The Bioinformatics CRO Podcast

Episode 14 with Dan Geschwind

Dan Geschwind, professor and associate vice chancellor at UCLA, shares his path to neuroscience, thoughts on precision medicine, and why one should use unbiased methods to prioritize hypotheses.

On The Bioinformatics CRO Podcast, we sit down with scientists to discuss interesting topics across biomedical research and to explore what made them who they are today.

You can listen onSpotify, Apple Podcasts, Google Podcasts, Amazon, and Pandora.

Dan is a professor and associate vice chancellor at UCLA. His group has pioneered the application of systems biology methods in neurologic and psychiatric disease, with a focus on autism spectrum disorders (ASD) and neurodegenerative conditions.

Transcript of Episode 14: Dan Geschwind

Grant: Welcome to The Bioinformatics CRO Podcast. I’m Grant Belgard and joining us today is Dan Geschwind. Dan, can you introduce yourself please? 

Dan: Sure. Thanks for having me here Grant. It’s really great to be here. I’m a professor of neurology, psychiatry and human genetics at the David Geffen School of Medicine at UCLA, and also in my role as associate vice chancellor and senior associate dean, I direct the relatively newly formed institute for precision health at UCLA. 

Grant: Can you tell us more about that? 

Dan: Sure. Basically the idea is that there’s a whole new wave of genomic information and a revolution on the horizon in medicine that even the most advanced academic medical centers have not really prepared for. Although again, like UCLA, we are beginning to prepare for it.

It kind of falls under this umbrella of what many call precision medicine. We call it precision health because it involves prevention and health. We’d like to be precise before people get sick. And the notion, I think Obama stated in the most succinctly and clearly of course–one of his speech writers did–every person gets the right treatment at the right time.

And so what that really means is that we take into account individual differences as we are treating patients. And that involves crunching a lot of data to understand the individual’s genetic background, as well as in the future, potentially a lot of other environmental and sociodemographic factors. There’s a huge amount of social determinants of health as well that can be put into these equations.

Grant: For what types of indications do you think we’ll see the earliest advances in this space? 

Dan: Well right now precision medicine is being done in about 6% of cancers. A mutation is identified in the tumor that requires a very specific targeted therapy. And of course, 10 years ago, that was less than 1% of cancers. So this is a rapidly changing field in the area of Mendelian disorders or rare genetic disorders, which individually are rare. These are disorders that are fewer than one in a thousand, but if you add up the percent of people who will have a rare disorder, or who have a common disorder, but a rare cause of it, it’s probably somewhere between 7 and 10% of people. 

And again, in major medical centers that might even be higher at tertiary care centers. And so those people get diagnosed. Often. They go through what’s called a diagnostic Odyssey where they go to multiple experts around the country. Nobody can figure out what it is and it’s not until they get a clinical sequencing study done that it becomes obvious, Oh, they have that disease because they have a mutation in that gene. 

And so we believe in taking a sequence first approach to people with rare diseases because we– being UCLA and my colleague, Stan Nelson and others– have shown that the yield of the test is somewhere close to 25-30%, depending on the indication. Again, congenital heart disease, rare neurodevelopmental developmental disorders, immunologic disorders, et cetera. And there’s really no single clinical test that has a yield as high as that. And as the sequencing costs get cheaper, it becomes the least expensive and most efficacious thing to do.

Of course, it’s not just an expense issue. It’s an issue of the poor patient’s family going through years of uncertainty and stress going to multiple doctors, et cetera. So those are the main areas that it’s being done. It’s in undiagnosed diseases, developmental disorders, of course the typical pediatric genetic syndromes, more and more in unusual clinical conditions and then in cancer.

And most of that is finding rare mutations that cause things, but there’s also common genetic contributions to diseases. And as more and more of that is understood we’ll be able to apply what are called polygenic scores, that is look at the common additive risk that people share in common that might predispose them to various common conditions. 

