Extrachromosomal DNA and Cancer Development with King Hung

Welcome to the Axial Podcast. Axial is an early-stage investment firm based in San Francisco. We partner with great founders and inventors investing in early-stage life science companies, often when they are no more than an idea. Axial is fanatical about helping the rare inventor who is compelled to build their own enduring business. Okay, we're recording. Hey, King, thanks for taking the time to do this podcast. Really excited to talk about your scientific journey and your research around genomics and X-chromosomal DNA. And, you know, we had like a 15-minute debrief already. Maybe we should record that too. We should record it all. Maybe, maybe I should start recording everything rather than just like doing some record then talk. But how are you?

I'm great. Thank you, Josh, for inviting me on this podcast. Very excited for this discussion. Discussion as well.

Yeah, me too. I think you're, you know, you're another superstar. We only talk to superstars, and so I'm excited to just, you know, talk about your story and then talk about your research. And we can always do this again, uh, because I'm sure, you know, I'm sure you're going to do a lot more. And congrats on the new paper. That's news to me now. So, um, and so I'm really excited for like the more research you do. So maybe to start off the conversation And to set the table, you know, King Hung is a PhD student in Howard Chang's lab at Stanford, and his research is really centered around using a variety of genomic tools, mainly to study extrachromosomal DNA right now. I'm sure he has other projects in the works, and he has published a really awesome Nature paper some time ago, and I'll put it in the description, using a variety of genomics-based tools to discover extrachromosomal DNA hubs. And I think we'll talk about that research right now. But maybe to start off the conversation, King, maybe you can talk about your story. How did you get your name? Where does the name King come from? How did you get into science? Stuff like that. Anything you want to talk about?

Yeah, okay. That's actually a great question. People ask me— so King Hung is Cantonese. It's not English. People ask me like, Is it— was it— was I named— like, I was not named because my parents wanted me to be King, or I did not name myself King because I picked an English name. It was actually just a phonetic Cantonese word for "be watchful," and that's how it was translated. But anyway, cool. Yeah, I grew up in Hong Kong where scientific research is really not a common career, so I did not know that it was a legitimate career, honestly. I've always been fascinated with biology. My family and I would go on hikes and watch documentaries, and I've always been fascinated. But when you watch a biology documentary, usually you see people exploring in the woods and, you know, going to all these like remote ends of the world finding rare species. And I always thought that that was not a VAD career. Like, I don't want to be just traveling to nowhere. So I didn't know that scientific research would have been like a stable VAD career until I, in the middle of college, when I started doing undergrad research. And I, so I worked with Dr. David Kamelman during my undergrad on embryonic development. So we did some work on looking at gene expression programs in certain cell populations in zebrafish embryos and trying to figure out how these expression programs, you know, drive formation of specific tissues in the embryo. And that got me really excited about just the complexity of all these different cell programs and how coordinated they are and how important that is for just normal, normal, um, functions, right? Um, and really, that's mainly what got me into cancer. If you think about cancer biology, You're asking the very simple question of you have some cells that are growing out of control, so they're not doing what they're normally supposed to do. So the first thing you have to figure out is what are they normally supposed to do, and then you say, okay, what are they doing differently, right? So pretty much all of cancer biology is trying to figure out, you know, what is normal and now what is different. And that's fascinating to me.

Interesting. It's so curious. I mean, like, at least for me, I always romanticized like biologists, you know, who go on ships or go out in the jungle. And then in reality for me, like, you stay in lab all day. So it was like, oh, so maybe you're the opposite. That's why you're so good. You just like, you love the lab. Sometimes people love the lab, right? Or, you know, I think I like being outside.

Yeah.

Maybe the opposite. That's so fascinating. And so when you were an undergrad, that seems to be like a pivotal moment. You know, was there like a certain time or was there a certain like thing that made you realize, I want to be a scientist? Was it mentorship from your PI? Was it, you know, joining a lab for the first time? Was it a class? Was it a paper? Or kind of what was the thing that triggered you to say, hey, I want to be— maybe you love science, but you want to be a professional scientist.

