Stellar Lab Radio Episode 4 Guest：Nozomu Yachie
When does life take shape, and through what processes does it grow into something so complex?
Can we truly say that we are seeing what is happening inside living systems?
To study cells and organisms, life science has long relied on destroying, stopping, and capturing a single moment in time. In many ways, the field has advanced through such “snapshots.”
But what if we could record life as it continues to move—without breaking it apart?
If that were possible, an entirely new understanding of life might emerge.
In this episode of Stellar Lab Radio, we welcome Nozomu Yachie, Professor at the University of British Columbia and Specially Appointed Professor at Osaka University. Professor Yachie is pioneering research that aims to record the inner workings of life over time by building a “video camera” inside cells and using DNA itself as an information medium.
In Part 1, we begin with a fundamental question: why has life science required destruction in order to observe? From there, we explore the origins of the groundbreaking concept of DNA event recording, Yachie’s cross-disciplinary approach to assembling a “camera for life,” and the long-term vision that drives his research.
Together, we examine how the idea of recording life may transform the way we understand what life truly is.
Listen from here (in Japanese)
A Fundamental Challenge in Life Science: Why Must We “Destroy” Life in Order to Observe It?
Yachie: I want to create technologies that have real impact. One fundamental problem in life science is that, for example, mice are widely used as experimental animals, but if you want to study a mouse’s brain, you have to kill the mouse.
Once you do that, you can capture a snapshot at that moment. But if you want to see what is happening inside cells, you have to destroy the cells as well. At the instant they are destroyed, you obtain extremely high-content information. However, you can never observe what was happening before that moment, or how the mouse might have developed in the future if it had not been destroyed.
It’s a bit like Schrödinger’s cat in quantum mechanics. You can observe what happens the moment you open the box, but you can’t see what came before or after. That may sound obvious, but I think this is a fundamental limitation of life science.
To overcome this, I began thinking about putting something like a video camera inside cells, so that we could observe what is happening inside them as they continue to live. In fact, even now, recording is happening inside cells.
By using a “lens” to detect events and writing that information onto a kind of hard disk, it becomes possible to create cells that continuously record what is happening within them.
If such a cell were a fertilized mouse egg, and it developed into a mouse, we could understand how a mouse emerges from a single cell over time. Ultimately, within ethically acceptable limits, we could extend this approach to other animal models, such as monkeys or cows to study the complex processes by which human and animal bodies are formed.
I believe that developing this kind of technology could be incredibly useful for many people, which is why I’m working on it.
Kokeguchi: I see. Thank you. That’s fascinating. Let’s dig deeper into this. One of the key words you mentioned earlier was “tool-making.”
When people hear “biotechnology,” they often imagine developing new drugs or machines. Is it correct to say that your focus is on creating tools that other researchers can use?
Yachie: Yes, that’s right.
Kokeguchi: Is this a large, established field, or is it something that you are leading yourself?
Yachie: When I started thinking about this around ten years ago, there were a few researchers, both in my lab and in labs in the United States, who began exploring similar ideas around the same time. The community is still small.
The challenge is that even if the concept is feasible, it takes more than a decade to realize. That makes it hard for people to jump in. But I genuinely believe this is cutting-edge work. It’s like launching a rocket.
Kokeguchi: Wow, that’s impressive. Which do you think will happen faster: this, or humans living on Mars? (laughs)
Yachie: This will be faster. (laughs)
Cells as “Video Cameras”: The Concept of DNA Event Recording
Kokeguchi: You mentioned “video cameras” as a second key concept. Does that mean recording information inside cells, using something like a hard disk?
Could you explain a bit more about what that actually involves?
Yachie: The human body contains about five trillion cells that carry DNA. DNA is made up of four letters—A, T, G, and C. Computers use sequences of 0s and 1s to manage information storage, right?
Living systems use sequences of ATGC to encode the blueprint for what a human body will become, or what a mouse’s body will become. In that sense, DNA itself is an information storage medium.
So we started thinking about using DNA as a recording medium, writing observed information directly into DNA as it happens.
