00;00;05;04 – 00;00;32;06
VO
Welcome to Beyond Lab Walls, a podcast from the Salk Institute. Join hosts Isabella Davis and Nicole Melnick on a journey behind the scenes of the renowned Research institute in San Diego, California. We’re taking you inside the lab to hear the latest discoveries and cutting edge neuroscience. Plant biology, cancer, aging, and more. Explore the fascinating world of science while listening to the stories of the brilliant minds behind it.
00;00;32;09 – 00;00;50;10
VO
En Salk, desvelamos los secretos de la vida misma y los compartimos más allá de las paredes del laboratorio.
00;00;50;13 – 00;01;16;21
Nicole
Hi, everyone. Welcome back to Beyond Lab Walls. I’m your host, Nicole, and I’m excited to chat with today’s guest. He was one of the first scientists to join the institute through the Salk Fellows program, which recruits researchers straight out of grad school to get a head start in running their own independent labs. He’s since been promoted to assistant professor and was recently named a Pew Scholar in the biomedical sciences, which is very exciting, and we’ll talk about a bit later.
00;01;16;23 – 00;01;35;07
Nicole
He’s a member of the Gene expression Laboratory here at Salk, and his lab uses molecular and computational biology tools to explore what happens when our genome gets folded incorrectly, and how that can lead to diseases like cancer. So without further ado, welcome, Jesse Dixon, we’re so happy to have you.
00;01;35;10 – 00;01;37;18
Jesse
Cool. It’s great.
00;01;37;21 – 00;01;49;27
Nicole
So before we get to hear, about your research, I’d love to hear a bit about who you were before you became a scientist. So where did you grow up and what was life like there? And what were some of your early experiences with science?
00;01;49;29 – 00;02;13;08
Jesse
So I grew up in Michigan, Ann Arbor, Michigan, and both of my parents were scientists, so I was sort of exposed to science and research from a fairly young age, though I have to say that they were not like super pressuring me ever to go into science. I’m sure that at times it was almost like extreme hands off, like, you know, choices.
00;02;13;10 – 00;02;31;24
Jesse
but yeah, so I think, I mean, honestly, when I was probably in high school, I was a, you know, a good student and I, you know, I was interested in science. But I think, honestly, a lot of my attention was occupied by sports. And, so I played competitive golf as a kid. and I was very much focused on, you know, thinking about how I can be the best golfer.
00;02;31;24 – 00;02;35;17
Nicole
For sure. Sure. When did you start playing golf?
00;02;35;19 – 00;02;55;17
Jesse
I, I sort of started playing with my dad when I was about seven, but I don’t think I really I don’t think he played very much until I was probably closer to like 10 or 11. And then we started playing more frequently, like almost, you know, every weekend or so. and then I think by the time I was like 12 or 13, I started playing in some competitive tournaments.
00;02;55;20 – 00;02;55;29
Jesse
Sí.
00;02;55;29 – 00;02;58;09
Nicole
So Tiger Woods by high school just no.
00;02;58;10 – 00;03;17;19
Jesse
No, no, I did play in college, so I went to Princeton for college. And so I played on the golf team there. So I mean, science was something I was always interested in. but I don’t think I really seriously started thinking about it as like a career, probably until I was actually in college.
00;03;17;21 – 00;03;27;06
Nicole
You’re also unique at talking that you have both an MD and a PhD. Can you tell us a little bit about, you know, what spurred that interest in both science and medicine in college? Yeah.
00;03;27;08 – 00;03;51;26
Jesse
So I think, you know, being interested in science, it was always sort of on the radar of the potential of going to medical school or, you know, becoming a physician. And then I think, you know, as I started getting more interested in science and research, in particular in college, I also sort of came to the realization that most of the instances where I was really interested in something is where it was very clearly related to human health.
00;03;51;28 – 00;04;11;24
Jesse
I feel like if there was a project we were working on or something we’re learning about in class, you know, the moment that it said, well, this is a way that this could be relevant for a particular human disease. That’s like really when my interest piqued. Got it. And so sort of recognizing that in myself, I began to look more into the possibility of doing medical school and ultimately did an MD PhD.
