jason: You’re developing technology to perform longitudinal recordings of biological processes. I think this is going to be amazing at the single cell resolution. That’ll be really cool because once you have a ticker tape, you can sort of fast forward that ticker tape and move cells quicker through a biological process or develop cells quicker, which I think will be really important. 

Grant: Cool. So Grace here is a UNC senior, just getting kind of started with her career if you were in her position, deciding what area of work and laboratory and so on to start in 2021 or 2022. What would you do? 

Jason: Yeah, well, I guess can I ask Grace first what are you interested in? What do you like doing? Because that’s kind of the most important thing. 

Grace: Yeah. I work in the lab of Dr. Ian Carroll and I work with the microbiome, which I’ve been obsessed with since probably junior year of high school. Yeah. I definitely do that in the future.

Jason: Okay. We, I mean, that’s the most important thing. Do the work that you’re passionate about because if you’re going to go into graduate school or you’re going to start being maybe a technician after undergrad–that’s what I did–then go to graduate school and then go to postdoc and walk your way along the academic path. It gets super frustrating at certain moments because science is difficult and the experiments don’t necessarily turn out as you expect them to. And so what gets you through that is like actually caring about what you’re doing.

Jason: The most important thing is that you really care and that in those sort of low lows that you’re going to have, no matter what field you go into, that you’re going to be like, “you know, I’m doing this for a reason. This sucks, but like, I’m just going to keep working at it.”

Jason: I think that’s kind of  the most important thing. Obviously, I think towards Grant’s question, he wants like, what do I think are important skills and things like that. I think having bioinformatics skills, especially for microbiome work is going to like make your life so much better and easier because instead of just doing an experiment and handing it off to somebody else, which I think is what Grant’s organization allows people to do, it allows you to do both of those things.

Jason: And having the ability to do both of those simultaneously really gives you a lot of skill sets. That will be super valuable because you’ll know because you did the experiment that like “I isolated this sample, and something was weird with this one. I don’t know if the concentration was low, but the quality control metrics didn’t look right.”

Jason: And then you see it in the data and you’re like, “okay this isn’t right. This is an outlier. I know why this is an outlier.” So you can throw that out. But if you’re only doing one of those things–like you’re only doing experiments or the bioinformatics–it becomes more difficult because then you’ve got to go back to the guy who did the experiment and go be like, “do you remember anything weird about the sample?” And they may or may not remember.

Grant: That’s a good segue to our next topic. Can you maybe tell us a bit about your path? Add some color. You know, start in Dayton, Ohio. What motivated you? What surprises did you find at different stages of your career? What do you maybe wish you had done differently? 

Jason: So, yeah, I was born and raised in Dayton, Ohio, the birthplace of aviation. Where the Grant Belgard–if you didn’t know for all those podcast listeners the T stands for The Grant Belgard–was also stationed for a little while. I went to high school at a public school. There were some pretty good science teachers that I had when I was in high school. Mr. Protrusio and Mr. Martin. If they’re listening, shout out. 

Grant: You could send them the transcript after.

Jason: If I can even remember how to spell Mr. Protrusio’s name, which I a hundred percent do not remember. They were good. They were inspirational and helped guide me along the path. 

Jason: When I went into college, I went to Northwestern. I studied at this program called the Integrated Science Program. And this was kind of an amazing place. Like we were somewhat segregated from the rest of the Northwestern population, which was kind of weird, but actually kind of good because we had our own little community. So we had the integrated science program, the ISP, house which was this little, somewhat crappy house.

Grant: Sounds like a real party house.

Jason: There were parties there. Yeah. But you know, people would study there, they would hang out there and we’d have parties there. We would have our classes there. So it was like your little community. It was only about 20 people or something like that. Coming from Dayton, we had very smart people that were at the high school, but like, there were some, really brilliant people in this ISP program.

Jason: And it was cool to just be around these people, interact with these people and have the attention of the teachers too, who were teaching these very small classes to us. So I liked it a lot and enjoyed the classes. 

Jason: One of the most useful things was they had this–I forget the name of the class–but they taught us basic computer programming skills like Unix, some Pearl. I don’t think R was even a big thing back then. So we didn’t learn R, but like a little bit of Python and things. And that was very useful because that has helped me in the future so much. I had a little bit of computational skills. 

Jason: So when I was there, I did research. But I did research in physics. So I worked at the Fermilab, which my friends called Fermi camp, which was a very weird, but kind of amazing place. So if you ever been, it’s outside of Chicago in Batavia, Illinois, and they have these roaming Buffalo that are just around on this large cut area and underneath there’s a particle accelerator, which was then the largest particle accelerator in the world before the large hadron collider. And they shot particles at each other and then they got a whole bunch of data and they needed some way to visualize that data. 