And we can use that in screening tests. So for example, you could imagine that right now the way women get screened for breast cancer is by age and there’s some consideration of family history. In the future, you could have polygenic scores and genetic testing to identify everybody with a rare mutation and increased surveillance at early ages in those patients. As well as try other preventative approaches that might come along. 

And then with those that are in the top fifth percentile, or depending on what percentile of the risk you’re in and your age, there would be a formula for when it’s optimal to screen you and what kind of screening you should have.

But you could imagine if you’re in the top 10% of risk for many conditions, you might have a risk that’s similar magnitude to some of those who have a rare mutation and one should probably be screened and treated that way clinically. And so as sequencing costs come down, it’s going to make sense for health systems and hospitals to actually do this as just part of their normal routine care. Hopefully insurance will reimburse it, but even if it’s not directly reimbursed, it’s going to have to become part of care one way or another. 

Grant: How do you think the rollout will happen? With this work being spearheaded at major academic medical centers, when will community health centers in Mississippi see this?

Dan: Well, that’s a great question. Well, number one, you know one of the pioneers in this area is the Geisinger Health System, which is a private non-academic system very much like Kaiser. I think those types of systems where they’re both the insurer and the provider really makes a lot of sense because it’s a very, very cost-effective approach.

So I actually see that it might happen there even more quickly than in some of our more disjointed systems where we’re actually not the insurer, just the provider. And I think smart insurance will make agreements with providers to actually provide these things because it will reduce cost and morbidity and the burden of illness later on.

But right now I think we’re in an area where there’s enormous promise. There’s some areas where certainly genomic medicine is used in clinical practice and should be used a lot more like I just described, but there’s a ton of research that needs to be done using the clinical records and genomics to identify those people most likely to benefit and to develop the algorithms.

So it’s a whole new area of research and medicine. One of my colleagues has likened it to thinking of the healthcare system as a learning healthcare system, where the patient records and other research data that we collect, like genetics, gets integrated. And we learn from that as we follow patient trajectories and then apply that information back into clinical practice.

And that’s our view of this at UCLA. And in fact, we’ve initiated a major, what we call population health initiative called Atlas, where we’re collecting 150,000, at least of our 45 million patients and genotyping them. And we have a collaboration with Regeneron genetic center to sequence those patients and get the data back.

Part of our plan is to return results that are clinically actionable. Like some of those that I mentioned, let’s say, if somebody has a familial cancer predisposition gene, or a rare mutation that increases their risk for heart disease in the lipid pathways, that those will get flagged and treated as long as the patient’s consent that they want that type of data back.

Grant: What if it’s not actionable? 

Dan: Then it’s not actionable. Then we don’t give it back because we’re not necessarily helping anybody. It might be of interest to people, but because we’re not really a commercial entity, we’re not trying to do things that entertain people.

For example, we’re really trying to do things that are serious and that help our health. We’re focused on things that are clinically actionable. And that our colleagues in medical genetics and the American College of Medical Genetics would largely agree are things that should get returned to patients because there’s actually something that you could do to help them.

Sometimes it’s just knowing that they have something that predisposes to certain things. And so they get an annual scan to check that out. And other times there’s actually an action or a medicine that should be taken to prevent things. So those are the types of clinical actions that should be taken.

Geisinger, who has kind of pioneered this in their sample, which is not an ethnically diverse sample, like our sample at UCLA, but still is a very informative sample and a very important set of observations have been published there. The prevalence of such clinically actionable findings could be as high as 3% and is going to increase as drugs get made. In fact, there are multiple drugs that have come out for certain diseases in the last two or three years that wouldn’t have been on that list five years ago. And so as that changes, we’ll be reviewing that and adding to that as well. 

Grant: What are your thoughts on the data management side of this? Do you think HMO’s will be storing a patient’s genome or exome sequences? 

Dan: Yeah. I mean, our data management is all on the cloud. Like we’re currently using the Microsoft Azure cloud, which our health system has been using and is very involved in. Microsoft’s been very collaborative on that, but I think most of it is going to be in the cloud.