00;04;11;26 – 00;04;14;11
Jesse
So it’s like a joint degree program.
00;04;14;14 – 00;04;16;11
Nicole
a lot of years do in both those degrees.
00;04;16;12 – 00;04;35;10
Jesse
Yes. It was eight years. I mean, it was it’s a lot of time and extremely rigorous education. And I think that that’s actually really valuable in the long run. And that like I don’t see patients now. But the MD portion, it’s like the most rigorous education you can have in like human biology and pathophysiology that I could imagine.
00;04;35;10 – 00;04;46;08
Jesse
And so I think that there’s a way that, you know, it sort of forces you to learn about things that would be very hard to replicate. Just being like, I’m gonna do this on my own and try to understand human medicine.
00;04;46;11 – 00;04;50;05
Nicole
Sure, sure. Yeah, I’m going to read some books now. It’s very different when you’re in the clinic.
00;04;50;05 – 00;05;03;24
Jesse
So I ended up actually took a year off between college and worked back in Michigan in a lab, and it was applying for, MD PhD programs. and ended up coming to UCSD and doing the medical scientist training program.
00;05;03;26 – 00;05;13;20
Nicole
So did you have a sense of what specialty you might be interested in? Did you know you wanted to in clinical practice or research afterwards? What was kind of your decision making process at that point?
00;05;13;25 – 00;05;43;11
Jesse
So I when I started, I was pretty certain I was going to do neuroscience or some neuroscience related clinical discipline. And when I came to UCSD, I went through the first two years, which is sort of the medical school coursework, and was still sort of thinking along those lines. And then when I actually started doing my rotations for the PhD part, the first rotation I did was with a Joe Gleason, who’s, at UCSD, and his lab does sort of the genetics of human neurodevelopmental diseases.
00;05;43;13 – 00;05;52;28
Jesse
and I kind of realized working in his lab that I actually like sort of the genetics aspect more than the neuroscience aspect.
00;05;53;00 – 00;05;53;15
Nicole
Sí.
00;05;53;17 – 00;06;14;00
Jesse
and so then I started really trying to pursue more opportunities, looking at sort of genetics and genomics and sort of how our genomes work. And that’s sort of what ultimately led me to the lab, to my PhD, which was, gaining being ran with at UCSD. and that has sort of just continually led to what we work on and continue to work on.
00;06;14;01 – 00;06;22;13
Nicole
Sure. Yeah. So what was it about, about the genetics that, you know, attracted you most? I, I’m a neuroscientist, so I’m a little biased.
00;06;22;17 – 00;06;51;29
Jesse
I think that one I think it’s very logical to about how we store information. It’s about how you encode information. It’s about how you kind of set up developmental systems to develop an organism. I also think that there’s a component where and I think that this is probably reflected in some of the things we work on today, is in instances where you can find, say, a mutation, there are certain circumstances where there’s maybe an immediate molecular consequence, but there’s sort of a mystery until mobile.
00;06;51;29 – 00;07;06;02
Jesse
After that, depending upon what the kind of mutation is or where the mutation is in the genome. Sure. And so I think that that was also really interesting is like this very open ended. Like what what is actually happening once you identify sort of the initial genetic alteration.
00;07;06;03 – 00;07;29;11
Nicole
I can see the attraction to the logic of it, but then there’s still so much mystery still to the logic to unpack. So yeah, and I totally get that. Well, yeah. Before we get into the details of, you know, your current work, maybe we can do kind of a quick genetics 101 sort of refresher for, for any listeners who don’t think about this stuff every day in each of our cells, we have a copy of all our DNA.
00;07;29;13 – 00;07;54;24
Nicole
This is this, you know, long sequence or code of instructions for anything the cell might want to do or produce. And certain sections of that DNA sequence are called genes because they can get transcribed into RNA and then translated into specific proteins or other products that can then carry out a particular function in the cell. But our DNA isn’t just this loose string floating around in the nucleus, right?