Yeah, yeah, so I think it's a combination of all those things. I did have very good mentors along my journey. My PI David Kimelman was also my biochemistry professor at the time and he gave just awesome lectures on— he just made very complicated ideas so simple and so easy to understand. And that's why I joined his lab in the first place. And I actually worked with a postdoc in the lab who taught me how to be independent, how to do rigorous science. And that had a huge impact on me as well. But really, I think I clearly remember a moment when I did One of my first in situ hybridization experiments in the lab. So you— the goal of the experiment was to detect gene, a particular gene that may be expressed in a subset of cells in an embryo, right? So you fix and treat the embryo, and then you add a probe that recognizes the gene. And then parts of the embryo lights up and you say, oh, those parts are expressing the gene. On paper, it sounds so convoluted, but then you do the experiment and then you actually look under the microscope and it's just, it's so beautifully shows, you know, patterns of gene expression that makes sense. And that's always been surprising to me, even now, is that things make sense. You know, because things are so complicated, wouldn't it be more likely that most of them don't make sense? Although, if things are beyond your comprehension, maybe you just they're not aware of those things in the first place. But I don't know, it's always been— actually, wasn't it Albert Einstein who said something like, um, it's surprising or it's incomprehensible that the rules of the universe are comprehensible to us, right? Like, shouldn't the rules of the universe be incomprehensible? But that's, yeah, that's always been something that I don't know if it motivates me to do science, but it certainly keeps me intrigued and fascinated.

Yeah, I think a lot of great scientists— did your first experiment work or did it fail?

Oh, my first experiment was probably It worked, but it was also meant to be a training experiment. Like, it was meant to work. It was meant to be easy.

Yeah, exactly. That's good, right? Get your confidence going. Yeah, I think developmental biology is really probably the most beautiful of the biologies because you can get these really distinct patterns. I think, you know, Drosophila amazes me. Like you said, right, you have like a whole systematic set of factors that drive, you know, development across fly embryo. And then that's across humans and zebrafish and everything, every other animal. So yeah, I totally agree. It's like, weird. I'm a geneticist, and so, you know, there's certain simple rules, simple formulas that describe genetic phenomena, which is like, shouldn't be the case, but it is. And yeah, it's— I think you're right. I think it's— most things shouldn't work in biology, maybe, but they do. And so that gets into maybe like, like why, why, why not? Why maybe, maybe how do we just uncover these rules then? That's kind of that thing into developing better tools. And so you went to Washington, went to labs in Washington, right?

Yeah, yeah, that's right.

Washington. And so, you know, you're making that transition from undergrad to grad school. How did that go? What was the logic there? Did you, I think you went straight to grad school.

No, I took 3 years off.

Oh, you did? That's why you're so good. No wonder. Exactly, you gotta take a break. You can't go straight to grad school. You gotta— so what'd you do in between?

Uh, yeah, I mean, it's good to get more experience, and that really helps when you're trying to get into grad school. They really want a lot of research experience. But aside from that, it's good to try something different, you know. You, you work in a lab, uh, for a year or two in your undergrad, and then that's really all the research that you know about. But it's good to try something different and ask, you know, do I want to go in a different direction? Do I want to stay in relatively the same field? After undergrad, I worked in David Rawlings' lab, also in Seattle, the Seattle Children's Research Institute, and I was still very fascinated by cell differentiation programs. So in that lab, it was more of a cell engineering lab. So I used ex vivo culture protocols and gene editing protocols to try to differentiate primary human cells to make therapeutic products. And yeah, that, that was also a great time for me. And I just decided, you know, I've gone the very basic biology route for a while, and then I've gone the, you know, more translational— or not translational, but more applied research route for a while, and I wanted something in between. I wanted some something that addresses, uh, basic biology questions but would also have the potential to be applied in the future, um, in therapy or just in any way that would improve people's lives. Um, so that's what got me into cancer.