Some will be on premises as well. It is a cost, right? And so for the smaller systems that don’t have academic laboratories and academic people to come and help with this, there may be commercial solutions to that, or local governmental solutions just depending on regulations and laws and things. So one example is that at UCLA, we’re building this enormous infrastructure and this involves the health system informatics people, who are experts in that.

But they’re also working hand in hand with people in engineering and computer science who were experts in databases and genetics and genomics–which the health system isn’t–to streamline these efforts and to make them work. And so that takes the merging of two very different cultures sometimes and working together. 

It’s been remarkable to watch it happen. It didn’t happen overnight. But it’s working really, really well in our system. We’re really fortunate to have collaborative, excellent people on both sides. So that’s really critical, but I could see how many of the smaller hospitals, or even large hospitals that don’t have resources like this and faculty who are willing without being paid to come and work on this thing.

Because number one, it’s so interesting. And number two, it’s so important. It’s such an incredible opportunity to have an impact. There are number of geneticists who are computer scientists, who have changed their direction of research because of this and are focusing almost a hundred percent on these biobanks and medical records now because they realize it’s a source of big data, but they can also have an impact with their research that was not possible before. But a lot of what they’re doing in the initial phases of this is quite voluntary. It’s really great for me to watch that and just see all the goodwill that comes with this new area. 

Grant: What do you think are the greatest roadblocks? 

Dan: I think it’s knowledge. There’s a lot we don’t know. In cancer, there’s 6% that are actionable. Even some of those aren’t cures. Right. We want to get to a point where we’re over three quarters or 90% or everybody. Right. But of course not all cancers are genetic, but even those with an environmental cause it’s of course probably driving something in the somatic tissue that one can identify.

And so the genetic and genomic technologies I think are there. And as the costs come down, they’re applicable. And I think culture has to adapt. I think it’s much easier to implement these things in countries like England, where they actually are now sequencing every child, who’s admitted to the hospital and certainly all of those in ICU. It makes total sense.

It should be done. It’s not being done here because it’s not an organized single payer system. And so it’s going to depend upon research philanthropy and a health system, deciding they want to waste some money on it, right? Because at some level it takes away from their bottom line. 

And a lot of health systems are profit driven. That’s by the very nature of our system. They have to remain financially solvent or they cannot operate. It’s a very different kind of system than most other developing and developed countries out there that have more organized public support for the health systems. And of course, with COVID, we’ve seen the remarkable failure of our lack of that.

So I’m hoping that we move towards a more organized, centralized system, so decisions are made rationally. In fact, COVID is an interesting example potentially of precision health. Where we, with a bunch of our other California health systems, have organized a consortium where we’re sequencing and genotyping our patients with COVID to identify mutations that make people particularly susceptible to very bad outcomes. 

Because as you know, most people get it and even if they get sick, it’s like a bad flu. But then there are those that have catastrophic complications, either neurologic, clotting, or pulmonary or immune. And we think that that’s likely due to rare variants in the immune system. That it’s due to host factors. 

There are probably two things going on. One is the dose that you get, the amount of COVID you’re exposed to and that gets into your airways and invades your body. And the second is how you respond to it. The host response, and that’s largely driven, not a hundred percent, but has a big genetic component and that can be measured. And there already are a lot of hints in published work that there are genomic features related to interferon and immune response that drive that.

And so down the road we could potentially, before anybody gets COVID know who’s at highest risk for these things and really that would be quite easy to prevent it, right. You just really keep those people away, make sure they have N95 masks and face shields and all that when they go out and really treat them carefully.

Then there are elderly people with a lot of comorbidities who should do really poorly, who just do great and are asymptomatic. And those people may have protectable alleles. And that could be super helpful in helping us develop drugs to treat it. So all of that is just a current example on many ends of how our system can approach this, but also how it’s failed in some ways, unfortunately, 

Grant: On the topic of COVID are you willing to go out on a limb and speculate about how we might see it play out this year? 

Dan: Well, I can tell you what my concerns are. And they’ve been there actually since the beginning. It’s viral evolution.

