3 Takeaways - The Age of DNA: Ginkgo Bioworks Co-Founder Jason Kelly (#100)

Episode Date: July 5, 2022

Steve Jobs once said that the biggest innovations in the 21st century would be at the intersection of biology and technology. Nature offers tantalizing examples of the magical properties of biology—self-assembly, self-repair, self-replication and more. Jason Kelly, co-founder and CEO of Ginkgo Bioworks, shares his dream of harnessing nature by reading and writing DNA to program cells like we program computers.Ginkgo is a synthetic biology company that programs cells for customers in the pharmaceutical, food, agriculture and energy industries.

Welcome to the Three Takeaways podcast, which features short, memorable conversations with the world's best thinkers, business leaders, writers, politicians, scientists, and other newsmakers. Each episode ends with the three key takeaways that person has learned over their lives and their careers. And now your host and board member of schools at Harvard, Princeton, and Columbia, Lynn Thoman. Hi, everyone. I'm Lynn Thoman. Hi, everyone. I'm Lynn Thoman. Welcome to another episode. Today, I'm excited to be with Jason Kelly. Jason's one of the founders of Ginkgo Bioworks.

Ginkgo's an R&D company which uses genetic engineering and synthetic biology to create new products for their clients. They want to program cells as easily as we program computers their work spans diverse markets from food to pharmaceuticals and i'm excited to find out more about biotech and what the next generation of innovation looks like many believe that it will revolutionize both products and manufacturing and that it will be as revolutionary as the internet. Welcome, Jason, and thanks so much for our conversation today. Yeah, thanks for having me on, Lynn.

My pleasure. Jason, what inspired you and your co-founders to start Ginkgo? We met actually as grad students at MIT. We were doing our PhDs in bioengineering, and there were four of us doing grad school together. And then the fifth founder was a professor at MIT, Tom Knight. And Tom was really the inspiration for starting GeekGhost. Tom was a computer science professor at MIT, so not biology, starting in the early 1970s. So I have like a great black and white photo of Tom with his master's thesis, which was this refrigerator-sized mini computer.

So he had gone from programming mainframes to these mini computers, and then he was a computer architect. And so when computers went to chips, he taught the semiconductor design course at MIT for many years, dyed-in-the-wool electrical engineer, computer scientist. Mid-1990s, Tom has this insight

that DNA, the A, T, Cs, and Gs inside your cells, is fundamentally digital code. And we can read that code with DNA sequencing, like genomics, human genome project. And we can write that code with DNA synthesis, DNA printing. Tom was like, well, if you can read and write code, and you have a machine that'll run it, which is sort of how he saw a cell, oh, that's programming. And so he moved, he shifted from computer science, electrical engineering, and started growing cells in his lab in the computer science building at MIT and sort of scaring his computer science colleagues and really shifted his focus. And he came to it with an engineer's mentality. That was really the beginnings of what today is Ginkgo, this idea that, well, if we approach DNA as code and think of it as something that can be programmed, maybe we can make some progress on using cells to address

some of the big problems in the world. So you're programming cells. How do you do that? I'll give you an example, right? Bacteria would have like, call it like a 3 million letter genome. So what is a genome? Well, a genome is just all the DNA that encodes for that organism. Give you a sense of scale, like a human genome is like 3 billion letters of DNA. Okay, so bacteria, much smaller, 3 million, just one cell. And if you wanted to make that bacteria do something new, basically what you need to do is find a spot in its genome

where you can open up the genome and put in some new DNA. And so if you've heard of something like CRISPR, for example, you might've heard of like CRISPR as a new technology. That's basically a way to find a certain location in a genome, like AA, GCCCAA. Oh, I just found that site. Let me open the genome there. And so you open it and then you put in new DNA code. Oh, if you wanted, for example, we have projects to make animal-free meat protein. So things like milk proteins and egg proteins, except without cows or chickens,

you might put in 10,000 letters of new DNA code. And then when you grow that cell up, it'll start to make that protein from the animal. So think of it almost like putting new code on your phone or something. It gives it a new capability and it'll start to run that code. So you start with an idea of what code you want to change. And from there, it's basically trial and error. Yeah.