Jason: And to be honest, I had no idea what I was doing, but they were like, “can you program a website to make all these graphs?” And I was like, “cool, sure.” So I worked on that and I didn’t probably do a great job, but they seemed to want to keep me around. So that was good. That was really good. 

Jason: So then after that, I didn’t know what I wanted to do. I did all kinds of stuff. I took the LSAT, the GRE, the MCAT, because I didn’t know.

Grant: You were really undecided.

Jason: I was very undecided. And then, I ended up applying to a position at NIH and working in the intramural program in Bethesda for two years under Andreas Meyer-Lindenberg. And we would scan people with schizophrenia, like do MRI scans of people with schizophrenia and their siblings, and then analyze the data as well.

Grant: Was it on the main campus?

Jason: Yeah. It was on the main campus, building 10. Yeah and it was a great experience. There was again, a really good environment there. Like there were a whole bunch of people, right out of college that were interested in science, good people, friends and stuff. They were all working there at the same time. We were called IRTAs, which is intramural research training awards. Yeah. Weren’t you in IRTA, Grant? 

Grant: I was in the GPP, so in grad school I was flying back and forth between Oxford and NIH, but I was at an offsite location around Twin Brook, where the intramural sequencing facility was.

Jason: Oh, cool. 

Grant: So it was kind of a cutting edge of the genome technology branch. 

Jason: Oh, nice. Very cool. Twin Brooks, so that’s not too far, but it’s like just not on the main campus. 

Grant: Yeah, it’s on the red line. So it was close enough. 

Jason: Nice. 

Grant: You didn’t have to fight as bad traffic. 

Jason: What year did we overlap? What years were you there? 

Grant: 2008 through 2012, but not really full time, except 2010 through 2012, but I would pop back and forth for like a week at a time. 

Jason: Okay, cool. We probably did overlap just like a tiny bit then. Okay. Yeah. So I scanned people with psychiatric disorders. That was really interesting. My computational skills were super valuable to those people. They found them valuable. 

Jason: You know, one thing I really hated was calling subjects. Like a big part of our job was recruitment, which means that you get on the phone and you say, “Hi, my name’s Jason. I’m calling from the NIH.”

Grant: Did you start robocalls spamming them? 

Jason: I didn’t, I didn’t make that. I should have made that. It would have made everyone’s life easier. So we had those assignments, recruitment assignments, scanning assignments, and then we would also do analysis, and I wrote scripts to make everyone else’s analysis easier. And then I was like, “if I do this, will you call my subjects?” And they’re like, “sure.” So that was great because I really hated doing that.

Grant: Motivation.

Jason: Yeah. So I worked for Andreas Meyer-Lindenberg, who’s now in Germany, but then was at the NIMH and he told me that I should work with Paul Thompson at UCLA for graduate school. And so basically, I applied to several places for graduate school, but I emailed Paul beforehand and I was like, “Hey, Andreas told me I should work with you. Can I work with you?” And then he called me and basically said, “sure, you can work with me.”

Grant: So, weather it had nothing to do with it. 

Jason: No, actually I was pretty anti-LA the beginning. You know, LA has a bad rap. 

Grant: You’re like the anti Neil.

Jason: Neil is totally the opposite. I think growing up in West Virginia, Neil must have really hated it.

Jason: So yeah, I was not into LA. And I was like, “Oh, this place is lame, but you know, whatever. The neuroscience is supposed to be good.” But I eventually grew to really appreciate it. There’s a lot of good things there, especially hiking. Hiking is pretty amazing there. And so close to the city.

Grant: And did the weather grow on you? I mean, would you have considered taking a position in Chicago afterwards? 

Jason: Oh, a hundred percent. Yeah. I like Chicago, actually. I don’t mind the Chicago weather and I very much appreciate changes in seasons. Like here in North Carolina, it’s beautiful. Like I love the springtime, which you don’t get. Grace is agreeing with me. Springtime here is amazing. Like you just get blooming of everything after the winter. It’s just such a contrast. So beautiful. So nice. 

Grant: Yeah. Florida is just one big, hot, hurricane season. 

Jason: It’s very muggy down there. Yeah. So I worked with Paul Thompson. 

Grant: Tell us a little bit about Paul Thompson. 