Cool. And Howard Chang is the best, you know, one of the best in the world for that. Um, so good for you. It's okay. You joined Howard Chang's lab, and Howard is He's well known for so many, so much stuff. Genomics, long non-coding RNA, cancer. And I never met the guy, but he seems like a sweetheart from his talks. He's a really nice person, which is pretty rare. And so, yeah, how did you— what was the kind of the initial projects in the Chang lab? And how did that build up? We can talk about the Nature paper. How did that build up to that, at least your first Nature paper?

Yeah, I joined the lab because I was mainly fascinated by the idea of gene regulation by the non-coding genome. So the idea that you have your DNA, which is the source code, which makes, you know, which have the protein coding genes, which get transcribed into RNA and translated into proteins. But most of the genome is not coding. So what is it, 98, 99% of the genome is non-coding. And it's not junk, you know. It has all these regulatory elements such as enhancers that interact with the coding genes, which then decide what genes are active and what genes are not active. And I just felt like there's a huge space there to explore, you know. And when I came to the lab in 2019, there had been a couple of high-profile papers on extrachromosomal DNA that had just been published in that year or the couple years before that. Suggesting— well, actually, this is— we can talk a bit more about this field, but, um, you know, the idea of oncogene amplification is a very important one in cancer biology. It's been known for a long time that, you know, oncogenes, um, provides a selective advantage to cancer cells and allow them to, uh, grow and proliferate. And sometimes one way to make more of the oncogene RNA is just amplify the gene copy. So if you have 50, 100 copies of the oncogene, then the cell is going to have selective advantage and it's going to be able to grow and expand. And it's also been known actually for decades that this sort of oncogene amplification can happen outside of chromosomes. So normally, you have the 23 pairs of chromosomes, 46 chromosomes. But you can have— in a lot of cancer cells, the genome is so messed up that you could have extra stuff. So this extra chromosomal DNA, or ecDNA, we're referring to clonally selected circular molecules that are megabases in size that carry oncogenes or really any drug resistance genes or any genes that provide selective advantages. We see them fairly frequently. I think ecDNA has been observed in something like half of all the human cancer types and a quarter of all cases. It can be much more frequent in certain cancer types. Types like glioblastoma or neuroblastoma. So, um, when I— but at the time I joined the lab, it seemed like a very interesting field. There was a couple of high-profile papers, but a lot of the biology was still unknown. So for example, at the time it was described that, um, ecDNAs, uh, are more— have more, um, have higher chromatin accessibility than expected. They allow higher transcription of oncogenes, but it's not well understood why that's the case. And so my initial project really was find out what is different about ecDNA compared to chromosomal DNA. And so we started from an imaging approach. The idea was just to try to validate this observation that ecDNAs make more of the oncogene transcript. So the idea was you take cells with ecDNA and then you do something called fluorescence in situ hybridization, which detects simultaneously DNA copies and RNA transcripts in the same cells. And then you image the cells and ask where things are coming from. And we did target both ecDNA and the chromosomal DNA and found that there's way more transcripts coming from ecDNA compared to chromosomal DNA. But what was fascinating was that most of the transcriptional events were happening within clusters of ecDNA. And so that's what started this whole project that led to the paper.

Yep, I think the key breakthrough of your research, the Nature paper, and I'm sure you have a lot more, is that like ecDNA, these extrachromosomal DNAs, you know, if they bunch together like in 10 to 100 between 10 to 100. I don't know that there's some sort of relationship in the paper, I forgot, but there's definitely a massive— not a massive, but kind of a significant increase in gene expression if they form these communities or hubs. And so maybe take a step back in terms of like the field of ecDNA. And I think the last time we spoke, you're doing some research that's like one of the most hottest fields in cancer biology right now. So what you're doing is really valuable. This, you know that. So just to plant that seed, you're not only doing research that's important, but really valuable. So that's great to hear. You're gonna do— you can post a lot of papers, but hopefully you'll do very well too for yourself. And that's— I'll be sure to help you there. But, um, let me step back about ecDNA and where the field has been, because it's one of these old fields that has become new. Yeah, well, really decades, maybe a century. If you had a microscope, you can And yet you see cancer cells, you can see these kind of high enough resolution, you know, see that X-chromosomal DNA are in cancer cells at a higher rate. And there's chromosomal abnormalities in cancer. It's one of these hallmarks of cancer. And so what changed recently to make it easier for you, maybe others, to actually study I don't know what to call these structures or whatever, these nuclear structures, and map out their architecture and function. That seems to be kind of a recent change the last few years. What do you think's— what's enabled you to do this type of work and maybe others?