That’s why we don’t have vaccines to common colds and why we get flu vaccines and they’re marginally effective. So the more that something gets in human populations, the more it has a chance to mutate and also there’s natural selection going on. It’s a natural selection experiment.

It’s not just the mutations, it’s the selection pressures. I think it’s a strong possibility that this is not going away. And COVID is just an example of this. But these types of illnesses will become more and more prevalent and there’ll be times where they’re low level. And then when evolution occurs, they could explode and we have to be ready for them. But my sense is that even though I’m getting my second dose of vaccine tomorrow, I would not be surprised if in the fall or winter, I have to get another shot, not just to boost my immunity to what’s out there now, but what’s new. 

The good news is that technology could bring us ahead of this. Right. It’s really extraordinary how the RNA packaging with the lipid coated nanoparticles and things that have been developed by Moderna for years for delivering gene therapy or vaccines. It’s really shown its amazing utility. So I’m obviously very, very bullish on technology in the future, but I think just like HIV and hepatitis have really changed biology, it just shows us how important infectious disease is and will remain. We’re never going to conquer infectious disease, although we think we have/. Because it’s commingling with us. It’s part of who we are. It’s part of life on Earth. 

Grant: So let’s talk about your lab. What would you say have been the highlights of your scientific career?

Dan: Well, that’s a tough one. I’m hoping that they’re still ahead of me. My goal when I started all of this was to use the basic science–genetics, genomics, and neuroscience–to understand disease and to develop therapies in the disorders that I work on. We haven’t gotten there yet. So I’m hoping that the big advances are still ahead of us.

And we have some glimmer of that. I think the methods and approaches that we and others are taking with genomics really does provide new opportunities to develop better, more targeted therapies that are more mechanistically driven, safer and more efficacious. So I’m positive about that. 

It’s an interesting thing. I don’t reflect very much, but I had about 30 seconds of reflection this morning as I was looking through my calendar and seeing that I have a great chance to talk with you, Grant, which I’ve really been looking forward to. It’s always wonderful to talk with you. And I was thinking, you know, if I’m so interested in disease, why don’t I work at a biotech or pharmaceutical company?

And that has crossed my mind over time. I bet you were going to ask me that. And I think in some ways there’s some lost opportunities there, right? Because that is the focus of biotech and big pharma. It is to develop these medicines and to relieve suffering. But there are two parts of it. One is in the earlier parts of my career, I really got an enormous pleasure and always have, and always will from seeing patients, just having that experience. And even patients with rare genetic diseases that we can’t treat, discussing with them and talking about the research that we’re doing to give them and their families hope. Telling them that somebody cared about them and was working on this. So that was a piece of it. Another piece was the training. There’s nothing I enjoy more than walking through the lab at 5:00 PM and seeing what people are doing and just sitting down and talking with them about what they’re doing.

Unfortunately, you know, being so busy and now with COVID, I don’t get to do it at all, but I recognized at one point that with the notion of enjoying yourself while you’re trying to have a significant life and making a difference in the world and having impact, I realized that one of the impacts that gives me the most pleasure is watching students and postdocs grow in the lab. Interacting with them, arguing, learning from them because you know I’m still in school all the time.

And especially from the more mathematically inclined people like you, that I was always asking a lot of questions about. It was really, really great for me to explore these areas that aren’t my areas of expertise with some of the people in my lab, but also then help students and other people find their way. The training part of it has been really, really satisfying to me just personally. I enjoy it. It’s fun. Kind of just like the way I’ve really enjoyed having a family, having kids and having some dogs, even though sometimes I regret that latter decision when the dogs are chewing up the furniture. 

Grant: So if you were a college student today, in what direction do you think you would go?

Dan: Well, I’d be tempted to work in some startup-y kind of thing. Cause that seems to be what all kinds of smart, driven people are doing. I’m sure I would. In fact, as you may know, when I was in college, I wasn’t sure what I wanted to do. I was a chemistry major–it was called chemistry modified with psychology. So it was like a chemistry major with a minor in psychology. And I was interested in science and was thinking about medicine and had a lot of family role models in those areas.