Kokeguchi: I see. So you think of DNA itself as a medium. That’s a fascinating idea.
Yachie: For example, the human genome consists of about three billion characters, which is roughly 750 megabytes. That’s remarkable. Within those 750 megabytes is written what will happen to you—what kind of person you will become.
All the instructions from the moment you are born, from a single fertilized egg to a fully formed human being, are contained within those 750 megabytes.
And when you think about it in combination with the laws of physics, it’s astonishing. An iPhone today has, what, around 250 gigabytes of storage? So the information that makes up a human being is about one-hundredth of that, or about one-hundredth of what’s stored on the phone in your pocket.
Kokeguchi: So that means it could store information equivalent to hundreds of people.
Yachie: Exactly. That’s right.
Kokeguchi: That’s really interesting. Is the idea of using DNA as a storage medium already widespread, or is it something more novel?
Yachie: If my memory is correct, back in 1999 a high school student collaborated with a professor at New York University and published a paper in Nature, one of the top scientific journals, showing, for the first time, that artificial messages could be written into DNA.
It was quite sensational at the time. But back then, technologies for reading and synthesizing DNA were still very limited, so people mostly saw it as a fun experiment like, “That’s cute, a high school student did this.”
Later, when I was a master’s student around 2005, I worked on writing artificial information into DNA in a more robust way. DNA mutates easily, so information can quickly be corrupted. About eight years later, I published a method for writing information stably into DNA so it wouldn’t break down. That work got a lot of attention at the time—we even had interviews with outlets like The New York Times.
Kokeguchi: Wow.
Yachie: After that, I got bored with the project. For me, writing artificial information into DNA had been something of a side project, almost playful, so it went dormant for a while. But when I started my own lab in 2014, I realized that with the technologies available at that time, it might finally be possible to take this idea seriously to write artificial information into DNA and truly apply it to biology. That’s when I decided to start again.
Kokeguchi: That’s fascinating. The high school student back then would be, what about 25 years older now, probably in their mid-40s. Do you know if they’re still doing research?
Yachie: I have no idea. (laughs) It was probably just a summer vacation project.
Kokeguchi: A school summer project… that’s incredible. And from that idea, you’ve gone on to create an entirely new field.
Yachie: Yes, that’s right.
Building Observation, Recording, and Readout as One System: Molecular Sensors and the Architecture of a “Camera for Life”
Yachie: Yes.
Kokeguchi: Going back to one of the earlier keywords, you mentioned that at present we can only take snapshots by destroying cells. That was somewhat surprising to me. I had assumed that many experiments involved careful observation, but in reality, destruction is unavoidable.
Yachie: That’s right. Of course, you can observe cells growing in a culture dish. But for example, the human genome encodes about twenty thousand genes, and each of those genes produces proteins.
If you want to study how those proteins are produced, you would need to observe all twenty thousand molecular species inside the cell. To do that, you ultimately have to break the cell open.
There are observations you can do without destroying cells, but they often yield limited information. If you want to obtain very high content data, you usually end up having to destroy cells or even sacrifice animals.
Kokeguchi: So to avoid that, you are building something like a camera lens and a recording medium that allow observation without destruction. If the recording medium is like a hard disk, represented by the ATGC code, then what kind of lens are you talking about?
Yachie: It is not a lens like a camera lens. What we use are molecular sensors. These sensors interact with specific molecules inside the cell, change their structure in response, and then rewrite DNA accordingly. That is the kind of sensor we are building.
Kokeguchi: I see. So by reading how the sensor reacted, you can tell what happened inside the cell.
Yachie: Exactly. To continue with the camera analogy, a camera has a lens, a detector, and a memory. The observed information has to be written into memory, so you also need a writing mechanism.
Finally, you need a readout mechanism that reconstructs the image from the bits written in memory. I think all four of these components are necessary inside a cell as well.
Kokeguchi: So it’s like building an entire film studio inside a living system.
Yachie: Yes. In Japanese, we would say it is built end to end. This is not research focused on a single element. It is about packaging. We take technologies that many brilliant researchers have developed over time and package them together to create a camera. That is essentially what this research is about.