Jason: Paul Thomas is an interesting human. He is a very brilliant man, very, very smart in mainly math and figuring out topologies of brain structures. He really led. He did really an amazing job at that and he is highly motivated to publish papers and you can see from his publication record. I think he’s in the thousands of papers (1250 Publications). So that was interesting. 

Jason: His lab environment was also kind of weird in the sense that we didn’t have lab meetings. There was no room. He was getting grants like crazy and recruiting many students and post-docs, and we were all shoved into his office.

Jason: There were like eight people maybe in his office. And I remember I was sitting on one side of a desk. And then on the other side of a desk, there was somebody else, like just facing you. Now that you think with COVID protocols, you’re just like breathing into the face of someone else. It was just crazy.

Jason: It was a good experience because Paul had access to so much data, and he really needed people that were willing to analyze the data. So he was really looking for people to analyze that data. And this was also at the time that GWAS was starting. I started grad school, I think in 2007, and 2007 was one of the first welcomes to GWAS.

Jason: There were a lot of candidate gene studies, which are, for those who don’t know, this sort of old school thing you don’t do now, versions of association studies. 

Grant: All BS. 

Jason: All BS. None of it has ever replicated. I would say that. The need for doing GWAS and large consortia was clearly there, but those didn’t really exist yet for brain structure traits like for things you can measure with MRI. That’s basically what I got involved in. So, Paul met with this guy, Nick Martin from Australia, who was one of our collaborators, over dinner. They basically said like, “Hey, we should form a consortium.” And then I was basically the one to lead that consortium, which eventually we called the Enigma consortium: Enhancing neuroimaging genetics through meta analysis. 

Grant: Who came up with the acronym? 

Jason: Paul came up with the acronym, which is a great acronym, but also the wrong side. You know, like if you think about it, like he keeps saying, it’s like the code breaking for the brain, which is great. I kind of like that, but it’s the wrong site. It’s the Nazi side that came up with the name. Like we should call it like Bletchley or something, something more positive, but unfortunately it’s Enigma. I mean, nobody remembers that. So I think it’s kind of great. 

Grant: Nobody knows history. 

Jason: Nobody knows history. It’s a great name. So yeah, I ended up leading this consortium and along with people from Australia, Sarah Medland, people from Holland, and they would all fly to LA and we would do all of this stuff together. It was a good experience because we were one of two consortia that were doing this at the time. We found a bunch of genetic associations to bring structure.

Grant: And then?

Jason: And then, I finished with Paul, although I didn’t really finish. I was still working with Paul on all these Enigma projects, but I was looking for a postdoc. I was trying to find a postdoc that I could not just find genetic variants associated with different traits, but what to do next.

Jason: And so I read some papers online about what to do next. And it seemed like things were converging towards using STEM cells to model variance. And I was like, “okay, cool. I want to use STEM cells to model variance. I have zero experience in wet lab biology. I don’t know how to hold a pipette. I don’t know how to run a Western blot. How can I do this? I need somebody who values computational experience and will allow me in a postdoc to like transition a little bit and learn some new things.” And so I emailed Dr. Daniel Geshwind.

Grant: I’m trying to  get him on. He said he had to listen to a few first

Jason: Did he? That’s funny. So yeah, maybe he’ll listen to this. So he wrote a Nature paper with Jenna Konopka, basically about hypothesis discovery research. Instead of all the classic wet lab scientists who were poo-pooing like, “Oh, you’re just on a fishing expedition. All bioinformatics work is a fishing expedition,” which was pretty prominent back in 2011. That’s what they would say. 

Jason: He was like, “Okay, there is hypothesis-generating research and that’s what we do. And we generate new hypotheses. And then we can validate those hypotheses, test them with model systems.” And I was like, “Cool. That makes sense. You’re not, poo-pooing all of the giant amount of work that all of this discovery science is doing.”

Jason: So then I tried to apply to his lab and I emailed him once. No response. I emailed him a second time. Maybe I got a response. I don’t really remember.

Grant: Three letters

Jason: Yeah. It never worked. Then a third time I was like, “Dan do you want to meet me or not” And then he’s like, “Oh yeah that’s fine.” And then we talked and I think this was when he was in London, maybe at the Institute of psychiatry. And so it was very difficult to get an appointment with him. But eventually I got an appointment with him. He allowed me to be a postdoc and yeah. That is where I met the Grant Belgard. 

Grant: So tell us about Dan.