00;07;54;25 – 00;08;15;16
Nicole
It’s actually stored in a much more complicated structure, as you’ll talk about. And only certain genes can be read or expressed at a given time. You’re kind of responding to the environment or, you know, you have different functions that you need to carry out at different times. Can you take things from there and kind of tell us more about that structure of our genome and the role of regulatory elements that you study in your lab?
00;08;15;19 – 00;08;35;26
Jesse
sure. So one of the ways that we sometimes sort of introduce this is, you know, you have 3 billion base pairs you inherit from your mom and 3 billion base pairs you inherit from your dad. And if you take that 3 billion base pairs and you stretch it out in kind of the classic watching in Crick double helix, that would actually measure over two meters in length.
00;08;35;28 – 00;08;56;25
Jesse
and so that has to be compacted down into the nucleus of every cell in our body, and then has to be stored in a way where not only you’re compacting it, but you’re actually able to access it and, read the information so that a cell can sort of respond to appropriate cues and developmental programs that can respond to stimulus or stress or nutrient signaling.
00;08;56;28 – 00;09;09;04
Jesse
and so, you know, that’s a sort of a significant challenge for a cell is how do you actually compact all of this DNA into a very small space in every cell, but do so in a way where you can actually still read out information from it effectively?
00;09;09;06 – 00;09;23;14
Nicole
Right, sure. Yeah. I’m picturing just like a little string of yarn if you needed to, like, crunch it into a tiny ball. But then still access specific parts of that string while it’s in this complex ball. That’s yeah, definitely a complicated task ourselves, but somehow they do it.
00;09;23;15 – 00;09;44;12
Jesse
Yes. Yeah. So I think the other part that I also think is interesting is, you know, you mentioned sort of genes and how they code for RNA and ultimately protein. So one of the things that we focus on a lot is how genes are regulated. So what are the sort of processes and signals that actually will turn on a gene in one cell type and maybe not in another?
00;09;44;15 – 00;10;08;16
Jesse
And at a very basic level, every gene is fundamentally conserved as a sort of gene structure all the way from humans to bacteria. And that, you have a part that’s going to code for protein. But at the beginning of the gene, you have a region that essentially has the information that allows the cellular machinery, particularly what’s called RNA polymerase, to basically sit down and make that initial RNA copy.
00;10;08;18 – 00;10;13;25
Jesse
And again, that’s conserved as a basic structure from humans to bacteria.
00;10;13;28 – 00;10;20;12
Nicole
So it’s kind of like this bookmark on the DNA that’s like this next part. You know if you want to read it come here. Start reading.
00;10;20;13 – 00;10;51;08
Jesse
Exactly. But you know even again in bacteria you don’t necessarily want to express every gene always at the same time. And so there are ways in which we can have other sequence elements that are usually adjacent to that beginning part of the gene that will allow the gene to be expressed in particular context. And so like the earliest examples of this are actually from bacteria where you have cues actually respond to different types of sugar, so that the bacteria will make particular metabolic enzymes that allow it to metabolize one type of sugar versus another.
00;10;51;11 – 00;11;00;11
Jesse
And that is basically controlled by proteins binding right next to the start of a gene, which would call a regulatory element, and turning that gene on or off, depending upon a particular cue.
00;11;00;16 – 00;11;00;28
Nicole
00;11;01;12 – 00;11;22;15
Jesse
and again all the way to humans, we still have regulatory elements that will respond to signals in our environment. but the thing that we start to do as you get to be in more and more complex body plans is for some reason, we started moving the regulatory sequences away from the genes that they actually regulate.
00;11;22;18 – 00;11;44;15
Jesse
and so whereas in bacteria and in maybe yeast, most of the regulatory sequences are immediately adjacent to the beginning of the gene. We start moving in far and farther away such that like in humans or other vertebrate species, you can have instances where the regulatory element might be within a gene and it doesn’t regulate that gene, it doesn’t regulate the gene next to it, but it regulates a gene that’s sort of further away than all of us.