A Turning Point at 2 a.m. in the Library: How an Encounter with CRISPR Defined the Direction
Kokeguchi: I see. So there are different people for each part, someone for the camera side, someone for sensing, someone for writing.
Yachie: For example, there is genome editing. Genome editing is expected to correct disease related mutations, such as those involved in cancer, by rewriting altered DNA sequences back to a normal state.
Originally, these technologies were developed to rewrite DNA in living organisms. But when you attach a sensor to a genome editing tool, you can rewrite DNA in response to what the sensor detects. It may be hard to grasp just by saying it, but that is how we are using these tools.
The moment that really made me want to pursue DNA event recording was in 2014. You know Hiroshi Nishimasu, right? Professor Hiroshi Nishimasu at the University of Tokyo.
He published a paper in Cell describing the structure of CRISPR Cas9, together with Professor Ryuji Nureki and Feng Zhang. I happened to read that paper and thought, since we are from the same generation, that there are truly incredible people out there.
At the time, I was living in Toronto as a postdoc. The apartment building I lived in had a small library. I still remember this clearly. Around two in the morning, after coming back from the lab, having dinner with my family, I went down to the library and opened that paper.
The paper described the structure in detail, and to elucidate it they had carried out many careful engineering steps, using slightly modified RNA molecules and various experiments. I read that paper about five times over the course of that night.
That was when I realized that if something like this was possible, it could be built into artificial cellular circuits, and DNA event recording could actually be done.
Kokeguchi: That sounds like a powerful moment. That was in 2014?
Yachie: Yes. I was so excited thinking, “This could work.” I spent the entire night reading Nishimasu’s paper.
What’s funny is that I didn’t even know him personally at the time. I had never met him.
Kokeguchi: Did you just happen to come across the paper?
Yachie: Yes, I happened to find it in Cell.
Kokeguchi: How did things develop from there?
Yachie: I realized this might be possible while I was still a postdoc. A few months later, I got the opportunity to start my own lab. I returned to Japan, wrote grant proposals, received funding, and decided to give it a try. Not long after that, I met Nishimasu for the first time. I finally saw the person I had admired so much, but I did not tell him then that I had read his paper obsessively. I only confessed that recently. (laughs)
Kokeguchi: How did he react? (laughs)
Yachie: He seemed really happy.
Kokeguchi: It must be gratifying to know that someone read your paper so deeply.
Yachie: Absolutely. I think anyone would be happy about that. And later I learned that he had poured a huge part of his youth into that work.
Kokeguchi: Did you end up collaborating?
Yachie: Many times. We are still collaborating now.
“That’s Incorrect, Right?”: How a Chance Encounter Accelerated the Research
Kokeguchi: So that covers the sensing and writing components of the system. How did you meet the other people involved, and what led them to join the project?
Yachie: The readout part of the memory involves developing new algorithms. After decoding DNA sequences, we use supercomputers to reconstruct the recorded information.
Kokeguchi: You use supercomputers?
Yachie: Yes. More precisely, it involves parallel computation, using many computers at once to solve large problems. These are problems that a single computer cannot handle, so you might need hundreds or even thousands of machines working together. I already had a rough idea of what kind of computational architecture would be needed.
Around that time, I happened to be teaching an undergraduate class at the University of Tokyo, for third year students in the Faculty of Science. After one lecture, a student named Naoki Konno came up to me and said, “That’s incorrect, right?” I replied, “Oh, you’re right. Sorry about that.”
I started wondering what kind of student he was. After the final exams, I checked his score and saw that he had gotten a perfect score, even though the exam was designed so that getting a perfect score should be almost impossible.
I thought, “This guy is incredible.” I asked around to get his email address and basically invited him to my lab to work with us over the summer. About six months after he joined the lab, he had already built the system.
Kokeguchi: That’s amazing. Was his background more in physics or mathematics?
Yachie: At the time, I think Konno had only just started learning programming. But now he is at Stanford and has become a real superstar. I thought he was impressive when I invited him, but he turned out to be even more impressive than I expected.
