Ginkgo Bioworks Holdings, Inc. (DNA) Earnings Call Transcript & Summary

Welcome everybody to Gingko Ferment. Super excited to have you all here today. I want to start by thanking Quinn and the Ferment team for this absolutely gorgeous venue, we all get to be in today. It's a quick round of applause for the team. Okay. So I have two big goals for today. So number one, I want you all to meet each other. As they carefully curated group of people really introduce yourself to someone you don't know and from today take advantage of the breaks and the party tonight. And then number two, I want us all to learn from each other. So one of the things I'm most excited about is today at Ferment the people you're going to mainly see up on stage. More than anything else, are our customers. So you're going to hear from 11 different companies talking about the applications, the products they're building on top of Ginkgo's platform. You're going to see panelists from our customers go check out these cool domes here in the back, you can meet with a bunch of customers and up in the library, see highlights of the products that they're developing. And then tonight at the party, you're going to get to eat and drink products from Motif and Ayana Bio, all right? And so super excited about this, and I think it's an important point to make because people are often confused about Ginkgo. We don't make our own products here. We are a platform services company, right? So we ended last year with about $1.3 billion in the bank. I'm not spending that on clinical trials or new food products or AG products, that's all of you all. I'm spending that on platform technology to make it easier for you to get products to market. So we want to learn from you, do not be shy, right? Tell us what is missing on Ginkgo's platform, what you'd like to see us change, timing, pricing, whatever it is. We want to hear it today, this really is about you, okay? All right. So before I get to our platform, I do want to take a minute and talk about what a year 2022 was for synthetic biology as a field. So this is an executive order from President Biden came out in September, it says, "we need to develop genetic engineering technologies and techniques to be able to write circuitry for cells and predictively program biology in the same way in which we write software and program computers." hell, yes. This is something -- those of us who've been in the field of synthetic biology for a long time have been waiting to see happen. We want to see the federal government putting, synthetic biology up there with AI and semiconductors as critical technologies for the country. I think you see that happening. This was the meeting at the White House. You see I was there, right after the EO, Mark Warner head of Intel, you have the Head of HHS, DOE, Agriculture, Deputy Secretary of Defence, Jake Sullivan National Security Council was leading the meeting. And the basic discussion was, what are you going to do in your agency to bring more biotech into it, okay? Very exciting for the field. And then, be about a month after that, right down the street, at the JFK museum here in Boston, President Biden came in and announced a new agency getting created. ARPA-H, okay? And this is modeled after DARPA, that's the research agency that gave us the Internet, right, bring that kind of ambition and risk taking into developing health and medicine. And who did he pick to run this agency, right there in the green, Renee Wegrzyn, who you're going to get to hear from today. And let me tell you what did Renee do when she was back at DARPA? She ran the synthetic biology portfolio, okay? So this to me is a sign. You're seeing this from the administration that the types of technologies we expect to be the DARPA hard projects in health are going to come out of Symbion. So excited to hear from Renee, she'll be interviewed by the wonderful Michael Specter, a highlight up in the library, you can get a free copy of Michael's new audio book [ little Plug ] so check it out, super excited for that discussion. The other thing that happened on the government side was there's a national security commission for emerging biotechnology created. It's got two senators, two Congress people on it. I'm actually honored to be chairing this commission. There was a similar commission 3 years ago for AI that Eric Schmidt was the chairer and this is highlighting -- now you're talking about the defense sector and the defense part of Congress saying, "Hey, we want to know more about what's happening in biotech for both national security and strategic competitiveness reasons, national competitiveness. And so very happy to have Michelle Rosso as the Vice Chair on that committee. She's going to be here today previously on the National Security Council. And there's an interesting interface between national security and public health that's going to happen here. And so we're also very honored to be joined by Richard Hatchett who is the CEO of CEPI, just a global health champion who really did enormous amount of work getting vaccines out to the well -- world during the pandemic and Megan Frisk, who currently is doing Biotech Policy coordinator for the U.S. Department of State. Okay, it's going to be a very exciting panel, and we're lucky to have the Lieutenant General, Tom Bostick to moderate at a friend of Ginkgo, super excited about this. We have not been on the sidelines at Ginkgo in this area, this interface between National Security in public health, and we're lucky to have Matt McKnight heading up this business for us, and he's going to speak to you today about our programs, led by the CDC that detected first sequence cases of BA.2 and BA.3 in the U.S., we're through this program, and we're looking to expand that internationally now and Matt will tell you a lot more about that. So excited to be working in this area. Okay. Also in 2022, we had our big industry meetings, SynBioBeta and Eric Schmidt gave the keynote and this is what he said to the crowd. "it's time to take the work of the last 10 or 20 years and make it scale to a global phenomena that is somewhere between $4 trillion and $30 trillion of economy, you all will do that. That's why I'm here."" That's a message for all of you, right, that are developing these products. And it's no surprise to me that you're seeing more of these tech leaders. Like Eric was CEO of Google '01 to '11. When the Internet went through exactly that change, became that huge part of the economy. He knows it when he sees it. And these tech leaders are coming in, in part, I think because they recognize that just like in computers, in biology, we have a common code, in this case, DNA code that crosses all these different markets. And they saw this with software. You had Amazon Web Services Cloud Computing. It's the same infrastructure for Finance and for Pharma. Same thing with Operating Systems. And so the argument was, could we do the similar thing for biotechnology. And that's what Ginkgo's been trying to do. A common platform for all markets in biotechnology to make it cheaper and easier and faster to develop cells. Lots of people said we would not be able to do this,okay that the robotics you would need for microbes versus mammalian cells wouldn't be the same. The AI models you would use would be different for Pharma or Ag. And I'm happy to say we're proving those folks wrong. So we have more than 80 customers on the platform now, this is just some of them. And I'll highlight, it's big pharma names Merck, Novo Nordisk, Biogen, in industrials, in the chemical industries, Sumitomo, Givaudan, Givaudan is the largest fragrance company in the world, Solvay in Europe in chemicals, in agriculture just added Syngenta earlier this week, Corteva, Bayer has been a longtime customer. So the largest players across all three industries engaging with our platform. And then importantly, many of the startups, the new future greats in these industries also. So for both small and large companies across industries, we are seeing adoption of the platform. And it's getting faster. So in last year, we increased the number of active customer programs by 60% at Ginkgo and the rate of new program launches by 90%. So okay, investors care about this. That's exciting. The company is growing. I would argue all of you should care about this too. And the reason is that Ginkgo has a platform scale economic. The more customers we add to the platform, the better it gets and by better, I mean, it reduces risk for the cells you're engineering. It takes less time and it costs less. And I'm going to talk more about that in the talk. So when we add customers, when you see a new announcement from Ginkgo, if you're one of our current customers, you should be happy. So please send folks our way. All right. So what I want to spend a little time about is I went out and talked to a lot of our customers about why are they partnering with Ginkgo, why they using our platform. I want to share a few quotes and kind of the big categories I hear from people. So this is from Brian VanDahl at Novo Nordisk and you're going to see him on the panel in a minute. It's not why I picked his quote. I actually think he's got this right. So he said, "Science is currently undergoing a revolution. Large-scale data sets coupled with AI are opening up a greater opportunity space within biology. We no longer have to limit ourselves to the questions that can be addressed by traditional research methods." And this is the key point. It is those large data sets plus AI means you can ask bigger and better questions in biology. And this is not just pharma R&D saying this. This is if you're an auto company, if you're a financial company right now, if you're a health care system, you're asking how is big data and AI going to impact what I'm doing, right? So I think Brian is 100% right about this in Pharma. We're super excited to have him. Chris Sander from the cBio Center at Data-Farber and Joshua Dunn, Head of Design, moderated by wonderful Claire Evans to talk about this in our panel about AI meets bio in just a minute. Okay. So these are the big categories of why I think people are adopting outsourced synthetic biology platform. The first, what Brian said, more data per dollar. Second, you shouldn't only have access to your in-house data right? Like you need much larger data sets to make this stuff work. You actually want to access data across the whole industry. Three, you want it to be fast. It needs to be faster than our current time lines in biotech. Four, if you're a small company, you want to cut CapEx, like you don't want to have to build a lab, okay? And then five, you want it to be a variable cost. when you want a lot of R&D, you want to spend a lot. And when you don't, you want to just turn it off, like knob. That's not how it works today with traditional biotech. All right. So I'm going to go through each of these. Cool videos, this is -- we acquired a company called Zymergen in California last year. I encourage you to go check out our facility. It's 10 minutes from here, but this is really the leading technology for flexible lab automation that allows you to swap in and out machines and great software to handle it and everything else. And the reason I show it to you is just as people developing products in synthetic biology, you should not need to be the world's experts in laboratory automation. You should not need to be the world's experts in liquid droplet management. These are things that Ginkgo can be expert in and you can just access the latest technology so that you get more data per dollar. We are committed to doing that. That's what we're spending all our capital on. okay? And you hear this need for these big steps from our customers, Alphonse who is at Biogen, when we did that deal said, we did the deal, with Ginkgo to explore a large number of design ideas. Marcus Schindler, CSO at Novo Nordisk, we wanted to be able to rewrite whole genomes and engineer new bespoke biological systems. It was to access the scale that wasn't otherwise available in-house. Second, it shouldn't just be your data assets that are going into producing your product. And what Ginkgo is committed to doing is getting together as much data as possible around