00;11;44;18 – 00;11;46;09
Nicole
And that seems kind of chaotic.
00;11;46;10 – 00;12;02;13
Jesse
Yeah, it’s like there’s a certain level that it doesn’t really make sense. Like, I think it’s kind of a very weird thing that our genomes have evolved to basically store information in places that it’s not near the sort of target of, of where that element would need to act.
00;12;02;15 – 00;12;02;29
Nicole
00;12;04;08 – 00;12;26;24
Jesse
and so there’s actually a very kind of one of the really famous examples of this is the regulatory sequence that’s actually involved in the control of the development of your hand. And there’s mutations that can actually arise that give rise to a syndromes called pilot activity. So basically means you have an extra finger right. And this is actually not uncommon in like cats and chickens in some instances where it happens in people.
00;12;26;26 – 00;12;48;01
Jesse
And there’s actually these like famous examples at Ernest Hemingway’s house in the Florida Keys, where it’s like overrun with these six toed cats. And actually know the exact mutation that leads to these. They call it like a Hemingway mutation. But the kind of wild thing is the sequence element that is mutated is about 150 base pairs long, and it sits about a million base pairs away from the gene that it actually.
00;12;48;06 – 00;12;49;04
Nicole
Oh, man.
00;12;49;06 – 00;13;04;28
Jesse
And like I sometimes like to say that you know, if you think about this in terms of like if you have a person who is six feet tall, the equivalent thing would be them being able to communicate with another person that would be like six miles away. But like, you don’t have a cell phone, you can’t send a letter.
00;13;04;28 – 00;13;21;01
Jesse
Or if you’re a, you know, sequence element. So there has to be some way that they can actually physically communicate to each other. And so that’s been kind of one of the major motivating aspects of a lot of the work that we do is trying to understand when you begin to move these regulatory sequences further and further away from genes.
00;13;21;03 – 00;13;32;06
Jesse
How is it that they’re actually able to find those genes? How can they communicate with them and relay information? And also how can that process basically go wrong? Know different human diseases?
00;13;32;08 – 00;13;40;13
Nicole
It sounds like part of that comes from, you know, unique forms of genomic structure. So can you tell us a little bit about that?
00;13;40;14 – 00;14;00;17
Jesse
Yeah. So the the the idea is, you know, if you do have these sequence elements and genes are actually really far away from each other, there has to be some sort of biophysical mechanism that allows that information to be transmitted. And so the way that a largely appears to happen is that you actually bring the elements in close physical proximity to the genes that they regulate.
00;14;00;19 – 00;14;14;17
Jesse
and so that you could actually have proteins that would bind at the regulatory element and proteins that bind at the gene, and they’d actually be able to come together and physically interact with each other. And then that can facilitate transcription or the expression of those genes.
00;14;14;19 – 00;14;36;21
Nicole
So if you’ve got one part of a string that needs to talk to another part of a string, then you kind of create a loop in the string to bring those two parts of it together. And so now they even though they’re along the line of the string, really far away from each other in physical space, you’ve pushed them closer to each other, but now you’ve got this say, loop or this, you know, new structure of the string all together.
00;14;36;21 – 00;14;39;07
Nicole
And that’s kind of what’s happening in our in our genome.
00;14;39;07 – 00;15;01;16
Jesse
I guess that’s 100% right. And I think that one of the ways in which we really got into this in the first place was trying to say, well, can we make a map of how different elements and how different genes are actually related or associated in three dimensional space, which sounds.
00;15;01;18 – 00;15;34;13
VO
If you’re enjoying this episode of Beyond Lab Walls, be sure to check out our other channels at Salty Edu. There you can join our new exclusive media channel, Sulk Streaming, where you’ll find interviews with our scientists, videos on our recent studies and public lectures by our world renowned professors. You can also explore our award winning magazine, Inside Sulk, and join our monthly newsletter to stay up to date on The World Within These Walls.