Kokeguchi: Incredible. It really was a chance encounter. He caught your attention by pointing out a mistake, and of course the perfect score must have stood out as well.
Yachie: It really caught my attention. It left a strong impression on me.
Kokeguchi: By the way, what was the mistake he pointed out? Was it a simple error, or something deeper?
Yachie: I was teaching about the mitochondrial respiratory chain, and I had gotten the number of electrons flowing through the chain wrong.
Kokeguchi: That’s a very detailed point.
Yachie: Yes.
Kokeguchi: It also takes courage to point that out to a professor.
Yachie: Absolutely.
Kokeguchi: You mentioned supercomputers earlier. Is AI involved in this work as well?
Yachie: Not yet. But I think it will be soon. The framework that Konno built is starting to look like it could be useful for AI applications as well.
Running a Lab Like a Startup Founder or a Designer?
Kokeguchi: That’s fascinating. So by integrating and packaging many different research fields, you are essentially building a “video camera for life.” At this point, what do you feel is most lacking, or where are you focusing your energy the most?
Yachie: What’s probably lacking the most is my own stamina. The reason is that while we are trying to build a project that may take ten years or more, the students and researchers in the lab are constantly growing and moving on to their next stages. There is a kind of metabolism within a lab.
Because of that, I need to keep the overall vision firmly grounded and clearly in mind. Every time someone new joins, I explain the vision and say, “Based on what you are good at, I’d like you to work on this part.” They may not get to see the completed version of the camera for life, but by contributing to a specific component, that work can still lead to new discoveries or new products. Those pieces can be published and stand on their own.
To make that work, I feel I have to keep track of everything and hold the full picture together.
Kokeguchi: That sounds a lot like a startup CEO, or a founder.
Yachie: Does it? I’ve never actually met a startup CEO. (laughs)
Kokeguchi: Startup founders usually have a very strong vision of what they want to build. Of course, core leaders often stay for a long time, but many of the people who help build things together come and go fairly quickly. That makes it essential to constantly communicate and reinforce the vision.
Yachie: And the vision has to keep being updated. In our field, new technologies appear one after another. In our lab’s Slack channel, people share about ten new papers every day. Even skimming through them and keeping them in mind is already a huge task.
Kokeguchi: In other research areas, such as drug discovery, the image is often very focused, like “we want to fix this specific part of the liver.” But in your case, you are looking at many different components and assembling them into a single system. If a new CPU appears, or a new writing technology is developed, it all matters. The scope you have to watch seems extremely broad.
Yachie: And sometimes you realize that while you were working hard on one component, something far superior has suddenly appeared elsewhere.
Kokeguchi: Does that happen often?
Yachie: All the time. And the people most affected are usually the students who are directly involved in developing those components. When that happens, part of the job is to make sure their work is not wasted. That might mean changing direction slightly, pivoting, and redesigning the project so it can create new impact.
Kokeguchi: That sounds much closer to the role of a designer.
Yachie: Yes, it feels like creating concepts.
Kokeguchi: Listening to you, I was reminded of a company like Apple. They have created innovative products like the iPhone, but they do not manufacture everything themselves. They travel around the world to find the best components, use glass from Japan, chips from Taiwan, and so on. Since everything keeps evolving so quickly, it is never obvious what the optimal solution is.
Yachie: That’s exactly right. It’s about combining the best technologies from around the world.
I Hate Creating Boundaries: A Stance That Refuses to Be Defined by Labels
Kokeguchi: Earlier you mentioned biology. Do you see your work becoming a new field? It feels like it crosses many disciplines. You have also used the term “super biology” before. What kinds of boundaries does it go beyond?
Yachie: That’s a difficult question. To begin with, I dislike creating walls or boundaries, whether it’s biology, science versus humanities, or any kind of categorical division. I have rarely thought about what kind of researcher I am.
There are people who like to label themselves, saying, “I am a researcher in this specific field.” I understand that it can give a sense of mental comfort. But once you put yourself into a label, you also become constrained by it. So I try not to define myself that way.