But I had grown up in the sixties and seventies, when I was a teenager. And I was rebellious enough against any kind of authority–even though my parents were certainly not authority figures by any stretch, they were very gentle and kind to people and really helped me a lot. Still by nature, I was a–I don’t want to use any bad words on the podcast, but just fill in the blank–a rebellious ______.

Grant: It might’ve been some chemistry modified by psychology as well as psychology modified by chemistry. 

Dan: There you go. I think I’ve always been a little bit like that. So I kind of started to think about business and all my family were physicists or chemists or doctors. So business was interesting to me because of the entrepreneurial spirit and all of that.

And so I went to work at Boston Consulting Group actually, rather than go directly into science. What really attracted me to the Boston Consulting Group was that it was just full of incredibly smart intellectuals of all different sizes, really a huge variety of smart driven people. Some with PhDs in economics and with law degrees, women, men, you know a really interesting place. That’s I think what drew me to it. 

If I were in college today, I’d be looking for a place of opportunity, a place where I felt like I would have mentors. Even if they weren’t doing what I wanted to do, people who I really respected and was going to learn from and who would challenge me.

And that certainly was the case. So I think I would look for something like that. And I think in some of the startups, there’s quite a bit of that, but I’d probably end up in graduate school one way or another in the end.  Just because I think graduate school is one of the places where you learn the most about yourself.

Grant: So why did you go to grad school? Can you maybe talk a bit about why you left BCG? 

Dan: BCG was filled with impressive people, many of whom have gone on to do really impressive things. I was reflecting on my life. And again, part of this was family. My father had sent me some articles written by Freeman Dyson, the physicist who did his great work while he was 19 and 20.I had had this thought that, well, you know, I’ll go and make some money. And when I’m like 30 to 35, I’ll kind of be a gentleman scientist down the road. But I’ll make some money first. 

And I kind of realized that maybe those first two decades of life after college might be my most productive years. Of course, in biomedical science that’s not the case. It’s not like physics or math and I didn’t have the brain for those fields or the drive. 

But I’d done some chemistry research. And realized that I wanted to do something that was connected to people. And that really fit with the medical kind of thing. Medicine just seemed like the right thing. To me, it seemed like the perfect blend of science and helping people. If you like science and biology and you’re curious about people and want to help do something good, it just all fit together.

And talking with a family friend, really a couple things helped me. One: I had this friend named Craig Bradley. He had gotten one of these honorific masters scholarships, and he was in Edinburgh doing a master’s in English. He came back and we were having beers in Boston at, I think it was the Cheers pub. And the whole time I was talking about brain science for like an hour and a half. He goes, Dan, geez, we’ve been here for an hour and a half. You haven’t told me anything about your work at BCG. What are you talking about? All this other stuff, and it just kind of dawned on me. Huh. You know, maybe I have other interests.

So it was that. And some of this realization that life is short and this real drive that I still have to really have an impact and to have a significant life. You know, you can have a significant life in many ways, of course. And with your family, it’s always going to be significant, but to have an impact beyond my first degree relatives, to really have some influence to help humanity in that way. I realized I had this drive and I think that’s really important. 

Grant: Can you maybe elaborate a bit? I totally agree with you that in the biomedical field, for the most part, your greatest impact comes later in your career. Something I found pretty impressive about you is you’re able to kind of pick up enough about so many different things and then tie them together in really nice ways. Can you maybe elaborate a bit on that? What do you think are the most important compounding returns that makes a biomedical scientist so productive later in their career?

We actually recently did some work, looking at citation metrics of computational biologists. And what you see as people get later in their career, is there’s this huge divergence in different measures of productivity, starting from a smaller divergence. I mean, it really does look like compounding returns.

Dan: Yeah, it’s interesting. So, some of that may have to do with the sociology of science and human endeavors, right. Successful people read your work and then that helps you get grants and other things. So success does breed success like it does in other fields. If you started a few companies and they’ve been successful, it’s going to be easier for you to raise money.  