the industry and make it available to you. So you can see this comments, Givaudan it was the breadth of existing assets is why we work with them, with Merck, it was the experience of the employees. So our people's ability to use these assets and at Sumitomo, the great transparency and sophisticated data set. So let's touch on that transparency for a minute. So with Bayer we acquired the agriculture R&D unit last year. All these other companies, we acquired the full set of IP, okay? If you're interested in the AAVs from StrideBio, the Circular RNA from Circularis all available to you transparently in our data set. But I'll highlight one thing here. This is a set of genomic data we have. Compared to the public database there's 246 million Novel genes, this is in Uniprot. We have $2 billion in our proprietary data set from all these acquisitions we've done. All available, right, to you as a service, if you want to train your models. Okay. So the third, we launched work quickly. We already have this infrastructure in place, to Trent at Microba said, "Hey, Ginkgo's expertise and resources, move your drug discovery project along at a pace that would not be possible with either internal resources or traditional CRO." Nicholas, CTO of Lygos; said, the team is talented, the early results on one of our projects are stunning and support Lygo's mission of accelerating the world's transition to high-performing sustainable products. That one is interesting. We just announced that deal started 6 months ago. So how are you getting stunning results in 6 months? It's because you're not starting from scratch like we have to break this idea in biotech that we start every project from scratch. I was actually building on an enormous amount of Bio Resource we had a Ginkgo plus leveraging automation on the first stack. And finally, Bob Reiter, Head of R&D; at Bayer they did a big deal with us to outsource both to us and to other companies. And Bob's saying, look, open innovation is letting me tap a wider set of data to bring products to market faster. This one is for the smaller companies in the room. It's a great comment here from Jasmina, CEO at Arcaea. "Arcaea was able to begin lab work in weeks without building out our own biotech capabilities." This was a brand-new company. They had just raised money. What would normally happen. You'd call Alexandria, they would show you overpriced lab space in Kendall Square. You spend a fortune on a bunch of rent, you spend a fortune on a bunch of new equipment and 6 to 12 months later, you get to start doing some work. Instead, Arcaea was able to start in weeks and it says we were able to launch a compelling, differentiated product in less than 2 years had we not worked with Ginkgo, we'd never be able to move this pass or deliver as great of a product, and you should go check out their product, go talk to Jasmina and the team, they're going to be here about what they've been doing it, but that speed is because the assets are already in place. The last point I'll make on this, this is what, I would say, a normal like startup life science therapeutics company looks like at the beginning, so that dotted line is the R&D team that they build up at the start. At the beginning, they wish they had more research people. They want to get to that drug candidate faster. They want to try more design so that they get a better drug to go into the clinic. Then is we got it. This is our shot. We're going to put it in the clinic. Well, suddenly, they want a smaller R&D team. They're like, I want to spend all my money on the clinical trials. I don't want to be spending money on some future stuff. I don't even know if this works yet. Then they get a good result. And suddenly, it's like, oh my gosh, I want a big R&D team because I want to grow my pipeline. And the really ugly part of this is when the venture industry contracts and tightens like it is right now in that little dip in the middle, what they do is they just lay off the R&D team. This is not an efficient way to run our industry. It would be more efficient for us to have services that provide variable cost R&D. You need a lot at the beginning, great. You need none when you start your clinical trial, turn it off. That would be more efficient. That's how compute works with Amazon Web Services, when you need it, you do and you don't, you don't. We want to create that in the area of research services. Does that make sense? So those are the five reasons. So let's say you hear all that and you're like, okay, great. I want to sign up. How do you sign up. Right? So, one thing is go to our website, you click, you say work with us and you're gonna hear from our technical scientists who go engage with customers and commercial team and the -- this is like very easy. You can tell them what you want to do at Ginkgo and they'll tell you if it's possible. But a lot of people, that doesn't really work. They're like whatever, I don't really know if it's relevant to me. Can you prove it to me, can you show it to me. And so one of the things we did this year was we launched our first service offerings, Ginkgo Enzyme Services. And this was meant to say, if you were working on a protein project, an enzyme project, you should really be working with us and we had a whole bunch of preexisting data and all these things. That worked great. So I'm happy to say today, we're announcing the launch of 4 new services, Ginkgo microbe services, cell therapy services, AAV services and RNA therapeutic services. And this is built in part on some of the acquisitions we've done recently, StrideBio and AAV, Circularis, the Bayer Agriculture, our biologics team out in West Sacramento. Those are all rolled into these services. And if you're excited to learn more about this, check out the events calendar, we have this great webinar. We're going to be sharing more materials in the coming months. You're also going to get to hear from some of the those scientists that our customers are raving about that are sitting on top of these platform services, Emily Wrenbeck, in just a minute on sort of enzymes, Shawdee for RNA and cell therapy, Magalie in microbes, and then Kennon, who just came in via the StrideBio acquisition,well throwing them right in the fire. So he'll be talking today about that. So the #1 request we're getting from customers is to help them de-risk their cell engineering. So yes, people want it cheaper, they want it faster, but they really want to take the technical risk project. And so you can see that, right, Nicholas at Lygos as a CTO at a growing company, "what can I possibly do better than working with Ginkgo to de-risk my plans?" Keith from Optimvia, "The likelihood of success was seen to be improved by the technical capabilities at Ginkgo." So this is a tricky topic. So in enzyme services, we have now done enough programs that we have a sense of what is very likely to succeed and what is not. And we tell our customers like we're like, and this is totally going to work. And they're like, "Yes, I don't believe you." Because no one in biotech really believes that anything is predictable, like it all is research. It's not like building bridges or cars, it's research. And so today, at Ginkgo, we're announcing we're putting our money where our mouth is on these services. So for our mature services here, enzyme, discovery, optimization and protein expression, we're moving to success-only payments. In other words, if we hit the technical goal, we get paid. If we don't, we don't. And that means that I am offering our customers what they want, which is de-risking. I'm taking that technical risk and I'm putting it on to us because we have confidence there and so you don't have to pay for that risk. All right, so really excited about this. If you want to learn more about it, [email protected] or check out the dome back there that has the T-rex skull on it. You can talk to some of our commercial folks about this. I think it's going to be a big deal and really excited to pioneer this in the industry. Actually, one quick story there, like an only in biotech story. So when we announced our deal with Biogen a couple of years ago, it said like there's this new deal has a $5 million upfront fee, okay. And there's this -- like I won't say the name of like a biotech reporter on Twitter is like, "$5 million upfront be like, what's this the deals for ants? Why is it so small? And I'm like, how crazy is this. Let me tell you something customers do not like giant upfront fees. And so like what I'm committing to you all is that we're going to work to make things less expensive to make them faster and to reduce the risk in your projects, like this starts with you. We are a services business. We want to do that. You won't see any like shooting for biotech you'll see me for making you guys more successful at a lower cost right? So let me be clear about that. I'll end with this. Biology lives in the world of atoms. We are made out of biology, the food we eat is made out of biology, the atmosphere is produced by biology. And as we develop these tools to program it, it is going to have deep impact on us and our lives. And we need start asking the questions, discussing the ethics and impact today. You're seeing this in AI right now. Holy c***, everyone's talking about ChatGPT, the biggest open issue is they're worried about all the regulations and the ethics and what's going to happen, you need to have these conversations early in the technology curve. And so I'm really excited. Again, if I have not proven today it's about learning, we literally have a library upstairs. Go up there and get free copies of these books, we really like on these topics. You're going to hear four amazing storytellers at the end of the day today to open your minds on these topics, our new addition of Grow magazine. Please please, I think it's important for our collective group to get smarter here. We're also going to have a great panel on the future of plants. I'm hoping we have Maddie Hall, she's having like travel issues, but just to give her background. She's the CEO of Living Carbon. It is a company that's genetically engineering plants to grow faster to take carbon out of the air in this case, Poplar trees and then Ruth Wagner, who's the Head of Data Science and Analytics at Bayer Crop Science, we talk about all of the big data applied in plants and moderated by Nadia, Berenstein. We should use plants to do more than make food. And so I think it's a really exciting area to talk about. And then we're super fortunate to have [John Maraganore, ] interviewed by Jen Wipf, our Head of Commercial here at Ginkgo. And John was the founding CEO of Alnylam. So he went through the experience of taking sRNA as a new modality and launching it to a $25 billion company with drugs and patients how do we make that go faster? What are the lessons we can learn or we really like to hear from John. So I want to end on this. You talk a lot about AI today, and so I asked ChatGPT. I said, what is more important to society, atoms or bits he says, "Well, atoms and bits are both important to society. They play different roles and different implications that goes on about atoms and what bits are and all this. I'm like, wow, this is really becoming human intelligence is like an impressive dodge of my question, right? So I'm like, no, no. ok ChatGPT, But if you had to pick one, what is more important to society, atoms or bits. Well, if I had to pick one, I would say that atoms are more important to society because atoms are the building blocks of the physical world, they're essential for our survival and are well-being. They make the food we eat, the water we drink, the air we breathe. They make the materials that build our homes and roads and bridges without atoms, society would not be able to function in its current form. I could not have said it better, all right? Synthetic biology is the programming platform for programming in the world of atoms. At Ginkgo we are trying to make that technology easier and better and faster. So all of you can have that impact in the world. I cannot wait to see what you build with it, and I'm delighted to learn with you here today. Welcome to Ginkgo Ferment. And i'm excited to welcome up Emily Wrenbeck.