Yeah, that's a great question. Like you said, it's a very old field that has a lot of renewed interest. EC DNA molecules were first described, I think, in the 1960s. Back then, they didn't even— they weren't even sure. I think they eventually figured out that these are actually chromatin bodies, but at first, it was not immediately clear that these dots that you see on chromosome spreads are not just contamination, right? You have your chromosomes, which have the characteristic ribbon-looking structure, and then you have these dots that are outside of chromosomes. They eventually figured out that these are chromatin bodies, and then subsequent work in the next decade or so beautifully deduced this logically that because these structures seem to have been selected in a lot of tumor samples, they must— they provide— they probably provide a selective advantage, and they probably contain genetic elements that allow cells to proliferate more. They didn't even know about oncogenes at the time. They just thought there's extra bits of chromatin, and it probably provides a selective advantage. So it's really beautiful how people could figure things out like that. And then since then, there has been a lot of groundbreaking work on just characterizing the features and frequencies of ecDNA in cancer. But I think the reason why there's been so much interest in recent years is because now we have all these sequencing tools, imaging tools that we can use to really describe these structures with much higher resolution than before. And what we find is that there's a lot of heterogeneity in the structure and copy numbers of of ecDNA within a cancer cell population that is really important for driving cellular heterogeneity in a tumor. And so to capture this high level of heterogeneity, you're looking at DNA sequences that are rearranged in complicated ways. You're dealing with data coming from a large number of cells. Really just having the computational tools and having these sequencing technologies helped a lot in answering a lot of these questions.

And so when you think about the research you did, what were some of the key tools, the workhorse tools to kind of discover these kind of ecDNA hubs? I'm assuming single cells, I'm assuming just sequencing in general, but there's also You did CRISPRi screening at the end to validate, but what kind of— get to like lay the line of your toolkit here.

Yeah, yeah, yeah. I do a lot of sequencing, uh, from, from short-read sequencing to long-read nanopore sequencing, um, and then, yeah, functional perturbation like CRISPR interference, uh, to, you know, try to— you say something about the functionality of a genetic element To prove it, you have to perturb it and say, OK, this is what changes. So all these things are, like you said, workhorse tools for me. We had a big imaging component in that paper describing the clustering and just general spatial distribution of ecDNA molecules in the cell nucleus. But really, that speaks— actually, that speaks to how important teamwork is in science.

Because you get TGen on the paper, if I remember correctly.

Yeah, yeah, because that's what they're good at. They do high-resolution imaging, and they combine, you know, all these modalities like RNA, DNA visualization. And we don't do that. So if you look at the paper, you'll say, wow, there's like 20 different technologies that are used here. But also if you look at the author list, there are probably 50 people on the author list. And really it's because it's impossible for one person to do all of those things. And I think that's probably illustrates a very important point, which is if you want to do robust, good science, I wouldn't say it's impossible to do it alone, but it's often much easier to do it with a bunch of other experts. And not only can you incorporate their expertise, sometimes just by talking to people who are good at different things, you come up with new ideas that would not otherwise come to you.

Yeah, I totally agree. I think collaborating with people who are the best at what they do is really awesome. So you can find people who are the best microscopy people and then collaborate with the best cancer people, and that kind of becomes very fruitful. Could you maybe describe for the paper like what the key experiment was? Like we had a broad overview of like, you know, you had to bring all these tools together, and I'm assuming all these different types of people, you have to take these samples and sequence them and image them? What was that kind of like, that supply chain? How did you go from a cell line and then to a readout? What was the kind of the steps there?

Yeah, that's a very interesting question. We had, I would say, we collaborated with probably 4 or 5 main groups.