Same thing with grants. So there’s that kind of practical piece of it. It’s very interesting because my early view of this was that stuff that I’m doing now is a combination of a vague vision that I had when I was 37. 

I had several beliefs. One of them is that we didn’t know very much. Number two was going to take methods that were kind of unbiased. I really believed heavily in screening and discovery. We needed to spend a lot more time on that rather than testing hypotheses. And I guess part of that came from the notion, which I think most of my colleagues would agree with—there may be some who disagree with me—but me and most of my colleagues know which hypothesis is the best to test.

And therefore, if there are potentially thousands of competing hypotheses, why are you testing that one? You need to have a reason for it. And omics techniques and other systematic ways of looking at things that was the kind of thing that I started with. And, that was initially not a very popular way to go.

People would call those fishing expeditions. It’s not hypothesis driven, especially in neuroscience. So I think that’s part of it. It took a long time for that kind of stuff to catch on. I went for five or 10 years without being recognized at all, getting a lot of heavy criticism. But you know, my feeling in the long run is that what works out, works out, what doesn’t work out will go away eventually.

And that’s one of the problems with biology and biomedical research in general is that unlike physics and mathematics and to a lesser extent, things like chemistry—cause there’s all kinds of laws in chemistry—where there are fundamental theorems that drive things, the theoretical basis of the field and in biology has to do with the central dogma and also evolution, which drives everything. But that’s such an overarching explanatory theory that, you know, where do you go from there? So we’re at a very descriptive stage in all of biomedical research. And so my belief has been in collecting large data sets and trying to interrogate them and let the data hopefully bring us somewhere.

But sometimes we’re helped in looking at those data sets by having some knowledge of biology and some biological insights. And I think that’s where the kind of compounding occurs is by having some knowledge and also by reading a lot and having a decent memory. I can remember what I read and I go, Oh, I just read about that. And that is related to this.

And so it helps you connect things. And so having a broad fund of knowledge in this area is helpful, even when analyzing big data, because sometimes it’s not obvious what the next steps are. 

Grant: I would say, especially when analyzing big data. Right. It’s great to be able to look at a slide and recognize some old friends among the genes that pop out with these unbiased approaches.

Dan: Right. And you want to be careful about that, right? You know, about that old friends approach, the dartboard kind of approach. I know that approach. So you have to be sanguine about that as well. And sometimes what happens is it can really help you by having read something and understanding how these nine genes are part of a pathway that I just read about two weeks ago, or I heard a talk about them. That can really take you somewhere.

But you have to have some relatively rigorous way of prioritizing hypotheses. Let’s put it that way. And so using these genomic methods and more systems methods, it’s really helpful in that regard. That’s been my belief throughout my career. And I think finally people have started to work on it, you know, and use these technologies.

It’s interesting in neuroscience, people really weren’t interested in any omics technique until the single cell analysis came out. Which again is quite interesting to me and yeah reflects a certain type of bias in the way people are looking at it. But of course the single cell technologies are wonderful and extraordinarily powerful. 

Grant: It’s interesting. It seems to be almost universal for omics technologies that they’re pioneered first and cancer and they reach neuroscience almost last.

Dan: Well, I’ll tell you why. I mean, this is my view of this since graduate school, when I wrote my qualifier on the notion of oncogenes and neuroscience, what they were and why.

Number one, we are a decade behind cancer, but the reason for that is that cancer is a practical field. Cancer is based on combating a terrible array of illnesses that afflict humans. It’s all about discovering what the problem is and making a drug or a treatment to stop it.

Right. And the problem with neuroscience is that I study cerebellum. That’s the most important organ in the brain. I studied this. That’s the most important organ. I’ve studied that. The ground truth ends up being what very charismatic opinion leaders decide is going to be the most important thing.

And that’s how awards for neuroscience and other things are given. And I really do believe that that’s pervasive. You know, it’s almost like the high school prom voting. That’s not to denigrate those who have made extraordinary advances. I think there’s some things that if you got a hundred people in a room, 90 of them would agree on things like optogenetics, single cell approaches, receptor trafficking, and synaptic function, vesicle release, G-protein coupled receptors.