Good morning. It's very nice to be with you all here today. So my name is Emily and I have a very fun job of leading a team of protein engineers at Ginkgo and so the focus of my talk this morning is going to be on how we approach enzyme discovery and engineering at Ginkgo. And I will talk about how we view this as a coupling of both AI and computational design approaches, coupled with our data generation engine, we call the foundry. And so in case you've missed it, proteins are really hot right now. And there have been a number of really big headlines hitting mainstream media, even talking about how we now have hundreds of millions of pre-computed protein structures at really good accuracy as well as generative AI models that can help you dream up, right, new protein sequences. And so this is also reflected in the boom in research publications in the field. And this is largely due to the fact that proteins and the technology for designing them have really been able to kind of ride the coattails of a lot of the other developments in AI research in other fields, especially in business applications. And so now more than ever, we're sort of like at a juncture where we're considering like what is the composition of my day to day going to look like in a business? Because AI technologies are advancing so fast. And GPT is from Open AI has a great topical example here. And the latest version of ChatGPT especially has made a major leap because not only are the predictions getting really good. It's also very user friendly, right? It's a really good product. And for example, I could ask, I had a similar idea of Jason is asking to chatGPT, "thanks from my talk of like -- could you write me a presentation on protein design in Ginkgo Bioworks. And after trying a few different prompts and asking it a few different times, the answer I gave was actually pretty good. And so I think to me, this sort of hallmarks like it's a major leap in terms of the felt impact. These technologies we're going to have in a very practical sort of day-to-day sense in our workday. And I think the same is true over in the world of proteins, right? All of the new technologies coming online are changing how we think about protein design in a very day-to-day practical sense. And so again, this is due to the fact that protein technologies have really been able to provide the coattails of AI research and other fields, especially natural language processing. Because folks were very quick to realize that proteins kind of have a language of their own. They have an alphabet out of it, of amino acids and there are motifs in proteins that look kind of like words and there are sort of patterns and very nuanced rules that are available to learn about proteins. And so the application of large language models, especially have been very transformative for protein applications. And the big success story here is AlphaFold 2 right? Like we now have the ability to generate protein structures at atomic resolution for most proteins that really rival experimental capabilities. So this is truly transformative and impactful on a day-to-day sense. And so there is a lot of focus, right, on the machine learning and all of the products that are made available because of these advancements, but I just want to take a moment to step back and talk about like the data for a second. Because none this is possible without data, machines can't learn from nothing. They need content and so similarly, you also can't really understand what a machine has learned without having the right kind of data to test and validate it. And so if you just really just kind of consider the relative investment of generating this corpus of training data in comparison to what it takes to develop these large models and build apps on top, it's really orders of magnitude difference. GPT and models like it are pretty much trained on the Internet. And there's no like real hard figures for this, but we're probably somewhere in the trillions of dollars of money spent to generate all of the data that's available on the Internet. And similarly, over in the world of proteins, the protein data bank, which are like all available experimental structures of proteins, probably somewhere in the billions of dollars spent in research funding. And so in both cases, if you really just kind of take stock, right, all of the assets that go into generating great products like ChatGPT, the data is kind of invaluable in a sense, right, because it's also reusable. You can take the same data and then train new models and try to get new insights. And so data is Queen. And we believe that the key to transforming what is possible in biological engineering is really good data, the right data. And so if you've ever visited one of Ginkgo's facilities. You will have seen our foundries that are full of all of this really cool robotics hardware and the corresponding software that make the box do what we want. And that's because we believe that physical experimental data is fundamentally important to transforming what is possible in biological engineering. And it's not just about like generating gobs of data so that you can train really sophisticated models, it's also just about maximizing the chance of success, right? maximixing the other's success and just getting the job done for our partners, right, being able to generate a lot of data is very enabling that way. And so how does this view inform how we think about approaching protein design and engineering at Ginkgo. So Again, as I mentioned earlier, we view this as sort of like a coupling of all the state-of-the-art sort of AI and computational tools in a design tool kit with our data generation engine we call the foundry. And so on the dry lab side, we are experts in the sourcing of natural libraries of proteins we often refer to as metagenomic sourcing. This is done through both public and proprietary sequence data basis. We have built a rather broad computational protein design tool kit including more classic methods like molecular dynamics and Rosetta. We've also built an automated platform to make use of sort of the latest and greatest AI models for printing prediction that come online, including Alphafold and ESM and EV couplings, ProteinMPNN, et cetera. So that using them in kind of a routine way as possible. Finally, because we have the foundry, we also often opt to take like a supervised machine learning approach to protein design which means that we simply just train models directly on the experimental data that we generate in the process, and we do this all in our platform we call Owl and then finally, we have the ability to fine-tune large language models also using the same data generated from the foundry. And so just to give you one quick story of the platform in action. In this study, we were asked to optimize an enzyme, which was critical to a customer project and ask was just make the enzyme more active please. And so about this enzyme, it's known in the literature to be rather recalcitrant engineering. It also has an unsolved reaction mechanism. And so both of these facts kind of make doing traditional rational protein engineering complicated. And so we opted to do a more data-driven approach to protein design. So kind of like the data be the guide in a sense. And so what we did was we engaged in four successive rounds of design, build, test and learn. In early rounds, we were using some of that design toolkit I talked about earlier, we're using sequence-based self supervised models, active site immunogenesis, Rosetta docking and design, things that hopefully are familiar to some here. And the goal of this was to really sort of explore the sequence to activity relationships of this protein. And so in successive grounds, all of this data was fed into our Owl platform to train models and build designs and finally, culminating in a final hit that was tenfold improved over the starting enzyme. And I just want to point out that the library size is here in the final round, we were able to realize a really big leap in performance, so the rather small library because as the models are given more data, they get much more predictive. And so you can often hit your goals in a much more focused way. And so just in closing, I just want to make sure not to say that our experience with enzymes is very broad. And so this plot is a bit more artistic than you can actually -- than the scientific, but the point here is that all of the rows here are different top-level EC numbers and then along the X-axis is how many tests we've done or sort of the far end is around 100,000 or something like this. So we've done a lot of testing of a lot a of enzymes in a lot of functional space. And so you can see some stats here at the bottom, and this list is kind of always growing and love to add your enzymes to the list. So if you want to talk more about our enzyme services, myself and my colleagues will be hanging out and I think the Earth Dome over there during the break, so please come find us and chat. And up next, we're going to continue this discussion on using AI and other machine learning technologies for designing proteins and developing medicines with the panel discussion. This will be moderated by the lovely Claire Evans, who is a musician and a writer and you'll have participating Brian Vandahl from Novo, who is a real thought leader in this space, and we've had a really enjoyed a really great partnership with Novo as well my friend and colleague, Josh Dunn, who is an index of knowledge you can ask anything from AI technologies to very complicated Yeast biology as well, Chris Sander,who is really leading the way and applying quantitative techniques to solving different problems and challenges in biological engineering. So with that, I'll say thank you and enjoy the panel.