That's a lot. A lot.

It was a lot. So, so there, even just for one project, it grew to a point where we would have almost subgroups that would be definitely to, you know, talk about experiments that would have to be done to, to achieve the same overarching goal, but different subgoals. So we had, you know, a whole group of people who focused on imaging, a whole group of people who focused on molecular chromatin conformation capture, which is looking at genomic loci that are in contact with each other in the nucleus. We had a separate subgroup with people who make computer software. So they— because a lot of this— Much of the study was very data intensive. We had to customize some computational workflows to deal with the amount of data. And there are people who make very robust, useful bioinformatic tools to deal with this kind of data. So we had a whole group for that. So it was a complicated structure, but it was very important for trying to do all these things that are each individually so complicated.

That makes sense. I think image files are really large. I think they're bigger than freaking sequencing files. So geez Louise, it's a lot of work. Maybe we can talk about then, like, you have this ability to sequence ecdna, image them as well, and figure out where they cluster. What did you discover? You really made a really fascinating discovery around not only these hubs, but then these tethers. Yeah, there's certain proteins that tether these hubs together. So if the ecdnas are clustering together, they're doing so via these tethers. Maybe you can talk about what these tethers are, what they do, potentially do. You know, you used chemical perturbation as well as genetic perturbation to kind of suss this out. Maybe you can talk about— maybe transition towards the hubs and why they're important, especially in disease, especially in cancer.

Yeah, yeah. So the main point of the paper was that ecDNA clustering seems to be very important for driving oncogene transcription. So when you have strong clustering, you get more transcriptional output. And then when you disperse these hubs, you get less oncogene expression. So the one aspect of the study was we use, like you said, chemical perturbations to target a protein family called the bromo— domain protein family. This family of proteins, specifically BRD4, has been shown to be specifically important for the Myc expression of the Myc oncogene. So we found out that when we perturbed these proteins, it led to not just a decrease in Myc expression coming from ecDNA, but also dispersal of ecDNA hubs. So the idea is, well, if clustering is important for transcription and you break up clustering, you can also decrease transcription. Could this be used as a therapeutic opportunity? So in theory, if this is universal, you could theoretically target the tether, the protein tether that's holding the ecDNA hubs together. And by breaking up the ecDNA hubs, you would decrease oncogene expression, which would be bad for the cancer cells but good for you, the cancer patient. But I would say it's a very promising concept. What probably people have to figure out is how universal this is. So is this true for all cancers that have ecDNAs? Is it true only for a subset of oncogene amplicons like MYC? It seems it's important. But there's also other oncogenes like EGFR, FGFR2, which have also been shown to be frequently amplified on ecDNA. Is there any specificity to the oncogene itself? And then also, let's say if this is a universal phenomenon, we also have to figure out, is the protein tether universal or is it also specific to different cancer types? In this case, it's the bromodomain proteins or BRD4. It might be a different protein in a different cancer type or a different oncogene family. So I think it could be a very promising therapeutic tool, but there's more work to be done. And yeah, there is a lot of work that's ongoing now in our group. That's trying to address these problems, these questions. But yeah, we're all very excited about this.

I mean, you just published a paper last year, so I would be very surprised if a drug gets created in the next few years. It's going to take a long time, at least a decade, to begin to answer all these questions you laid out of the context of these hubs and these tethers. But at a high level, It's a pretty important breakthrough in cancer because a lot of these— Myc, for example, is undruggable right now. And so if you can find a new strategy to disrupt Myc expression in cancer selectively through these tethers or to ecDNA, that's really exciting and can solve a massive problem because, you know, if you can— probably the biggest opportunity in cancer is the undruggable genome. Is there's these oncogenes that are really hard to drug. And so if you can find indirect ways to pursue them, that's going to help a lot of patients. And so, yeah.

No, that's a great point. I mean, their proteins are undruggable, but now you're targeting upstream of that. You know, you find out something abnormal about the source code.