Jason: I don’t know. What do you want to know about Dan? He’s a very brilliant man. He has a million projects running simultaneously. Somehow he knows and can provide useful insight to each of those different projects. He also promised me that I could meet Bob Dylan if I got a paper in a fancy journal and that has not happened. His neighbor is Bob Dylan. So I’m still waiting for that. And I hope that happens soon. 

Grant: You’re on the spot, Dan.

Jason: Yeah, I would really like to meet Bob Dylan. I think that that would be–other than meeting Grant–really a defining feature of my life.

Grant: So, what did you do in Dan’s lab? 

Jason: So in Dan’s lab, I worked with this other postdoc named Luis. Who’s now an assistant professor at UCLA. Luis and I formed a team, which we call team middle earth because we had the middle bay and we’re both nerds. And we basically did everything together. So Luis is a very brilliant molecular and cell biologist. He was trained at Harvard. He’s the most careful scientist you will ever meet. And he knew nothing about bioinformatics and was very curious about bioinformatics.  And I knew nothing about wet lab biology and he knew everything. So he taught me everything I know about wet lab biology. And then I taught him a little bit about bioinformatics and how to code. And I think mainly like how to interpret what the possible confounds are for bioinformatics experiments and stuff. 

Jason: That combination was amazing. I can’t speak for him, but it really helped me out. Like really helped me develop into a much, much, much better scientist because now I have more skillsets. I’m not just analyzing other people’s data. We can generate our own data in the lab. That was really great. 

Jason: So Luis and I worked on developing human brain tissue. We acquired that. And then we studied multiple aspects of it: how well STEM cells model the actual development of the in vivo brain. And then we also studied how chromatin accessibility changes during neural development and the developing human brain.

Grant: We use those data sets. I feel like everyone does.

Jason: You do use those data sets? Oh, nice. I’m glad. I’m glad to hear that. That’s cool. 

Jason: Luis is great. You should have Luis on. I like Luis’s insights into anything. Anytime I try to do something new, I always ask Luis. I’m like, do you think this is a good idea? If he thinks it’s a good idea, then I do it. If he thinks it’s a bad idea. Then it’s usually a bad idea.

Grant: Cool. So, what’s the wet versus dry lab balance in your lab now? 

Jason: So I’d say it’s pretty 50:50, and leading a little more wet now. So it was leaning a little more dry before. The initial experiments in our lab were growing these hundred different stem cell lines that Luis and I generated in Dan’s lab. And then, I shipped here to UNC and then we ran these QTL studies here and now we have a lot of data.

Jason: So we have a lot of hypotheses because you can get co-localizations between your eQTL and caQTL with GWAS data. Okay. So now we have both the system for discovery and an experimentally modifiable system to see what the effects are and why those effects exist. So now we’re in that stage where we analyzed a whole lot of data, and then we have all these different experimental hypotheses. So now we need to do all the validation for those experimental hypotheses. 

Jason: Yeah. So I’ve been really fortunate to get students who are interested in both sides that were willing to come with me, even though I’m not the best wet lab biologist in the world, but I still have decent resources.

Grant: They don’t know any better. 

Jason: They don’t know any better. That is the definite truth. 

Grant: Yeah. I think back to when I was a student, I mean, I had no clue what I was doing. It was just pure luck. 

Jason: So yeah, I’m definitely taking advantage of that, you know? The main thing is being nice to people. I think if you’re nice to people and you try to be a good mentor, then word gets around that like, “Hey, this guy’s not a jerk. He cares about science.” And so other people hear that from the other grad students and they were like, “Oh, maybe I want to work with this guy who is not a jerk.”

Jason: I think we’re like 60, 40 now. I mean, it’s still pretty heavy for dry lab stuff, but. A lot of, even the bioinformatics students are doing wet lab experiments now to try to validate their hypotheses. 

Grant: Nice. So tell us about getting started. What were your biggest surprises? Obviously you would have been as well prepared as anyone going into it, but what weren’t you expecting?

Jason: Yeah. I mean, I think how long it takes to do anything is like a big surprise and kind of a disappointment. Also how lonely it is at the beginning. First of all, imposter syndrome is overwhelming. When you come in and you have an empty lab that’s a little bit dirty and you have your office space and you have nothing. No people, no experiments going, no data. It’s just you. And it’s like, awful. I think when you just start out that first week.  The process of building the lab, getting equipment, getting people takes a very, very long time, especially because you want to hire the right people. You want to make that lab environment good, so people actually enjoy working there.

Jason: You want to have very competent people who have skill sets that are complementary and not identical to your skill sets. And the major recruitment that you get is graduate students who have a defined schedule of being able to join your lab. So you can only get graduate students on the rotation schedule and then they join your lab.