Good morning, everyone. My name is Claire L. Evans. I'm a writer. I write an occasional column for Ginkgo in-house magazine Grow about computation and biology, which is why they ask me to moderate this panel because that's what we'll be talking about. So as it has in so many fields, including my own, AI has turned biology upside down. It's only been a few years since we've effectively used AI to solve the problem of protein folding and progress has continued since then at a kind of staggering pace. Now we're not only predicting the structures that proteins will take with almost experimental accuracy but now we're also synthesizing and generating new ones to better suit our purposes. So whatever next, I'm joined this morning by a very illustrious panel. We have the computational biologists Chris Sander whose lab at Harvard applies computational tools to solving biological problems; Brian Vandahl, Vice President of Global Research Technologies at Novo Nordisk; and Ginkgo Head of Design, Josh Dunn. We to talk about how AI models are changing the kinds of questions we're able to ask in biology and how what we learn can be put to work. So we don't have that much time, so let's just jump straight into it.

The question of how proteins fold was a grand challenge in biology for nearly 50 years. Now that it's been more or less put to bed by AI, what happens next? What do we do with all this new data? And what kind of questions can we ask now that we weren't able to ask before. Anyone who wants to take that?

So you may say it's a giant step that we may now be able to somehow predict the protein structure, but it's just a small step because the biology is so much more complex than that. So one thing is how proteins have fold another thing is how they interact with each other and then how they function in the complexity of the human volume, that's a much, much bigger question.

I think. What I would add to that is you really want to think about why we wanted the structure is in the first place, right? And so one use case, one of the reason you want to structure that's close to my heart is you really want to know what a protein does before you actually test it. And the best way to do that is to look at the structure. But the structure doesn't tell you every aspect of how our friction functions. It doesn't tell you right now what the substrate is, what the product is, what the reaction is. And I think that's a really obvious target for a next-generation model is going straight from sequence to function. There are some other obvious targets like drug interactions and things like that. As far as kind of the broader future, I think we can take a lot of the techniques that have been applied to proteins and apply them to DNA, RNA and begin predicting functions, structures and all sorts of regulatory elements that we could begin designing in the same way.

Yes, go ahead.

So just AI and biology, also unsolved problems, of biology, of course. We heard about the protein structure were amazing progress and so on. Imagine the protein walks in the door, you have a structure, you don't know what the function is new proteins. And then yet and function has to do with interactions themselves. So you want to actually put the protein structure and the new designs and the context of Cell Biology. So a major challenge in compared AI and biologies in fact, competition of cell biology, how do the cells work, all the components, not just a few and then how the cell cells interaction and organism. And there's this amazing challenge of how to now take the next step and do a the computation of cells interacting in an organism and how that relates to disease and also in technological applications, how you bring that to bear in environmental and health problems, which I can talk about later.

I mean speaking to that, I think machine learning tends to open up this new possibility space that allows us to find structures that might not have been time or cost efficient to find beforehand. But once those structures have been found or predicted or synthesized, you still have to take them out of silicon and bring them into the lab and verify them or build them. Can you speak a little bit to the relationship between models and lab work and if there's like a feedback loop there, how that works?

So I think it will remain an iterative process, at least when you're talking about protein and peptide therapeutics also for a long time in the future. we have AI assisting in designing molecules that we will then go back to the lab to test. And that is what Jason also spoke about, so the interplay between high throughput lab automation and AI because we need AI to help design the experiment in a way so that we get the maximum information and value out of the experiments that we conduct so that we can feed those data back into the model. But it will still, as I perceive an iterative process for many years to come.

And I think what I would add is actually just to repeat a message Emily, my colleague delivered a few minutes ago. Data is Queen. There's no substitute for data in validating the model. And one thing that I think is a really important point that she made is that as we are iteratively engineering enzymes taking predictions for models and testing them in the lab, we not only improve the function of the enzymes, but we improve the function of the models. One thing she showed very clearly is that the models get better at predicting the next mutations that make so that we can make further improvements in creating, testing here and fewer things with each cycle. One of the beauties of having a foundry in Ginkgo is we can actually start are wide and a lot of experimentation and then zero in build up a reasonable knowledge base that we can apply to some the same enzyme family in the future or some of the problems.

Yes. I mean with, for example, large language models they say bigger is better, right? And -- but the more data we can give them, the more emerging capabilities they will have, the better they will be. Is that true? And how can we leverage that? I mean, is there actually enough good verified data out there to train these models at a scale?

For the first question, like is bigger always better, I don't think we know yet. And I think the only way to figure that is just to try and see if we saturate. My hypothesis is a biologist is that these things are going to saturate at some point. But we've got to be bold, and we've got to push AI large language models to see if we hit that ceiling. And if we do, we'll go on. For the question about whether there's enough verified data, there is and there is a it really depends on the topic area, right? So if you want to ask where are the large corporate publicly accessible data that are well described and that are underutilized. DNA sequences, RNA sequences are the first ones to go after. There's a long tail of smaller things after that, but there's a lot out there that you could live the same models and do some really incredible things with, and I think we're going to see that in the next few years.

Is there anything, you want to add?

Yes. I mean one challenge is, as you know, it's one thing to have an AI engine and we all have the technology. But there's another problem to actually, first of all, hit some domain knowledge in order to use the AI engine in the right way. To give you an example from the protein structure, the amazing observation is that proteins and evolution are constrained. And over millions of years, all the information about the functional constraints of proteins is the positive in genetic sequences and so one of the key inputs to all those new methods we heard about is, in fact, the fact that molecular evolution is giving us the information. So for other new applications of AI input engineering, you have to have some domain knowledge to actually that useful, number one. Number two, you actually have to have a ton of data. And we all have seen examples here, Ginkgo had lots of data. If you don't have the right kind of data and you don't prepare it in the right way, then you can't really very successful. I give you one example, which is genetic variation. We all have genomes and we all -- some us have angular genome sequence, we have mutations in our genomes. We like to know what's the relationship between the variation and the differences among people, number one; number two, propensity for disease -- and so how do you now get enough data about all the genomes and that's on progress. That's a big challenge. We're getting more and more genomes, number one. Number two, the relationship to disease. So as we speak, AI methods are being applied to analyze genomes increasingly and then relate to the variation to [ preventative ] disease, which in the end also has implications for therapeutic development. So that's a major challenge, and we look forward to doing that.