Nobody’s going to argue about how seminal those things are, but there’s a whole bunch of stuff around that. Again, it’s just in camps or almost tribal. What’s so interesting. If you look at the Society for Neuroscience, the organization of the meeting has not changed since 1985.

It’s not changed since then. And maybe since the sixties, when it started with this idea of motor systems, sensory systems, development, then there’s neurobiology of diseases as a separate little section. None of that organization has changed despite what’s been learned. And there’s also been chronically this gulf between kind of systems neuroscience and molecular neuroscience.

Right. You study how a receptor works using physics and math at the same level that you can study functional brain networks using similar methods. And I think as technology advances, those things will begin to merge. And my hope was that things like transcriptomics and these molecular techniques would provide intermediate quantitative merging phenotypes that would allow us to bridge molecular and systems neuroscience. And I still believe that although it’s of course a very, very tough path. 

Grant: Yeah, what do you disagree with most of your colleagues and why are you right? 

Dan: I don’t argue with anybody. It just doesn’t get you anywhere. I’ve learned that at home, after 35 years of successful and wonderful marriage. Have you ever read the book, which I read recently, called The Righteous Mind by Jonathan Haidt?

Grant: It’s on my list. 

Dan: Yeah. I think you should read that because it really gets to fundamental issues about people’s beliefs in general and how people argue. But I guess a couple things over the years that I’ve noticed that have been frustrating to me. Let’s put it that way.

Number one, when we started doing microarrays initially in neuroscience, everything was like, Oh, it’s a fishing expedition. It’s just a useless fishing expedition. Yeah. Well, how did you figure out that hypothesis that you’re testing? How do you rank that hypothesis? Quantitatively, show me the equation that you use to discern that hypothesis A is better than hypothesis B, you know? And so we would have those types of conversations where I would say what I’m trying to do is use these data to help me rank hypotheses using data, rather than some emotional attachment to something, which is how most of this has been done.

Not to denigrate the power of those emotional attachments. Those can be really important motivators and drivers, and that led to great discoveries. Don’t get me wrong. So I disagreed on that, obviously that was pervasive for about a decade. And again, it has to do with the fact that neuroscience is not necessarily focused on disease.

Neuroscience has a lot of different avenues. And in fact, many of us started studying the brain because it’s so interesting. Understanding the brain is a fundamental human endeavor. There’s so much of that, and it’s so difficult and it’s so challenging from so many different angles that all of that basic work is essential.

But I guess my belief has always been that even though I can’t know what’s important, like you might in mathematics, the notion that I’m working on disease makes me at least think or believe that what I’m working on is important. At least to some people whose diseases I’m studying. So that’s one thing. That’s an older thing now, cause that’s kind of water under the bridge. 

More recently what I’ve noticed among a certain ilk of extraordinarily brilliant people, geneticists, especially—it’s not everybody—some people have not understood or have very, very strong opinions, who heavily criticized transcriptomics or these omics techniques. They say they don’t have any place in genetics. What are they really doing? There’s no causality. It’s just a phenotype, blah, blah, blah statements.

Like none of these network methods work. Like what do you mean by that? Like what do you mean by work? And I think some of it has to do with the fact that the methods aren’t understood. People come from a math background and don’t understand how to connect it to biology. Sometimes it’s a lack of imagination. Even among the extraordinarily brilliant, imaginative people in one area like math or whatever, who don’t have imagination outside of that, or there might be actual, real criticisms that are extraordinarily valid, but I just haven’t heard them or had a chance to discuss them.

So there’s some of that going on, right? Again, this territoriality around that. As I get older and older and move along more, that becomes less important and we focus more and more on the output. In other words, we are moving things forward to a point where we might have therapeutic targets that then we can work with.

One area is dementia, which I think is helped by the fact that we’re looking at a phenotype that we can screen in a dish, which is a depth of a neuron. And in some instances in neurorepair, I think we’re actually making progress using these techniques. And we have proofs of principle where we can really show that these networks identify real targets that we can predict and then verify.