It's selective though. That's being selective. That's a— eCDNA is a hallmark of cancer, at least a lot of eCDNA. So if you can like begin to under— research you're doing is really important to characterize the architecture, so to speak, of these extra-chromosomal arrays and then begin to— that's what— you know, Howard Chang's known for that, right? Structure-function relationships using genomics, right? He's like the master of it. So I think there's— It's like a decade's worth of work here, probably. But maybe you can also talk about— maybe you did this research. When you're doing this research, often good work creates new problems. And so what kind of new problems did you unveil with at least that Nature paper? And maybe it led to this new paper you're having getting published soon.

Yeah, that's also a great question. Um, um, some of the questions that it raised, definitely include those things like how universal these features are, and if they are universal, can you find the universal factor that's tethering the ecDNA hubs? If they're not universal, could you find out more about the the relationship between oncogene structure, oncogene amplicon, and cancer type to the formation of ecDNA hubs and utilize that information to better inform what to do therapeutically. That's definitely a big question that would have to be answered next. And then also from this study, What we found was there's just a lot of structural diversity in the repertoire of ecDNA molecules, even in the same cancer sample. And then also, we tried to use a lot of sequencing techniques to identify these structures, and it was very difficult because not only do you have a lot of diverse ecDNA species containing different genetic elements that are coexisting in the same sample, you also have chromosomal DNA still. And you have to somehow figure out what signal is coming from ecDNA and what signal is coming from chromosomal DNA. And so that led to the next part of my work, which is finding a way to isolate and separate EC DNA molecules from chromosomal molecules and using that as a tool to profile EC DNA in a more targeted way. So that's what I've been working on in the past 6 to 9 months or so, is finding a method that would allow us to do that. And it turns out to be Very simple, and I can talk about it a little bit. So this idea came from a few papers that were published in the '80s where people described cancer cells containing ecDNA. And so they— what they did was they took the genomic DNA and just ran it on a gel. So this is similar to your normal DNA electrophoresis, except this is pulse field electrophoresis, which allows you to separate larger pieces of DNA. But they ran the DNA on the gel and found that the DNA, when it's intact, gets trapped in the well. But if you bombard it with, uh, a lot of radiation, it runs into the gel, which makes sense because your DNA is getting fragmented. But if you have a circular molecule it runs into the gel and shows up as a very distinct clear band. And that makes sense because if you bombard it with radiation but it's breaking the DNA infrequent enough, most of the time you'll have a single break in the circle, right? So a single break in the circle would lead to a single size of DNA fragment,, and it would show up as a distinct band. So what this proved was when the circle is intact, it gets trapped in the well, but when you break it, it runs into the gel. So the idea was, now we have CRISPR, so instead of just using radiation, we could use CRISPR to make a cut on the circle, and then the circle would run into the gel, and you can extract it, and you can do all sorts of genetic and epigenetic analysis on on these circles. So we were able to optimize this method, which is called CRISPR-CasH. And you can use it to just find genetic mutations that are specific to ecDNA. You could also look at things like DNA methylation. And if you have multiple species of ecDNAs with different structures, they would show up as different bands. So you could extract the different bands and identify all the complicated structures that are coexisting in the sample. And so I think I'm really excited about this study because I think, you know, it really opens up the window for a lot more targeted analyses on yeast DNA.

You probably find more diversity than people expect, right? I'm assuming that's gonna happen.

Yeah, yeah.

Chromatin accessibility modifications.

Yeah, cool. Yeah, yeah, exactly.

She's a weed. You know what, your research only creates so many other research projects. It's like, it's like, oh, you gotta do paper. It's like, well, now we have a decade's worth of work now, so Good research, right? Good research creates a lot of new problems.

Yeah.

So you're like a professional problem creator. You just research and then you create problems for other people and then they have to solve them. And so maybe on that note, how do you think about the broader field of genomics, extra-chromosomal DNA? What excites you? Do you see yourself kind of focusing on ecDNA or do you have maybe go into non-coding genomics in general, and like long non-coding RNAs and small RNAs? How do you think about your career and also the field in general and how you want to fit in over the next maybe 5 to 10 years?