Jason: So all of that leads to huge delays in the ability to make a functional lab with enough people and enough equipment. And enough, whatever the experiments are to do anything. So you can have all these ideas, but you can’t do anything at the beginning. So that was like a six month process.

Jason: And, you know, the people here were really nice and they also explained this to me. They’re like, “Oh, it took me six months to run my first gel” or something like that. And I was like, “Oh, thank goodness.” That makes you feel much better when you hear that. 

Jason: I mean, that took a while, especially the first project. We had this QTL project, which was my R00 project. It took a long time to get running, but it finished and was running and I had a great technician to help me out with that. Now we have all this data that allowed me to get more grants.

Jason: And now with the students and stuff, I feel like I don’t have to be here. I mean, I still have to be here, but they’re rolling. And they’re smarter than me and they can just do it. You know what I mean? I’m just providing advice. It’s not quite at the point where Dan’s lab is: like a super mature lab where he has like many postdocs who have really, really strong experience. Like I have a very grad-student-heavy lab. People still need to be trained and stuff like that, but it is much more to that point than it was five years ago, which feels good. That feels good. 

Grant: So going back to that, that startup phase, would you ever consider starting a company? 

Jason: You know, I’d consider it, but I don’t know what I have to offer yet, The Grant Belgard. I feel like my end goal would be wanting to make some sort of therapeutic. That would be an ideal for me. And I don’t have that yet. Like my research hasn’t led to that yet. So if I ever do feel like I have something to offer that I can sell that other people would want to buy that I think can help the world. Then yes, I want to do that. But right now I don’t have that. I don’t have that, that thing. 

Grant: Yeah. It’s interesting. These days the virtual biotech model is in Vogue. So. It’s quite common for a very small company. One, two, three, four people to develop an asset and essentially get to the stage where it’s ready for later stage clinical trials and have pretty much all the work for it outsourced. And so really the core team is finding organizations for that, interpreting the data, raising money and so on, as the asset progresses. 

Jason: That’s interesting. 

Grant: And there even organizations today that do that with pretty complex biologics. So, you know, gene therapies and things like this. So it could happen one day, maybe Jason Stein. 

Jason: Yeah. I mean, yeah maybe, I don’t know. That sort of idea is quite different from the way academia works, where you farm out the experiments to other people. So in that case, you’re mainly doing either bioinformatics or just even interpretation of the experimental outcomes? Like how does that work?

Grant: Interpretation, deciding what to do next, there’s a lot of coordination that has to happen and quickly and it’s very interdisciplinary, right? So you’re working, not even primarily with bioinformaticians. You’re working with chemists with biologists, with biostatisticians, with clinicians. You have to draw on a lot of different skill sets. 

Jason: Yeah. That’s cool. For me, first I don’t know anything about business. Like you know a lot about business. I don’t really. I don’t know how to do that, but I feel like the essential thing of a business is that you need to have something that you’re going to sell to other people that they want to buy.

Jason: And so like right now, I feel like I have research. I have research ideas. I have the ability to do research. I have people and the resources to do right now. I hope that that research leads me to something that someone else wants to buy and that would be helpful to the world. But immediately, I don’t see it. I don’t have that right now. 

Grant: You could always spin something out. 

Jason: Yeah, maybe. For example, I talked about the Angelman’s syndrome gene therapy treatment. That’s a thing where it took 10 years. He has something that’s a gene therapy in mouse and human cell lines. I can definitely see that moving towards a company or even farming it out to other people to do experiments because there’s a very clear path. I participated in that, but I’m not like leading. So, for my own research, I don’t really have that yet.  I hope one day to get it. And if I do, it’d be great to either form a company or work with somebody who does. 

Grant: Sweet. So 30 or 40 years from now, what would you like to have done? 

Jason: Like in my career? 

Grant: Yeah or actually more broadly.

Jason: I think about this a decent amount. So I think first it just to be remembered as a decent human to other humans. I feel like, especially now with political situations, that simple kindergarten thing is totally forgotten and is totally the most critical thing for a functioning society. And just the most basic thing is to treat other people with respect. 

Jason: I think in terms of career aspects, if I can, I would like to be a part of developing treatments for psychiatric disorders, which right now don’t have treatments like schizophrenia. I mean, they have treatments, but they’re not good. And a lot of people don’t do well. And there’s nothing really for these people. Like they’re homeless. They’re in state hospitals. It’s a huge burden on society and people are just suffering. But there’s so much good neuroscience that’s happening right now. So much good genetics that I feel optimistic about what could happen for these people. And so if I can help contribute to that, I feel like that would, that would be amazing, like an amazing cherry on top of being a nice person and my career, 

Grant: Very nice. How do you like UNC?