And that’s where it becomes really exciting when there’s experimental verification. So we have some projects in our lab that started over 15 years ago that are finally coming to fruition and that with other collaborators, we’re developing drugs based on these targets. And it would be a dream, you know, for these drugs to eventually go into people and have some impact.

Grant: Right. What lessons do you wish you had learned earlier in your career? 

Dan: Well, one of the things that I’ve learned, because there’s a lot of frustration in this work is I think the better you can accept criticism, the easier it is. I’ve always had a hard skin and belief in what I’m doing, but also accepting and not only accepting, but integrating the criticism has always made our work better.

And early on, that can be tough, but I think it’s really, really important. And I wish there were more criticism in a collegiate way. Another is there’s a lot of frustrations in academia that happen. A lot of people have written about this. There’s so much politics because there’s so little at stake.

I haven’t run into a lot of that, but I think when I do run into things that I think are wrong, either at the institutional level or otherwise are happening in the field. I think understanding how to reframe things and how to be positive and how to come up with solutions rather than criticism is another thing that has helped me adapt better than just identifying the problem and complaining with my colleagues about it, which a lot of academics do. It’s like, okay, that’s a problem. How do we solve that? So I would say those are two things. 

Grant: Do you have any parting words, words of wisdom, words of Dan. 

Dan: Buy low, sell high? 

What else? You know I took a circuitous route as you know, Grant, and we decided not to talk about that. I think we talked about things that are more substantive actually in some ways. And I appreciate that. 

Grant: We really wanted to talk about your path, but ran out of time.  

I mean all I could say is that I think it’s important when you’re young to explore your interests and your curiosities and not to funnel too much into what either your parents or society or other people are telling you to do. And I think some of my early wanderings had to do with that. Like for example, taking a year off college, and going skiing and making ski movies, going to the Boston consulting group all of those things were extraordinary growth experiences for me as a person. 

And that sat with me for my whole life. I do believe in experiences and I believe in following your curiosity, but I also recognize that I came from an extraordinarily privileged background, not from a let’s say financial standpoint, but from a familial support and background standpoint.

And that I had parents who were incredibly supportive and academic and intellectual. So I was exposed to a wide variety of things and there was no reason why I couldn’t do any of them. Right. I was expected to think about all of them. And in our society, that’s not the case for many people.

And I think even more so now because of the expense of education and how hard these things are to reach. And that’s been one of my real joys of being at UCLA, because it is a public institution. And although it’s an elite public institution in terms of academics, it’s not elite from a financial or other family background standpoint.

And there are a lot of first-generation Americans that I have the privilege to work with. And I have to say that that is some of the most satisfying work that I’ve had at UCLA. I think as a country, we really have to face that because so much of our greatness comes from people with drive, who come from outside as well as inside.

So, I do realize that I had a head start in many ways. I had many pathways that I could take. I’m grateful for that, but I also think that, no matter what your opportunity set is, exploring and following them one step at a time. I don’t know if you remember Grant, but one of my favorite statements in the laboratory, especially for graduate students, not as much to postdocs, was this great French proverb that I just love: Little by little, the bird makes his or her nest. 

But the notion that it’s a step-wise thing. And I really do view my career in that way as well. It was little by little. It didn’t happen overnight and it’s been constant work and high level of motivation, which I think is the most important thing in life: to work hard.

Grant: Yeah. It’s interesting. You’re actually our third podcast guest that essentially their message was you have to go and create your own path. The first was a Thiel fellow, who’s now going to be manufacturing organs in space. 

Dan: I think it’s really essential. There are amazing opportunities for people and careers and stuff that we couldn’t even imagine.

Grant: Well, thank you so much for coming on. It was really enjoyable. 

Dan: Yeah. It’s always great to talk with you, Grant, and I hope we can talk even outside this podcast sometime. Thanks for doing this really, really enjoyable. 

Grant: Thank you.