Yes. This is kind of a philosophical question, but something I find somewhat confounding about in this context is that it allows us to solve really big problems without necessarily understanding what is going on underneath like we're able to predict protein structures. We don't necessarily know the dynamic processes that create those structures. We have these structures. We don't necessarily know the function of those structures. So is there a disconnect for you in this? I mean, is it enough to just to know that something works, don't you also need to know why it works.

I love this question, I just wanted to give other people an opportunity before I start spieling.

Spiel.

Spiel mode on, activated. Okay. So I was talking to one of our founders, Tom Knight, the other day, and he was telling me something at one of his mentor shared with him when he was younger, which is when you're learning something new, it is a really poor understanding to understand something only in one way, meaning to really build insight about something, you have to understand it through a bunch of different lenses. You have to compare it to things you know. You have to be able to make predictions about it. When will this method work when we will -- and so where are we now with large language models, right? These things can design proteins -- those proteins work -- and we're going to see incredible pressures both in academia and in industry for people to operationally use these tools without spending the labor and the time to understand why and if we're not careful, we'll end up in a world where as a fuel, we'll have a labor pool and a workforce that has a core understanding of how these tools work and innovation will stagnate. And so what we really need to do is make sure that as we operationalize these things and use them as which we should, we're training a fraction of our workforce to really understand the corners of these methods and figure out when do they fail, when do they not fail, where did we need innovation, and we need to dedicate sub-fraction of this to really understanding, okay, this first great, why what new insights can we bring? And I would argue that our central challenge now is moving from one where in a world where you've got scientific progress, moving with almost unbounded velocity and an inclination for insights to liquid in machines rather than people, how do you bring some fraction of that back into your people?

Yes. I mean, it's all well and good when things work. But when things go wrong, you need to understand the mechanics of play right?

Yes. I just want to give you one example of something that might motivate us in the future for AI. And we know we have all kinds of problems in the world. There was war and peace, there's global warming. And so let me take you one example, environmental problems. So one example is this is not the only one, but it's an interesting one. We have lots of garbage in the world and the plastic garbage patch in the Pacific Ocean. So people discovered an enzyme, they actually degrades plastics. How do you take that forward? So the number of different groups that are involved and to engineering could be designed to now take this plastic degrading enzyme and improve it. So it's one thing to have the mthods we've seen on the screen just to get the structure, but that's not enough as you want a design, you want to actually have the active side pocket. So now you have to actually develop the methodology, not just to get the structure right, to make sure you have a stable protein that design, which amazing progress there, but also how to now get the right kind of function the part of the protein does the right kind of reaction. And then synthetic biology, which we've heard about, put those proteins, as you know, into bacteria or certain organism that you have been out there in the wild to actually help do the environmental clean up. Now those are major challenges of engineering and the community has to come together to solve these in the right way.

[indiscernible] When we're talking about the protein therapeutics, do I care whether we know why it works. Yes, I absolutely do. And I may also help us that if we get the systems biology into to play old but it's so much easy to investigate how something works when you have something that actually does work. And that is where I believe that the [ interplague ] between high throughput lab automation data and AI can help us. So if we just take example of what it takes to develop protein in therapeutics. You need safety and you should also probably consider convenience and today also scalability. So that is how to produce the molecules. There's a lot of different things to optimize for at one time. And if you look at how position, it has been a scientist with a lot of insights into the specific protein at Novo Nordisk, we've been working with insulin for 100 years. We have scientists who have been working on insulin in their entire career. They know if they change something maybe amino acid residue position 22 in order to optimize the potency of receptor interaction. They also have to do something in position 13, maybe to keep the stability of the molecule. But the way that we've been working more or less those 100 years is the scientist is having a picture in their mind of so how can we make this work. Then they go into the lab. They produce a 10, 15, 20, maybe 100 molecules today if we are thinking big. And they see that's the hypothesis driven research, see if it works, go back and take that in, but it's just totally impossible to get your mind around the entire design space if you want because you have just in a medium-sized protein, you have maybe 100 residue -- you have 20 naturally occuring amino acids. You have all of the non-canonical amino acids have all of the chemistry that you do to the side chains and so on and so forth. That's the design space cannot comprehend with the human mind. And that is where AI comes in because you can -- AI can help you probe design space in a clever way, so the data in these iterative steps bring you closer and closer to that final molecule faster and better than what you can do just by your own mind.

What you're implying here is a future approach to science that is no longer hypothesis driven, but some new form I mean that's quite provocative.

It's at least a different kind of the hypothesis-driven research where you don't necessarily as -- and that is what triggering my answer, you don't necessarily have the path to what you are envisioning to achieve in -- laid out for you. You are just probing a design space to learn where you may make the next library of molecules in order to take a step towards that function that you're desiring to achieve.

What do you think about that? Does this technology fundamentally change the way we do science?

I'd say yes and no. And so computational biology has existed for about 40 years since the blast algorithm was published, for example. And if you talk to your biologist today, it's pretty hard to find someone who hasn't used that algorithm in the last two weeks, right? So we've been using tools for a long time and tools will generate hypotheses, whether that is a pricing structure from a language model or an alignment from glass, and it's up to the experiment or to really understand, okay is this hypotheses reasonable -- do i want to throw this in the garbage can, do I want to reorient my experiment around it, right? And so what hasn't changed is there's still an onus on the experimenter to really understand their tools. What has changed is velocity, like last is 40 years old, still use it there's new fangled stuff that works better blast is the to go-to. But the treating language models, there's -- first of all, there is no 40 year old model. And it's quaint to use a model that's 5 years old. And so really, the question then is how do you as a scientist, as a graduate student as you're navigating your career now understand when you adopt a new tool? How do you understand when you expect your whole field to be disruptive, maybe not even just your project in the skew but the whole field being disrupted. And so what that means is scientists need to be trained as strategist. They need to be thinking about a bigger ecosystem about a world that's bigger than their project. And I think there's actually a real great opportunity here, both in the academy and in industry to bring some of the topics that normally you'd see in business school into PhD education or into on-the-job training for scientists, and I think that would tremendously benefit the bioeconomy.

Chris, you've been in this game a long time. I mean, what does it feel like to be working in the midst of such accelerated change in the last few years?

One thing that comes from my mind, which is there are other major biomedical challenges. And talking about COVID-19, one important one, and this is a collaboration with Debora Marks, who couldn't be here today, is to actually take all the information we have about viruses and now use AI to anticipate how a virus is going to evolve and how a virus is going to evade the immune response and then we work with the community, and this is an example where the whole community has to work together in a multidisciplinary way to then to actually anticipate the kind of vaccines, multicomponent vaccines or broadly effective vaccines that are going to be able to actually anticipate what's going to happen. So we don't have these cycles of infection and vaccines and therapy and that, I think, is an example of the community is being challenged the right data to collaborate and to work on the important biomedical problems in the way I just described.

What about the back and forth because neural networks, genetic algorithms, reinforcement learning, computer vision, these are all staples of AI research that are influenced or inspired by biology in some way. Does that influence go both ways? I mean, beyond the models, are there ideas or approaches in machine learning that can be usefully adapted to biology?

I'm inclined to say, am i still here, still here Certainly, right? So I think like if you look at technology, there's some really provocative papers that have been coming out in the literature over the last few years where people actually implement, for example, physical neural networks, where they're growing neurons and culture and demonstrating that they have the capability to solve interesting problems. One of my favorite papers this came out, I want to say, like 14 years ago, there studying a slime mold called physarum polycephalum and they showed that it could actually solve the traveling salesman in problem, which is the famous problem we can get our science in non-polynomial time. And the reason why this was interesting was because biological systems can compute things in fundamentally different ways and we kind of theorized, we as a field not me personally, obviously. But theorized that slime molds for example, may be able to embody some of the philosophical computations that people have. hypothesizing in theoretical computer science might actually exist to see living in the implementation of that is pretty stunning. So there's a whole field of this called biomimicry. So if you're interested, you can read up on biomimicry about where you see biological paradigms going into technology and then also the vice versa.