00;15;34;15 – 00;15;50;14
Nicole
You’re creating this map of all these different folds in the DNA, and you know how that helps these regulatory elements get to the genes that they want to regulate? Are are those structures are those folds the same for everyone?
00;15;50;16 – 00;16;12;05
Jesse
I would say for the most part, yes. If we were to look in a normal cell in one individual and another, they’re going to look very similar in terms of the patterns of folding. If you look between different cells, there can be differences that might be related to sort of functional differences in one cell type versus another that you might have some elements that are in closer proximity to a gene.
00;16;12;05 – 00;16;29;27
Jesse
And that may be part of how you actually can have regulation of that gene. Basically why it’s on in one cell type versus another is that you’ve actually brought the element nearby in the cell type where it’s actually going to be expressed versus it’s sort of remains separated in one where you don’t want it to be expressed.
00;16;29;28 – 00;16;51;14
Nicole
Got it. So if you’ve got a gene that helps make a heart cell a heart cell, then you want the regulatory elements that turn that gene on to be close to that gene. And the DNA folded in that way to promote that in a heart cell. But you don’t necessarily need that loop, say, in a liver cell where you don’t want that heart gene expressed as that.
00;16;51;16 – 00;16;58;10
Jesse
So yes, there’s a component of that. That’s true. We can also see that there’s other parts that are actually fairly well conserved across a lot of different cell types.
00;16;58;13 – 00;17;02;25
Nicole
Sure. I mean, all our cells have to do some similar things across all cells.
00;17;02;26 – 00;17;29;11
Jesse
Yeah. And so that was actually one of the things that we, discovered when I was a graduate student is we were essentially making these maps of how your genome was organized, and we could identify that there were particular patterns where there would be a stretch of your genome, where it would look like lots of genes and regulatory elements were in close physical space, and then sort of a neighboring region where lots of elements would be in close physical space.
00;17;29;13 – 00;17;49;11
Jesse
but the two would be sort of separated from each other. And so we call those domains or we call those topologically associated domains. It’s almost like a neighborhood. It’s like who you see, who you interact with on a given day. And for the most part, we can actually see that a lot of those domain patterns are reasonably well conserved between different cell types.
00;17;49;13 – 00;18;07;26
Jesse
There are instances where they they’ll absolutely change, but there’s a lot of similarity. There. but within them, I think you can actually see a lot more sort of dynamic changes from, from one cell type to another. that I think are more important for mediating these sort of specific effects of, you know, when the gene is turned on or when it’s not turned on.
00;18;07;26 – 00;18;10;23
Jesse
You have these differences in the loops that are forming.
00;18;10;24 – 00;18;31;01
Nicole
Sure, sure. From cell to cell, it’s got similar kind of city organization, similar neighborhoods. But again, there’s always going to be little differences in the streets and things. Yeah, that’s from cell type to cell type. You said across different individuals for the most part that looks similar, though. What happens in the cases when it’s not.
00;18;31;07 – 00;18;53;12
Jesse
Yeah. So I mean, the most obvious case where it’s not is in the context of cancer genomes, because in cancers you begin to have mutations that don’t just change one base to another, but they actually break entire chromosomes apart and essentially stitch them back together in ways that are not reflecting what’s going on in every other cell type.
00;18;53;12 – 00;19;28;08
Jesse
Cancer chromosomes almost look like a mosaic of a set of normal chromosomes. And so those because you’re really fundamentally breaking DNA and reorganizing it, almost like by definition, it has to alter how your chromosomes are folded. Since I basically come to sort, that’s become a really major focus of the lab, is actually trying to understand, how the these kind of mutations that will break and rearrange chromosomes, how that actually is altering chromosome folding, and how this might lead to inappropriate expression and why that potentially could contribute to why the cancer cell becomes a cancer cell.
00;19;28;10 – 00;19;49;04
Nicole
If in a particular person, you have a mutation that is breaking the DNA in a certain way, and then it gets rearranged in an abnormal way, then these regulatory elements aren’t getting folded towards the genes that they normally would regulate. You’ve got some kind of weird mismatch. So I can see you. Yeah. Then how that would lead to issues.