So I think my past— the past 3 years or so of my work on ecDNA has gotten me really excited about the idea that DNA is not just a two-dimensional molecule, right? It's not just ATGC from start to finish. It's a three-dimensional molecule that is organized and folded in a very specific way in the nucleus. And ecDNA is just an abnormal form of this, where now you have a piece of DNA that breaks off from the normal site and is now somewhere else in the nucleus. And this change of spatial organization has a lot of implications on how gene expression and really probably many other important cellular processes are regulated. And so I think in the long term, I am really excited about just generally asking how the spatial organization of the cancer genome is altered and how this altered spatial organization changes fundamental cell processes like gene expression, DNA replication, DNA repair, and I think there's a lot of directions I can go, but generally I think this is a very exciting direction for me.

Yeah, maybe it's like connected to your college research where it's like getting fascinated in development, and you know, I've done fish experiments before and they're pretty cool, I'm gonna be honest with you. It's like, whoa, what a cool experiment, like things light up and you start seeing patterns, and it's cool when you do like a time-lapse video of something, especially cell line. Or I remember doing like multistaining. It's like, whoa, look at all the colors. And I can see why you got it. I got it not as much as you did, but it was fascinating to me. And so maybe in cancer, you're kind of doing something similar. You're mapping out kind of these core components of cancer development and maybe uncovering parts of the genome that people think are hidden or dark or junk. And connected them to cancer and metastasis of cancer, the growth of cancer. And that ultimately will probably have an impact on patients. Like what you're doing sooner or later will lead to something valuable for patients, whether it's a drug or something else, because there's often these like, you know, ecDNA is one of these fields, one of these old fields. I'm always fascinated with old fields that like are new. And so like X-chromosomal DNA is in that bucket of something that's known for decades, but only recently has been kind of become better understood. And so like if you had to like give a perspective or some thoughts on like what do you think the challenges would be to translate this research into products?

Yeah. I think there'll be— there'll need to be a lot of screening, right? So I think, let's say, let's say we find that there's just something fundamentally different about gene transcription on ecDNA, maybe because of the spatial organization of it. Maybe it's because of the inherent circular structure of it. Maybe it's because of how the amplicon was formed initially. But let's say there is something just unique about the regulation of transcription. What you have to find, what you really want to find, is unique transcriptional dependencies of cancers that have ecDNA. What that probably means is that you'll have to do a lot of chemical screens, do a lot of CRISPR screens, and figure out, you know, just what is— what do ecDNAs depend on? Like, are there things that are important that when you take out, it will target those specific cancers, right? So, and then in parallel I think there's— you know, nowadays you see a lot of genomics papers that are trying to characterize large numbers of patients and their genomic alterations and gene expression that are correlated with different therapeutic outcomes. And the same has not really been fully done on ecDNA. And part of that is because it's not just about digging the genomics data. It's also figuring out when you have amplification of an oncogene, where is it amplified? Is it extrachromosomal? Is it tandem repeats? I think what the past few years of research has shown is that the context of genome alterations really matters. Yeah. And so trying to incorporate that into large-scale studies that also include looking at patient outcomes in response to maybe different chemotherapies, immunotherapies, I think, I think those studies will really answer a lot of these open questions.

That's a lot, a ton of work. I think you're You're kind of spot on though. The context matters here, whereas your research is kind of laying the groundwork in the series of principles of ecDNA. But, you know, it's biology and everything's context dependent. Depends on the disease, it depends on the cell type, it depends a lot of things. So a ton of work left to be done, and probably you'll start your own lab one day and, you know, you'll be pursuing these problems. I'm assuming there's a bunch of other people working on this stuff too. But cool, I think it's a great conversation. Thanks for doing this. I think it's gonna be really useful to a bunch of people. But it's a sunny Friday right now. It's Friday. We'll probably publish this next week, but it's Friday. It's sunny down in Palo Alto right now. Yeah, so do you have any lab— I'm assuming some lab experiments or wrap up. But, you know, it was a great chatting and, you know, have a great Friday and we'll talk soon.

Yeah, this was very fun. Thanks for having me again. Absolutely.