What we'll do at Ginkgo, our business model is think of it like an Amazon Web Services. Companies come to us with an idea for what they want a cell to do. And our job will be to ultimately design the genome of that cell to make it do what the customer wants. And you're correct. We'll try many thousands or even hundreds of thousands of genetic designs to do it. We don't ever write a DNA code just out of our head. You're really pulling from nature. You're learning how biology does this by going out and reading the code in nature and then finding the parts that'll be useful for this project, putting them together and moving them into a yeast or

bacteria or a single cell mammalian host, depending on what the customer wants us to do. And then you take those cells or that bacteria and what do you do with them? I'll give you an example of the animal-free meat. In that case, the customer would run a facility, think like a brewery. What is a brewery, right? It's a big metal tank and you put in yeast cells, you add sugar and water, basically like food and liquid for the cells and they start growing. And as they grow, in the case of a brewer's yeast, they'll turn that sugar into ethanol. And they'll also make like a bunch of flavor components and all the things you think of when you drink a good beer. What our customer would do in the animal-free meat space is they'd have that similar yeast, except now it would have the DNA code for a meat protein. And so when it grows up in that tank, it actually produces, in this case, a protein that makes blood red called hemoglobin.

It makes that protein and then you add that back into a plant-based veggie burger. And suddenly it smells right, tastes right, cooks right. That sort of ingredient is one of the things that's making these sort of animal-free burgers taste so good nowadays. And so that's an example of a project in the food space. We believe, just like you'd have only one or two operating systems to serve many different apps in computers, or you have Amazon Web Services and Azure to run many different types of websites, we believe you can really have these kind of large platforms to enable you to program sales for any application.

That's the idea. But when you're talking about this as a platform, you're not publicly releasing all of the codes for other people to use. You're more like a consulting firm that has a platform to do consulting. You're doing R&D under contract to different companies. 100% right. The only caveat is when we do these projects for customers, we'll retain some reuse rights on the data and learnings and genetic code so that we can make it available to future customers. That is one of the core ideas behind the company is we want to get better as we get bigger. And so one of the ways we do that is what you just brought up. We have this learning and data that we can share across projects as long as you're on our platform with

multiple customers. And then the other way we do it is the actual way we do that programming of the cell requires a lot of lab work. My background, getting a PhD in bioengineering at MIT, it's like five years of moving liquids around by hand on a lab bench. It's a bit tedious. So what we've done is we've moved a lot of that onto robotics and automation. And once you do that, you get a scale economic. Think like making cars or semiconductor chips. The more of it you do, the lower the cost per unit. That's what we've been seeing. We call it our foundry. That's the other way we improve with scale is like more automation. Can we take a step back?

What is synthetic biology and biotech? Biotech really got started in 1978. There was two fellows, Herbert and Boyer, who developed the original technology for basically cutting and pasting a gene from one organism to another. And the very first products of this were developed by a company called Genentech. And the first product they came out with was human insulin. My dad's a type one diabetic actually. And prior to Genentech human insulin,

you got insulin from pigs. Wasn't perfect, it wasn't human protein, it had side effects. And so what Genentech did was they took the actual gene for human insulin, cut it, pasted it into E. coli bacteria, and then they grew it up in a tank and produced human insulin. And so that was the very first biotech drug. But Lynn, the key is like it was cut and paste. So you couldn't make changes easily to that particular protein design. So it took a lot longer for other types of drugs to come out.

Eventually you got things like growth factor, like EPO and things like that. What synthetic biology is, is building on that to say, I don't want to just cut and paste something. I want to be able to design a new piece of DNA. The core technology is called DNA printing. You go on your computer, and this is God's truth.

You go ATC, GGGG, you hit what you want. You hit print. And out of our labs here or companies like Twist Biosciences in California, you get that piece of DNA built for you chemically. A, T, C, G, G. And then they ship it to you. And then you install it, like I said earlier, into the genome of a cell.