Kokeguchi: Then how do you usually introduce yourself?
Yachie: When introducing myself, I might say that I am a life science researcher. In practice, most of what I do is related to life science. But at the same time, in order to contribute to life science, I end up having to be involved in many different things.
Kokeguchi: That’s interesting. Precisely because you do not stay within a single domain, you are able to create things that cut across fields and are truly disruptive. In that sense, it feels very close to what we at SS F are trying to do as well.
Yachie: Please support us. (laughs)
Kokeguchi: Of course. Looking back, are there any particularly memorable challenges you have faced, and moments when you overcame them?
Yachie: I do not think of my work in terms of challenges very often. The reason is that we are not doing research to make discoveries in the traditional sense. We already know what we want to build. The question is how to design a path and systematically reach that goal.
So part of the job is actually to eliminate challenges before they arise. That means preparing many possible paths and building detours in advance.
I often tell people in the lab, “Let’s do this, and this, and this as well.” It is like a game of shogi, where you think many moves ahead. I want us to invest in multiple directions while considering what might be needed in the future. From the perspective of the people doing the experiments, the reaction is sometimes, “Why are we doing so many unrelated things at once?”
But I do not always have the time to explain everything in detail. So I might say something vague like, “Just try this too, it will be interesting,” or “This might turn into something good later.” It can sound a bit rough. But by preparing options in advance, I think it helps prevent us from getting stuck in a dead end where progress becomes impossible.
At the same time, I may not fully understand what the people on the ground are experiencing. They are at the bench every day, pipetting and running experiments. They face real challenges daily, and when they overcome them, there are moments of joy, like going out for drinks to celebrate. Those moments exist for them.
For me, those moments are no longer there. That part is a little lonely. When I was doing experiments myself, I had those feelings every day too.
Research Cultures in North America and Japan: What Differences in Education, Dialogue, and Research Environments Reveal
Kokeguchi: When do you feel the greatest sense of joy in your work?
Yachie: Without a doubt, it’s when I see people growing. That’s the biggest joy.
Kokeguchi: At what stage do students usually join your lab? As undergraduates?
Yachie: Sometimes as early as their second year.
Kokeguchi: And then they may continue on through graduate school.
Yachie: That is common in Japan, but at UBC and at universities in North America, students almost always change labs when they move on to a master’s or doctoral program.
Kokeguchi: Is that a formal rule, or more of a custom?
Yachie: It is almost a rule. For example, staying at the same university from undergraduate through master’s and PhD, then becoming an assistant professor, lecturer, associate professor, and eventually professor in the same place is considered very unhealthy, even toxic.
There are probably many reasons for this. For instance, when a university hires a professor, they evaluate that person’s achievements. Even if someone has an outstanding record, it could simply be because their former supervisor was exceptional. But if someone has moved across multiple universities and succeeded in different environments, it becomes much easier to see their individual strength. I think that perspective encourages people to move across fields and institutions.
Kokeguchi: That’s interesting. Would you say this is a major difference between Japan and Canada, or more broadly North America?
Yachie: I think it’s similar in Europe as well.
Kokeguchi: So staying in one place for a long time is more characteristic of Japan.
Yachie: Probably of Asia in general.
Kokeguchi: I see. Are there other differences between Canada or North America and Japan that you notice, especially in research culture?
Yachie: Students there tend to throw out ideas more freely, even without studying everything thoroughly first.
Kokeguchi: What do you mean by that?
Yachie: They will say things like, “I just thought of this,” and casually share ideas one after another. They do not hesitate.
Kokeguchi: In a positive sense?
Yachie: Both positive and negative. Sometimes I think, “You could have looked that up,” or “You could just ask ChatGPT.” But still, I think it’s a good thing. I want them to keep sharing ideas freely.
In Japan, students study very thoroughly. Some of them know more than their professors. But they are often extremely shy. Even if they have a good idea, it does not come out easily. Of course, there are exceptions.
For example, in Canada I teach about seventy eight hours per year. I teach three times a week, Monday, Wednesday, and Friday mornings. After every class, about ten students line up at the blackboard to ask questions. They are energetic, and teaching them is a lot of fun.