I didn't think we get the opportunity to talk about plans on this panel, so props for bringing that up.

Bring it up every time I can.

We were talking a little bit about sort of the relationship between biology like between industry and academia in this context? Because I think from what I understand, developments in AI and machine learning and the adaptation of those developments to biology don't necessarily happen in tandem, right? Like AI researchers will present new tools and then biologists will sort of adapt those tools. Is there a way to shorten that gap a little and encourage more sharing, where we source sharing more information securing between the life sciences and computer sciences or perhaps more sort of germane to this group between academia and industry. Brian.

So I absolutely believe that the collaboration and partnerships is the way forward just like what we are collaborating with Ginkgo on. So there may be technologies where it's not the right thing for farmers to have everything internal because it is a better guarantee to be at the forefront to collaborate with partners who kind of does the same job across an industry or across industries. So I really do believe very much in the collaboration between big pharma, biotechs and academia in order to get the best out of all the worlds.

What about the academic perspective on this.

I'll pass on that one.

Okay. Well, we're getting close to end of time. So I want to ask a sort of big picture question. I mean, protein folding was a problem of some stymie biologists for half a century. What's the next big challenge in your mind? And will it take another 50 years to solve?

So I would say, end-to-end, fully automated synthetic biology from design of my organism through testing in the lab strain construction, testing in the lab and iteration. I would to be able to come in, engage a ChatGPT like interface and say, "Make me a cell of some type you choose that has these properties in these genotypes. And I don't think this is a 50-year problem. There are elements of this that are already happening. Parts of it might be a 10-year or 15-year problem, maybe in 20 years but I'm keeping my eye turned on that.

I'd say that in pharma, it's probably translatability that remains the biggest question. So how through the things that we do in the lab, the in vitro assays, the biological assays and also the in vivo pharmacology, how does that translate into human relevance -- so -- and if we can have AI help assist in that, maybe in the first place, simply to define the best biomarkers to get more value out of the clinical studies as a step towards what may be the final vision of goal, the digital twin of the human body where you can do your clinical development in silicon.

Happy to give you one more example of having to do with a big important biomedical challenge. It's not about drug design and proteins and so on, but this is about cancer biology and trying to not treat people late in cancer, but as possible. So one challenge is looking at people's clinical recurrence and use machine learning before they get cancer, identify people who are at higher risk, so you can catch the cancer as early as possible, pancreatic cancer, for example, you want to be able to do surgery before it gets too densely to be late. What do you need for that? Yes, you need new AI methods, but you need the data. So the big challenge for the community is, and the government is funding this, in fact, with all of us programs to get all genomes and then get clinical records which is very, very difficult, especially in the U.S. health care system that is accurate and then wearable devices where you get people's information about their status, the health status, nutrition, et cetera, et cetera. We're not there yet but once we have that data, we're going to be able to treat these aggressive diseases much, much earlier, and that could potentially be a contribution of high of AI to human health that will take several years, but it's a major chance.

Well those were very optimistic visions to end this panel on. I thank you all for joining me. Thank you all for paying attention. And yes, let's design a better future with AI.

I'm excited I'm excited. I'm excited for the next session. Hi, everybody. How are you doing? How is the audience doing. Yes. It's not all about us. It doesn't have to be all about us, right? Sometimes it's about you. You look great, everybody looks very smart, very excited for this audience. And I'm very excited for the next session. We are about to hear some application lightning talks. So if Ginkgo is an enabling technology for synthetic biology, we're about to hear from the leaders who are deploying that enabling technology to create products. These are our customers. These are our partners. They're also our developers. I really like this term developers to refer to people to companies that build products on top of a platform. So just like a software developer builds an application on top of an operating system, Ginkgo developer builds a product on top of Ginkgo's foundry. Steve Ballmer, former CEO of Microsoft knew how important the developers are for a platform company. So if he was here, if Steve Ballmer were here, he would be literally on this stage, jumping up and down, shouting the word developers, developers, developers over and over again. That's what senior Microsoft leadership does you literally cannot emphasize enough, how important right developers are. I don't quite have that energy yet. Maybe this afternoon, we'll see. We'll see. We might do some stage jumping. But for a platform company like Ginkgo, the developers absolutely are the life's blood. So this session, we're going to go by Lightning talk rules. I'm going to introduce all three of our speakers and I'll get out of the way, so I don't get in between them and the audience here. We're going to hear from Allonnia is a bio-ingenuity company dedicated to extracting value where others see waste. Synlogic is advancing therapeutics based on synthetic biology, working at the intersection of biology and engineering to treat disease. And Sumitomo Chemical is a major Japanese chemical company with a history going back over 400 years. We'll be hearing from Nicole Richards CEO of Allonnia; and Analise Reeves, Head of Synthetic Biology at Synlogic and Satoshi Okamoto Chief Research Coordinator at Sumitomo Chemical. So please join me in welcoming our application lighting talk speakers.

Good morning, everyone. It's great to be here. Allonia is just over 2 years old, and we've launched and sold our first product targeting the world's toughest environmental challenges. These first products epitomize that waste is a failure of imagination. And that we can fast-forward nature using synthetic biology and our partner, Ginkgo. To find natural solutions to remediate big problems, we have set our sights on these 3 areas: mining sustainability, plastic degradation and industrial contaminant remediation. So today, I want to focus on this last area and give you an example of why we're so excited and confident that nature can heal itself and use the tools of engineering biology to help that. We're going to focus on one exciting product that we're commercializing as this example to remediate 1,4-dioxane. 1,4-dioxane is a toxic chemical that is found in 10%. In fact, I just read this morning, it can be up to 20% according to the NIH in our drinking water. So 10% of our drinking water includes 1,4-dioxane. 1,4-dioxane has many health hazards, including it's been linked to a carcinogen. And it's been difficult to treat until now. So I want to take you on a journey that zooms in to show you how this innovative product works. Scale alters our view of the world. From a distance, our home is a natural expanse of fertile land and rich bodies of water. But as the scale changes, we start to see the consequence of our impact on the planet. And in particular, the impact of industrial practices, which can do harm forever and less remediated, like the chemical 1,4-dioxane. Decades of unregulated industry has polluted thousands of our most precious resource, drinking water. This harmful chemical permeates into the soil and into the groundwater and eventually into the water that we drink. At a microscopic scale, 1,4-dioxane molecules are inundating the aquifers under our feet. But thanks to the detectives at Allonia, they sticked through billions of clues within nature and have made a breakthrough discovery in the fight against 1,4-dioxane. To quantify this feat and to give you an example of the biodiversity in nature, one tablespoon of soil includes more organisms than there are people on earth. Now let's meet our hero, a naturally occurring bacteria that has an affinity to source its carbon energy from 1,4-dioxane. This incredible discovery is a breakthrough, but must be engineered and deployed into the contaminated sites. We have completed this work, and we're now ready to wage war on waste, starting by destroying 1,4-dioxane underground. 1,4 D-Stroy is an engineered vessel that contains our hero bacteria and is deployed into the contaminated site at precise intervals and depth. Exemplifying the power of biodiversity, our hero bacteria joins billions of other bacteria that have not yet evolved to destroy 1,4-dioxane. Our hero set to their task of absorbing the microscopic molecules, turning them into carbon dioxide and water, rendering them completely harmless. In situ, the bacteria rapidly remediate the contaminant. Now the story doesn't end here. We are -- work with the bacteria continues. We are using biological engineering to enable the bacteria to fluoresce so it can be detected in the field. We're also engineering this to improve the efficiency and allow it to work faster. Our here -- our detectives have their sites now set on another target to solve PFAS. Using the same example and tools and methods that I showed before, we're working on remediating PFAS down to below 4 parts per trillion, which has recently been announced by the [ ETA ]. So to give you an idea of what that is, 1 part for trillion, it's less than 1 drop in an Olympics size swimming pool. These are precisely, the kind of challenges that Allonia was created to solve, so we can help return our planet to its natural glory and be transformative in nature. Stay tuned for more on the solutions that we're working on, targeting the world's toughest environmental challenges. So I'd love to connect with you today or at another time if you want to talk about how to reduce your environmental footprint or how we can enable regulations to help with bio solutions for the environment or just talk about your passion for reducing waste. Thank you.