00;19;49;04 – 00;19;52;26
Nicole
So what’s how does it directly lead to cancer though. Yeah.
00;19;52;26 – 00;20;15;08
Jesse
So lymphoma is one of the very first cases that people recognize where you could have these kind of rearrangements, where basically a regulatory element is now near a gene that it otherwise wouldn’t have seen in evolution. And it’s now driving the expression of this gene. And so lymphomas are cancers of your immune system. And they can either be essentially B-cell derived or T-cell derived.
00;20;15;11 – 00;20;21;11
Jesse
And so B cells are the cells in your immune system that make antibodies in response to an infection.
00;20;21;14 – 00;20;22;03
Nicole
00;20;22;06 – 00;20;45;05
Jesse
And in these lymphomas, it’s very common that you’ll actually have these kind of rearrangements of DNA where the regions of the DNA that actually encode the genes to make antibodies, the regulatory sequences that would normally tell those antibodies, hey, we’ve we need to make a lot of antibody, because that’s our job as a B-cell. And we need to fight infection and whatnot.
00;20;45;07 – 00;21;13;15
Jesse
Those regulatory elements are essentially broken and switched and essentially put adjacent to genes that might be regulating cell growth. And so the idea then is that the cell thinks it’s trying to send a signal from these regulatory elements to the genes to say, let’s make a lot of antibodies. But what it’s doing instead is it’s sending a signal from these regulatory elements to a gene that’s now telling the cell to grow uncontrollably.
00;21;13;17 – 00;21;14;10
Nicole
Sí.
00;21;14;12 – 00;21;30;15
Jesse
and so the fact that you essentially rearrange this now allows this kind of inappropriate communication to actually occur, and that it’s really kind of like, at least in these lymphoma types, is really kind of V or one of the causal reasons as to why these actually become tumors.
00;21;30;15 – 00;21;49;24
Nicole
So like in that case, the cell that B cell sees, you know, some kind of pathogen, it’s wanting to make antibodies. It’s thinks that it’s telling this regulatory element to go make antibodies. But by instigating this regulatory element, because it’s now in the wrong place, it’s not telling the antibody gene to go. It’s telling the cell growth gene to go.
00;21;49;24 – 00;21;59;28
Nicole
And so the B cell is accidentally making itself cancerous. Yeah okay. That’s rough. And.
00;22;00;03 – 00;22;23;26
Nicole
So in your lab you’re kind of starting with these genome sequences that you’ve either obtained from tissues. You’re interested in your lab or from publicly available patient databases. And then you’re using these computational tools to identify these patterns in the genome sequence and structure. What’s kind of the main question you’re asking? Or, you know, the main goal then of your lab.
00;22;23;29 – 00;23;00;22
Jesse
Yeah. So one is if if you can learn the rules of when you have these regulatory sequences driving genes that they otherwise shouldn’t be driving, in many instances, those can be potentially sort of like a causal or essential mutation for that tumor to grow. And so that gives kind of a couple of different opportunities. And one of them is if you know which events are actually important in an individual, and you may have a gene that is potentially has a drug that you could use to target it, that’s that’s potentially an opportunity to think about these kind of like strategies for personalized medicine.
00;23;00;24 – 00;23;37;11
Jesse
the other is that given that, you know, it is sort of rare that a mutation actually leads to these kind of altered expression patterns. We’re really interested in trying to understand like, well, what what is the kind of, again, biophysical mechanisms that allow that sort of inappropriate communication to occur, because that ends up being, at least in my mind, that would be a very interesting or fruitful thing to think about developing therapies in cancers that if you can essentially eliminate instances of this kind of inappropriate communication going on, that that would really be something that would be much more selective for affecting expression and growth of a cancer cell, as opposed to a normal
00;23;37;11 – 00;23;38;05
Jesse
selling your body.