Kokeguchi: And in Japan?
Yachie: Did you attend a Japanese university?
Kokeguchi: Yes, I did.
Yachie: Then you probably remember how students often look down at their desks, even when professors are speaking passionately. I always wonder what they are looking at. They occasionally glance up, but it feels a bit sad.
Kokeguchi: You really enjoy active dialogue.
Yachie: That student I mentioned earlier, Konno, was one of the few in a class of about thirty who always looked up and listened attentively. There are certainly such students in Japan too. But I’m speaking about the average tendency. Many students are quiet and quite passive. It might sound like criticism, but they are very good at passively accepting what the teacher says and responding, “Yes, I understand.”
Kokeguchi: How long has it been since you moved to Canada?
Yachie: I moved in September 2020, so a little over five years now. This is my sixth year.
Kokeguchi: Where do you think these differences come from? It doesn’t seem like something that starts only at university.
Yachie: When I look at my own children’s school life, I see a lot of clues. My daughter is in high school, and in class they make videos with friends, create presentations about their cultural backgrounds, and share them. My son also makes presentations using PowerPoint and presents them in front of others.
Club activities are also very important from junior high and high school. My daughter plays in a band and an orchestra. These activities value things beyond desk based studying, such as group work and collaboration.
Kokeguchi: You are currently based in both Osaka and Vancouver. Do you see differences in management or research style between the two labs?
Yachie: Yes. The lab at Osaka University is relatively new, about two or three years old. It is a more mature lab in the sense that there are four associate and assistant professors. There are still few students, but the structure has stabilized and we hope more will join.
Kokeguchi: When you say a mature lab, does that mean there are no students?
Yachie: There are students.
At UBC in Vancouver, I am the only professor. There is a lab manager, two technicians, and about fifteen students. I still meet with every one of them biweekly. In Osaka, there are associate professors who operate very independently, which is of course how it should be. I might call and say, “Try this,” and then check in every couple of weeks or once a month to see how things are going.
Kokeguchi: So the Osaka lab is quite autonomous.
Yachie: Yes. And if someone at UBC runs into trouble, I might say, “Ask this person at Osaka, they understand this well,” and encourage discussion on Slack.
Kokeguchi: Are the UBC and Osaka labs doing the same research?
Yachie: No, they are doing different research. I sometimes feel the need to emphasize that. At UBC, the lab functions as a prototyping lab. We design circuits using stem cells and cultured cells and test whether they work.
At Osaka, we focus on putting those systems into animals and seeing how they function inside living organisms. That said, Osaka also works on developing new sensors that we do not build at UBC. Another important part of the Osaka lab is led by Mori, a former student of mine who is now an associate professor. He leads the AI related research module.
Kokeguchi: So in Vancouver, you focus on building and testing, and in Osaka, on actually using those systems.
Yachie: That’s roughly right, although there is also development work in Osaka for experiments inside mice.
Kokeguchi: Do you ever think of yourself as an outsider in terms of research style?
Yachie: I do. But maybe not in a positive way. To be honest, I’m not very good at memorizing things. Because of that, I cannot really do research that builds steadily on accumulated knowledge.
If Google had not existed when I was a university student, I probably could never have become a professor.
Kokeguchi: What do you mean by that?
Yachie: My grades were terrible as a student, except in mathematics. I could not memorize things at all. My memory is genuinely very weak… (to be continued)
Kokeguchi: That brings us to the end of Part 1 of this episode featuring Nozomu Yachie. How was it?
Viewing DNA as an “information medium” offers a fresh perspective and invites us to rethink a fundamental question: what is life, really? It’s a way of seeing biology that can subtly but profoundly shift how we understand living systems.
In Part 2, we will dive deeper into why Professor Yachie chose to become a researcher, the origins of his journey into the world of science, and the passion that drives his work today. We hope you will continue listening and join us for the next part as well.
Stellar Lab Radio – exploring world-changing research that nobody knows about yet.
Listen from here