Hi, everyone. I'm Anelise Reeves from Synlogic. And metabolites are all around us. They're in the air we breathe, the food we eat, and they're being made and modified by our microbiome right now. Most people never need to think about metabolites. But for people with disease, the wrong metabolite can have devastating consequences to their health. Take these metabolites for example. If they build up in the human body, they are toxic, and they cause diseases that have very limited treatment options available. The first one, phenylalanine is the cause of PKU or phenylketonuria. These patients can barely eat any protein in their diet, they can be physically weak and have cognitive defects. In the middle is [ L-theanine ]. This causes homocystinuria or HCU. Left untreated, these patients can go blind. And lastly, oxalate causes a hyperoxaluria, a very painful condition with frequent and recurrent kidney stones. And what's missing for these patient populations are therapies that can effectively reduce the concentrations of metabolites in their body. If I could have the video, please. So this is where Synlogic comes in. We have developed a platform based on synthetic biology to engineer bacteria into a new type of medicine that we call Synthetic Biotics. And the way Synthetic Biotics work is by combining a commensal probiotic microorganism called coli Nissle with pathway engineering to both sense and degrade toxins in the human gut and convert them into something harmless, like a biomarker, as a new form of therapy for these patients. So what you're looking at now is real human data after the use of our Synthetic Biotics. And what we've been able to achieve in the past year is between a 25% to 40% reduction in metabolite concentration across all 3 of these disease applications in both Phase I and Phase II clinical trials. And so what we're super excited about to do next is to take our lead program in PKU into Phase III clinical trials this year, and it's happening basically as I speak. So we're the first company to do this, to reach this critical milestone for delivering this type therapies to patients. And we're not just going to do this in the U.S., we are doing this globally. So please look for us. But remember, the metabolites are all around us. So we really think we're just scratching the surface with the potential of our platform by looking at metabolic diseases and validated biology. But what about other diseases affected by metabolites, diabetes, obesity, inflammatory diseases. What if we could treat these patients, these people with engineered bacteria, we imagine that we can. So if you want to learn more, please reach out to me, find me and follow the journey of Synlogic. Thank you.

Hi, and good morning, everyone. My name is Satoshi. So I'm very happy to be here today. So today, I will introduce our collaboration with Ginkgo Bioworks. So first of all, I'd like to introduce our company video. So could you start the video. [Presentation]

Yes, that is a nice video. So Sumitomo Chemical, in the past, we started [ 5 types of ] business to solve the social problems. Now we have 5 sectors in our business area. So there are a lot of social issues existing in our business area. So how to solve it. It's not so easy. So one solution is to find a good partner. The other one is how to get the powerful tool -- obtain the powerful tool. So that is kind of the trend of the chemical product -- production. So now the chemical industry is moving towards -- to the precise fermentation. Precise fermentation become a powerful tool to solve the social issues. So why a Japanese company, like me, doing such kind of the fermentation. So as you can see, the Japanese, in Japan, we have a traditional history of the fermentation industry. So we have confidence of the control of the microorganism to industrialize. So that's why I started the collaboration with Ginkgo for the precise fermentation to produce chemical product. So there are kind of the 3 products here. So one is agriculture plant growth regulator. So currently, due to the global warming issue, so the grape color become worse. So that's why we need the improvement. So using the plant growth regulator, we can improve it. So we started one project. The other one is for animal welfare. So [indiscernible] still the cosmetic industry are using the animal-derived product. But Sumitomo Chemical want to solve this kind of problem. So that's why, with Ginkgo, we started the development. And the last one is basic core chemicals, like [ phenol ] or something for engineering plastics or something. So that is the current 3 big projects with Ginkgo Bioworks. So here is 3 daruma. So in each project, we have a daruma. So daruma is, in Japanese custom, when we start -- when we kick off the project, we fill in the one eye. And when we success the project, so we can get two eyes in the daruma. So now already, we kicked off the two projects. So the soon we can get two -- both eyes in daruma and immediately, we can start the final project so that daruma has one eye. So finally, this is our final image. So much daruma with both eye we can get with Ginkgo. Thank you.

All right. All right. Thank you so much. Thank you so much to Allonia, Synlogic and Sumitomo Chemical. We will now take a short break as it -- we seem to be standing room only, I have been advised to inform you that we are now streaming the event from the mezzanine. And so if you have trouble getting a seat, when we come back, you might want to check out the mezzanine space, very cozy up there, very comfortable. We'll now take a short break, 30 minutes, we will reconvene here at 11:20. [Break]