And you can get whatever gene you want from whatever organism you want. That's synthetic biology. It's this larger scale ability to design, in particular using DNA printing to do that. And what is the range that that can be used for? Even if you just look at Ginkgo's customers, I mentioned we're doing stuff in animal-free meat. We also just announced an expansion of a partnership

with a bear, CropScience. This is a big ag company, biggest ag company in the world. The project we've been working on with them, to give you a specific example, is in the area of nitrogen fertilizer production. And then there's tons of work in pharmaceuticals as the other big category where you're seeing like a lot of activity. Those are some of the biggest commercial opportunities are really in those categories. How about drugs? Are the COVID vaccines from Pfizer, BioNTech and Moderna using a similar process? Yeah. So if you listen to like Stefan Bonsall at Moderna talk, he'll explain how they have the ability. I think he said it took two or three days for them to come up with the design for the vaccine once they knew the sequence of COVID-19.

So how? How did they do that? And the idea is, well, once you knew the genome of the virus, you could identify the particulars of the spike protein and things like that. So you had a sense of what it looked like. And then you could essentially design the way the mRNA vaccine works. It basically takes like a little fragment of the coronavirus and shows it to your immune system. And then your immune system turns on and you are ready in the event that the COVID-19 virus shows up to take it out. So that ability to essentially treat the virus as code required DNA sequencing.

And then the ability to design an mRNA vaccine, which is not a small molecule. It's not like aspirin or whatever. It is literally a piece of RNA code that's being used to turn on your immune system. I think there's actually an interesting point here. We think of it as if the vaccine companies are the ones that sort of get us out of COVID. It's not exactly accurate. At the end of the day, what's getting us out of this is our immune system. And what the mRNA vaccine allows

you to do is talk to your immune system in a language it understands. You can now, for the first time, like really talk directly with the biology of your body in the language of mRNA and DNA. And so what we did was we basically went and begged. We're like, hey, immune system, here's a message. Please get us out of this. And it did. It is incredible. So some of the earliest product of biotech you mentioned was insulin. The mRNA vaccines are other products. What is the range of products that now exist from biotech and synthetic biology? I'll tell you the

current big commercial ones, but I'll tell you some fun ones. Big commercial categories are therapeutics. That's the dominant. Like you just said, all your protein, basically your protein-based drugs. So antibodies, things like insulin. Now you're seeing all this genomic medicines with mRNA and gene therapies. I think that whole category is just going to get bigger and bigger. That'll keep growing. Biotech today makes up, I want to say it's about half of new therapeutic drugs going into trials. And then in agriculture, you've historically seen a lot in GM crops. around reducing the carbon footprint of ag. And so things like the nitrogen fertilizer project we talked about a minute ago, but also direct carbon capture. Today, there's not a market for this, but I think you see a pressure coming. And so you are seeing the big ag companies, in fact, we mentioned with Bayer, we're going to be working on a project in this category to say, hey, can we actually do more carbon capture by engineering

either the bacteria or the plants themselves to capture more carbon? So I think you'll see that as the growth area in ag, but that's the other big category. And then you see like a lot of fun stuff. We have projects in the flavor and fragrance industry, for example, to make sense that you would normally extract from plants, being able to make those less expensively or more renewably. We work in the cannabis industry. So I think one of the things, Lynn, that's coming is we saw this with computers.

Like back when Tom Knight, my co-founder, started in computers in the 60s, like mainframes, you know what they were used for? The military, launching missiles, like all these ridiculously expensive things because the cost of compute was unbelievable. But as they improved, nowadays, what do we use computers for? Sending cat pictures, you know, right? Like whatever we want, right? That's what I think people should start wrapping their heads around for synthetic

biology and designing cells is as the cost falls, you're going to see people express themselves with biology. You're going to see artists and designers coming into biotech, and you're going to see new consumer products like a glowing petunia that are biotech products. That's a thing I think people have to start thinking about. I think it's really exciting, and it makes the technology more approachable also. If I asked you, what do you dream of accomplishing at Ginkgo? What is your big dream? So really our mission, our mission at the company is to make biology easier to engineer. And we want to make that so that ultimately as many people as possible have the ability to design biology, to let humans flourish and do amazing things. That's our dream.

Many people at Ginkgo are here because they have a certain application in mind. They really want to work on carbon fixation or cancer or whatever it might be. And we have our customers all have those interests, but we as a company don't do that, right? We're a platform. Our business model is like, think like an app store. We get a royalty on when people develop those applications, we get a piece of the value of their application, but we ourselves are judged by how well we enable others to design biology. And so that's what I dream of, right? Is that we've made it easy to do someday. But aren't you more like a consulting firm that does work under contract for clients?