00;23;38;06 – 00;24;06;00
Nicole
So I can see, because it’s getting more common to be able to do genetic analysis of people and people’s tumors. But it’s not always clear from that sequence what what the outcome will be, what will actually go wrong. So you’re kind of developing that computational model or computational tool, kind of figuring out what the rules are. Then in those cases, what will actually happen and then how you can use that to understand that person’s cancer, to understand how that person might be best treated.
00;24;06;02 – 00;24;13;03
Jesse
I mean, that’s right. That’s the dream.
00;24;13;05 – 00;24;24;13
Nicole
I want to highlight that again. You were, recently named a Pew Biomedical Scholar. So can you tell us a little bit about this distinction? And what does it mean for you and for your lab?
00;24;24;16 – 00;24;34;21
Jesse
Yeah. So PUE is it’s considered sort of like a prestigious award for sort of young or junior faculty. And so at some level, it’s very nice to be honored like that.
00;24;34;24 – 00;24;37;06
Nicole
Of course, it is a big deal. Yeah.
00;24;37;09 – 00;24;53;04
Jesse
Christy Towers actually was a pew they call a pew Stewart and so it’s, you know, it’s a very cool thing to people say that they like the work that you’re done in the past and that you’re, you know, proposing to do in the future. I also think it’s, you know, it provides support for the lab, which is also fantastic.
00;24;53;04 – 00;25;22;10
Jesse
And, you know, we really want to try to use the some of the support to actually begin to develop systems to really test in much greater scale these kind of, effects of mutations that that we can see. And so I think, you know, maybe they give a little bit more context to that. We use these CRISPR-Cas9 genome engineering as a way of introducing some of these breaks and rearrangements of chromosomes, but you’re sort of limited to doing that essentially like one event at a time.
00;25;22;10 – 00;25;31;02
Jesse
And so that is challenging. It takes a couple of months to make any individual cell line. And so, you know, if you want to make a bunch of these to study a bunch of different genes, that’s a lot of time.
00;25;31;05 – 00;25;41;05
Nicole
and so this is again, you’re, you’re kind of introducing a mutation to see how it would affect the genome. Okay. Got it. Yeah. So doing that one at a time would be slow.
00;25;41;08 – 00;25;57;11
Jesse
Yeah. And so I think one of the things that we really want to try to move towards is finding ways of doing this, like in much higher throughput and being able to really introduce more and more mutations simultaneously so that we can actually really begin to understand these different patterns of ultra expression that we’re seeing in cancer genomes.
00;25;57;13 – 00;26;21;10
Nicole
So kind of instead of just waiting to get patient samples that happen to have different mutations, and then using that to learn what’s going on, you can kind of tweak the system yourself, like keep creating these little breaks or mutations and seeing how then the genome and the gene expression response. And the more of that information you collect, the better you can build your model.
00;26;21;10 – 00;26;53;29
Jesse
Yeah. Yeah. I mean, I think that that was and ends up being, you know, if you’re if you’re looking in, say, a tumor from a patient, you can see where the mutations are, you can see what genes are expressed. And there’s some tricks that I think you can use to try to relate those to each other. But from a, you know, really causal standpoint, like being able to actually introduce the mutation de novo in a cell in the lab, is the clearest way of actually saying, what is the consequence or what is the effect.
00;26;54;01 – 00;27;13;28
Jesse
Sure. And so that’s actually really where a lot of the kind of predictive models that we’ve been trying to develop are really focused is if we actually engineer these events, can we predict what the outcome is like, what gene will change, whether it will change, really kind of use that as a system to both learn the basics of the system, but then also as a tool for predictive modeling.
00;27;13;28 – 00;27;38;06
Nicole
So yeah, the more that you can play with the system and start to understand its rules and define its rules with your models, then the more if you get a case with a new mutation, you’ve learned enough about the language of this and the logic of this, that you can kind of predict how a new mutation would change, the structure would change gene expression, would change cell function and potentially lead to disease or not.
00;27;38;08 – 00;27;45;01
Nicole
Yeah I got that. So then where where do you see this going in the next five, ten, 15 years.