Jason: UNC is a good place. A lot of, a lot of nice people, not a lot of like ego. I mean, there’s still some ego. Science has that for sure. But like people work together really well. Everybody’s been really supportive from a young assistant professor kind of thing, so that’s been great. There’s some administrative things and things that definitely could be corrected or made better, but I don’t really want to be the administrator to do that. And usually the way they have it here and I’m sure in many places is the squeaky wheel gets voluntold to fix the problem. So you’re limited in your squeaks based on how much time you want to spend fixing the problem, which kind of makes sense. 

Grant: Well, you know you’re getting someone who’s passionate about it. 

Jason: Right. Yeah, exactly. Exactly. We only have time to do so much. And if you want to spend your time doing research, then you spend your time during that research and find alternative solutions to problems.

Grant: The ego thing in science is pretty interesting. It’s been pervasive, from even quite early days, back to Newton and Hooke. Ego has always been a bit of a problem. For a lot of people they are quite ego motivated more than anything else, but not everyone. I mean, some people are just nerds and like doing it right. I don’t know. What was your sense of that? I would say probably more people are just nerds in like doing it, but ego thing is pretty common. I don’t know. 

Jason: Yeah. Yeah. It’s hard. I can’t say I’m immune to it. 

Grant: I feel like it’s enriched in academia. No offense. 

Jason: I think you’re absolutely right. There was this paper written–it wasn’t a paper more of a website–where somebody proposed a new model for academia. Basically how it is now, I’m supposed to form my own very small business where I get people to work in my lab. I get grants from the federal government or foundations to support my work. And like I had my own business and Hyejung or whoever’s next door has her own business. Mark Zilkha has his own business. 

Jason: There is not this overarching effort, as is seen in physics, for example, where we all work on the big problem and we all do our individual small part for that. And some organizations, like the Allen Institute, I think have done pretty well. They work on giant problems that cost crazy amounts of money and they have each little person contributing to that.

Grant: And then on the other hand, you have the human brain project. Right? 

Jason: Right. There’s definitely examples where it didn’t work, but the ideal situation for me would be like, I have Luis and we work really well together. If Luis and I could form our own lab where we’re just two people running a lab, that’d be cool. But universities don’t often do that. They don’t recruit two people together. It’s like, why? Why not? We worked so well together.

Grant: Bring middle earth to UNC.

Jason: Yeah. We have to get Selene to want to move to North Carolina. Yeah. I think those things would be really cool, like to try to make something like that or to have very large projects. I know the human brain project didn’t work, but if you have very large projects where you actually have a very clear definable goal and steps that you want to get there. Where everybody says, “I’m not going to try to be the one who discovered it. I’m just going to do my part of the bigger entity.”

Grant: That’s basically, what academics refer to as “industry.”

Jason: Is it? I’m not sure I believe that.

Grant: Yeah, you know, we are in the same boat rowing towards the same goal. At its best. 

Jason: Yeah at its best okay. Because I feel like industry is dominated by the pursuit to make money and increase stockholder wealth right?

Grant: If you don’t make anything that’s useful, then you’re not going to make any money. 

Jason: Right. But that’s the essential problem. With research we don’t necessarily know if we’re going to make anything. You have to figure out what’s wrong first. And then you can make something. With psychiatric illnesses you have to still figure out what’s wrong first. And then you make something. There’s been a lot of drug development where you don’t know what’s wrong. Just throw a lot of stuff at it. And then you, you see minor therapeutic advantages.

Grant: Well, a lot of big drugs have just been discovered through pure serendipity with no known mechanism of action. Right? I mean, a lot of the big discoveries of the fifties and sixties, that in many cases took a very long time to improve on

Jason: Yeah. But clearly improvements are needed and there hasn’t been much in a long, long time. Especially in psychiatry. Yeah. 

Grant: Yeah. A hundred percent. Well, we are at about time and we just need to upload. 

Jason: Okay, cool. 

Grant: Thank you so much for joining us today, Jason. That was very enjoyable. 

Jason: Absolutely. Thanks for having me on the second edition? Is that right?  

Grant: I think this is episode four episode four.

Jason: Well anyways, thanks for having me. I really appreciate it, Grant. Hopefully I didn’t talk your ear off. 

Grant: No, it was great. Thanks, Jason.