Yeah, this is a really good question, Lynn. Yeah, so at the moment, a customer comes to us, they ask for a cell to do something. It is kind of similar to like a consulting arrangement. Like, okay, we have a team of scientists whose job it is to do that. And they have access to our platform, our robotics and our data to deliver that to the customer. Someday I would like the customer to just directly use our platform so that your scientists, just like today, your computer programmers use Amazon web services, but the technology maturity is not there yet to have that work. But someday I would love to have that work.

And that was, by the way, the history of Amazon Web Services, right? Originally, only the Amazon engineers were using it. And then they basically built the interface clean enough. the software teams will only talk with APIs from now on at Amazon to force a level of standardization between the services and the teams who are developing the website. Well, once that was clean enough, that team developing a website didn't have to be in Seattle anymore. They could be anywhere. And so suddenly the cloud computing was born as a real industry that we would like to get there. But today, as you say, it's only our scientists using it, but I hope to change that in the future. Okay. So you would like to create- Then eventually I'd like it to be your kids. Today, anyone can get on the web and design a

website and do that. Eventually, that's where I'd like it to get. It should be that anyone could design biology someday. So your dream then is essentially to create the standardized tools that anybody could use. You got it. Okay, that's exciting. Jason, before I ask for the three takeaways you'd like to leave the audience with, is there anything else you'd like to touch upon that I haven't already asked you? What should I have asked you that I did not? One thing we think about a fair bit on the risk topic is the people that are building the platform. There's been a long bit on the risk topic is the people that are building the platform. There's been a long history on the technology side of the backgrounds of the people building a technology, getting embedded in that technology. So specifically like the first heart valves being too small for women's hearts.

The original Kodak film didn't resolve black people well, right? Because the picture of what they used to do while the film was a white family. So you have this history of people having not be able to see around the corner on things they haven't experienced. And so one of the things we're big on at Ginkgo is a strong worker ownership culture in the company. Everyone's a shareholder here. We have super voting shares for the employees. And we want to have people at the company who come from backgrounds that might have been negatively impacted by biotechnology previously. You know, the global south has had a hard time in biotech and agriculture. As we just talked about, like with infectious disease and COVID, I think military applications of biotech are going to be a thing coming up.

We want to have veterans at the company and underrepresented minorities groups. They're on the short end of the stick on many technological changes. And so we do try to make sure we have a diverse set of folks actually building the technology to help us see around those corners. That ends up being quite important, I believe, in the long run for the tech. Absolutely. What are the three takeaways? I think from my standpoint, we're sort of entering the DNA age, right? Like if you think of like the last century, at least the last half of the last century as the electronics age, reading and writing DNA and the pervasiveness of biology in our lives, I think it will mean that this ends up being more transformative than electronics was

because electronics is the world of information. So all those computers, what did they really impact? Media, finance, telecom, like the information industries. What didn't they impact? Hamburgers, building materials, all these things, the physical world. And so biology is like computer. You can program, you can make it do new things, but it doesn't move information. It moves atoms. So I actually think it's good to the average person to have a bigger impact than computing. That's number one. Number two, I think Jurassic Park is right. Biology is wondrous and you have to approach it with humility. So as people, as more people get the ability to design biology, approach it with humility, but be respectful of how powerful it is.

And then the third, if we're successful and we do make this available broadly, you know, if you look at kids, they're born with wonder about biology. They really are. Like you cannot stop a kid from being obsessed by, I have a 10 year old, seven year old, like dinosaur obsessed, the zoo, the whole thing, right? And then we teach them, that's not important. We're like, no, no, no, no, no.

The computers are important. Let me teach you how this machine works. We push them away from that fundamental, like obvious wonder they have for biology. Don't do that. Don't do that. Don't do that anymore. The kids who grow up this generation and preserve that wonder, they're the bioengineers of the future. They're going to own the next century. So make sure your kids are keeping that love for biology.

Jason, this has been great. Thank you. Thanks, Lynn. If you enjoyed today's episode and would like to receive the show notes or get new fresh Thank you. Thanks, Lynn.