00;27;45;03 – 00;28;05;29
Jesse
I mean I think that I think if we, if we’re able to really increase the power and throughput of, of generating these kind of mutations so that we can expand this to make predictions, I think would be a really kind of important way to go. I think the other thing that actually helps a lot from the standpoint of thinking about these predictive models, is actually deep learning and AI type methods.
00;28;06;01 – 00;28;26;13
Jesse
there’s actually some very, very cool papers that have come out in the last few years that we’ve used a little bit trying to predict the folding of the genome from the sequence alone. So if you if you knew the pattern of the mutation that you might have from like a whole genome sequencing of the tumor, you can then begin to infer what are the three dimensional structural changes that are occurring as a result of that.
00;28;26;15 – 00;28;43;00
Jesse
and so I actually think, you know, these kind of AI type methods are going to be really powerful for understanding these effects in the future as well. And.
00;28;43;02 – 00;28;52;04
Nicole
I know that you have collaborations with some other folks at Soc2. So how are you using these tools in these? Yeah. You know, computational models applied to other questions.
00;28;52;04 – 00;29;13;09
Jesse
I think a lot of it stems from the fact that many of the things that we’re interested in are really basic. It’s how your genome is organized in space and how that allows different elements in the genome to communicate and have functional relationships. And so I think, you know, that was a big part of what brought me to Salk, or why I thought it would be interesting and fun to be here.
00;29;13;09 – 00;29;44;07
Jesse
Is that as an institute, there’s a fairly good amount of, I would say, scientific diversity in that you have people working on plant biology or people working in neuroscience, cancer biology from the standpoint of like some of the tools and questions that we’re interested in, those can be applied broadly in a variety, different contexts. And so we’ve had collaboration with like, yeah, Chang’s lab looking at T cells and sort of how genome organization differs in different classes of T cells and how that may be related to sort of different functional states of T cells.
00;29;44;09 – 00;30;08;29
Jesse
We have collaborations with Joe Ecker where we’re looking at genome organization and brain cells, and how that differs between different types of brain cells and neurons versus not in their own cells. And then actually also having work that we’re doing, collaborating with Julie Lau, looking at genome organization and plants O which is, I think, actually really interesting in part because plants don’t really have the same kind of patterns of these elements.
00;30;08;29 – 00;30;26;26
Jesse
Now moving further and further away from the genes that regulate. So they actually keep a lot of the regulatory elements fairly close by. But we do see that there are some kind of instances where you get to actually have these longer range communications taking place and kind of trying to understand why that’s happening and what what that’s really allowing the cells to do.
00;30;26;29 – 00;30;55;11
Nicole
Right. Because any of these other scientists might get to a point where they see a mutation, but they don’t really then understand how it’s breaking these genomic rules. And so you’re able to kind of use computational approaches to understand those rules better. And then you can apply those tools, those methods, your way of thinking to all sorts of questions and kind of help them then understand what that outcome is in a neuron, in a immune cell, even in plants.
00;30;55;11 – 00;31;14;12
Nicole
That’s pretty cool. Yeah, yeah, we’re excited to see you do it. So if you listening, want to learn more about Jesse Dixon’s work, visit the Salk Institute website at Salk Edu. And yeah, check out future issues of Inside Salk magazine where we might get to learn more about his work. Thanks again, Jesse, for chatting with us and thank you all for listening.
00;31;14;19 – 00;31;15;23
Nicole
See you next time.
00;31;15;25 – 00;31;31;25
Jesse
Thanks for having me.
00;31;31;27 – 00;32;02;14
VO
Beyond Lab Walls is a production of the Salk Office of Communications. To hear the latest science stories coming out of Salk, subscribe to our podcast and visit Salt Edu to join our new exclusive media channel, Sulk Streaming. There you’ll find interviews with our scientists, videos on our recent studies and public lectures by our world renowned professors. You can also explore our award winning magazine, Inside Salk and join our monthly newsletter to stay up to date on the World Within These walls.