Hello, everyone. Thank you. Welcome back from your break. Thank you very much for joining us. So I'm going to be talking about the agriculture program today, also known the Microbial Services. So what is going today on in the world of ag. Why are we here today? Why did Ginkgo invest in this new platform? So as you know, we are a growing population. So we are going to have feed our population. And we're going to have to feed this new populations on the amount of land that is not going to increase. And in a sustainable manner, meaning that we also have some finite amount of resources. And in order to do that, however, we have some different challenges, such as, for example, tightening regulations in the world across the globe, where our food is being grown, and as well as some climate changes, as we all well know, is also impacting how we can grow food. So what that means here is that growers need new solutions. So the solutions we have today in the market are now going to be able to help us grow in the food that we need. So growers need new solutions and they need consistent solutions and high-performing solutions. And we, at Ginkgo, we think that biology can actually help developing those type of solutions and be part of the solution and the systems that we have. So because growers need new solutions, that means that now innovators, so all of us here today in the room, we also have pressure because we need to be able to bring best solutions to the growers. So what that means here is there is different risks that innovators are going to be faced, either technical risk or regulatory risk or acceptance risk. And here at Ginkgo, we have developed a platform that can help you address all those different risks that the innovators are being faced today. So when you think about the value chain as to all the different steps, if you want, that innovators need to have in order to bring a new solution to the market, there are really 3 main difference -- different buckets here. So first of all, innovators need to have new starting points, differentiating starting points. Starting points are going to meet the demand from the growers. Innovators also need to have those starting points being able to perform at the performance that we need to boost all this productivity that I was talking about earlier. And most importantly, which has been very challenging so far, for the agriculture solutions specifically based on biologicals is consistency. So not only consistency in terms of the manufacturing, so making sure that you can develop products that are consistent from day in and day out. But most importantly, also consistent in their performance in the field. So we are going to have a lot of different environmental conditions across the globe. Those products are going to be sold at scale, and therefore, we need to be able to have very consistent products. So I'm going to drive you through every single of those different main elements, if you want, and give you some data and some insight on how we are addressing that. So the new starting points that we had here, so this is extremely important that we have differentiating starting point. So we have a very extensive microbial collections that we can actually use. This [indiscernible] collection has been heavily characterized, has been heavily also sequenced. And we are able to find starting points for really all the main solutions that can be found and are needed to boost the productivity that I was mentioning before. So think about your typical crop protection solutions, but think also about new emerging needs that we have today around carbon, for example, nitrogen and in general around nutrients. So the way we do that is we're going to be using genomics. And of course, we are also layering that with a bunch of different or mixed technologies, if you want, metabolomics, [indiscernible] in order to really find the diversity that we have in production and to be able to really address what we need and find some differentiating bonds. Then after that we pass all those different leads to a series of [ assays here ]. They are not going to be single assays. We're actually using selection index that are allowing us to pass those leads to a series of attributes, if you want. And we are really choosing our leads based on those attributes that are going to best fitted for what the client needs and therefore, what the growers needs. So in this example here, we have passed some of our leads through [ phosphate ], screening cascade here, which really allow us to have predictions and to really be able to identify leads that have the best productivity possible in order to then perform into the field. We also -- the second [ vertex ] that I mentioned was around performance, right? So here and is also related to regulatory risks. So I was mentioning that for innovators, regulatory is also like a [indiscernible] that they have to face when they are bringing new technologies to the market. And here, we have a wide variety of different tools that we can use from more like random metagenesis to something that is going to be fully engineered like what Emily was discussing today. So we are able to generate millions of different variants that are really going to bring this diversity that we need. Also, we need to have the screening capacity to be able to screen all of those variants. And we actually do have that with a new acquisition that Ginkgo has done through this encapsulation and the screening technology, which is really a high ultra-throughput wave of screening through millions of different variants to really find what we need in terms of improvement of [ product ]. The last [ vertex ] that I was mentioning was around consistency. So there is 2 pieces here of consistency. It's consistency of manufacturing. So we have a platform where we do have the knowledge and the know-how and we have people that actually have both commercial products to the market and to really being able to take this predictive fermentation process development and to really scale it up to manufacturing and also help clients to do this technology transfer to whatever manufacturing you need and at the scale that you need. The part of that fermentation is, of course, extremely important when you develop different products. We have experts that can develop formulations, not only from the prototype, but also from a commercial scale and a lot of different type of formulations from dry formulations to liquid formulations. We also have teams that are going to help with set of specifications to make sure that, as I was saying, all the products that are being brought to the market are extremely standardized, which is extremely important when you talk about biologicals in specific. So lastly, I also wanted to mention biology and field testing that we have. So that is really linked to the predictability of the [ assays ] that I was mentioning at the beginning in the lab. We have done a lot of work to increase this trustability between the lab and the field. We've heard this morning a little bit about the trustability on the human side. It's exactly the same for plants, right. We need to be able to, whatever we have in vitro or in the lab, we have to be able to make sure that it is working also in the field. So this is extremely important here. We can work on mode of action. We can work on product placement to really work with the clients and their own portfolio to really understand how new solutions is going to work with their own solutions, either as a complementation solution or as a replacement solution. And all of that is also with the regulatory team and a field trial team that can really help being a thought partner into all of the development. So I'm going to stop here and just reiterate that Ginkgo has now a very vertical platform that can address different type of food concepts from biological, from live microbes, to products that is being made by microbes to also plant traits. I think you have heard of the deal that we have -- that we just signed with Syngenta around plant genetic traits. And we can also help and support clients to the whole chain of what it means to develop a product, either through discovery, all the way to industrialization and manufacturing and also development. So I'm going to stop here, and I'm going to present our next panel. So we had a little bit of a problem, one of the panelists, unfortunately, could not make a flight. So in addition to Ruth Wagner, who is the Head of Data Science and Analytics, we also have Corey Huck, who is the Head of Biologicals for Syngenta. So Corey didn't know, but this morning, he thought he was a guest and here he is now being part of the panelists. So please help me welcome them, and they are going to be moderated by Nadia Berenstein, and they are going to be talking about the future of plants. Thank you.

Hello, everyone. Thanks for being here today. I am here with Corey Huck, who is the Head of Biologicals at Syngenta. He is a last minute emergency addition to this panel. So he doesn't have his head shot up. And Ruth Wagner, who is the Vice President and Head of Data Science and Analytics at Bayer. And I'm Nadia Bernstein. I'm a writer and a historian of science. So let's dive right in.

This panel is about the future of plants. But to be clear, when we're talking about the future of plants, we're talking about the future of life and especially our life on Earth. Plant problems tend to be our problem too. And so I'd like to ask -- begin by asking both of you, how we are -- some examples of how we're working with plants to solve and address some of the major challenges that are facing human life as well as plant life?

Great. I'll start since he's going to collect his thoughts. Thank you for joining me, by the way. I said I'm not going to do a one-on-one interview, so I almost ran out the door. So thanks for joining us. It's great to be on stage with one of our fiercest competitors, all right. So yes, it's good, right? We need someone from Corteva over here too. Anyway, I'll answer the question. So I think one of the biggest challenges that we face, and it's been brought up already today, is climate. And while agriculture is a major contributor towards gas emissions and things like that, the nice part is, as an industry, we can also use farming practices and science and technology to address that issue. And I personally, I like that because it's taking its responsibility. And it's something we care a lot about is the land and the health of the soil and the future of the planet and ability to feed people is super important. So there are different farming practices, things like special carbon initiatives, where farmers can change their practices, things like no-till, things like special cover crops, especially ones that can capture carbon, are things that farmers can do and be incentivized by. In addition, we can have technologies. I'm going to move this over a little. Technologies that we develop in the sciences. So things like Magalie talked about microbials and biologics, which you can talk about, but also genetics approaches to producing plants that do a better job of reducing the climate impact that we have in agriculture.

Can you -- Ruth, really quickly, can you give some examples of how the genetics of plants are being enhanced to make them more resilient to climate challenges or to growth challenges?

Sure. Some of the nice examples are on drought tolerance and withstanding floods. Things like floods for rice agriculture are huge issues. Droughts are becoming bigger issues in many places in the world. So genetic traits that can help the plant survive through those situations. Another one that Bayer is really proud of right now is our Preceon system, and this is short-stature corn. And this corn is growing several feet shorter than traditional corn. And what that allows growers to do is go through the field and apply chemistries, for example, pesticides exactly when they're needed and not just when the tractors can get over the corn because once it's too tall, they can't drive those tractors through. So being able to do things like adjust the plant architecture to make it easier for farmers to go through when they need to, helps this as well.

So helping plants so that they can -- so that -- so changing plants so that can help them. Now Corey, I know that your work is focused more on soil.

Yes. Yes. So I'll start with the original question around plants and why our plants are so important. Frankly, none of us would be sitting here today if it weren't for plant, everybody touched a plant or a derivative of a plant sometime today. So I have the opportunity to grow up on a farm in Nebraska. And the way that we managed plants, crops and everything to compete with them, then versus now is substantially different and continuing to change for all the things that Ruth just mentioned, mostly around climate change, continued to be a really big -- I can see it on our own farm, and I've had the opportunity to travel just in India a few weeks ago and really see the impact of the climate change on farming and what it means to produce a crop. So the need for technology in plants and the need to help plants become more resilient is really at the heart of what we try to do and something that I've spent most of my life doing. So yes, I think there's a lot of really great opportunities to really unlock the science of plants. And to your point, you just made, in terms of soil health we've always really looked at the plant itself, but there's really a deeper [ symbiotic ] relationship between the microbiome and the plant and the plant roots and the way they speak to each other and the way they depend and the interdependencies of each other. So there's a lot of science really to unlock to understanding that and how we make plants more resilient.

Yes. In our conversation before this panel, you described it in the way it was sort of like the difference between medicine that treats only the symptoms and medical approach that looks at the conditions that allow the whole organism to thrive and flourish. So I think...

Yes, it's actually something that -- there's been really a change in our organization, I think, across our industry as a whole, is we were very biotic-stressed to manage and focused in the past, symptoms. You have an insect, you have a fungus, you have a weed competing with the crop. Let's try to take care of that. And we're taking a step back and say there's maybe a little more to think about here is a more resilient plant, a healthier plant, can potentially help stave off some of those stresses. So that's -- I lead the biologicals group, and that's really what we're doing is looking at bioplant, biostimulants. Nutrient, it's efficiency. Love the technologies that Ruth mentioned as well as if that plant is healthier and off to a better start, it will have a better chance for, first of all, finding its full potential and second of all, fighting off all the challenges, whether it be climate change and everything else that's coming after it. It's more resilient plants. It's really what we're after at the end of the day.

Maybe can I add to that? My technical background in the sciences around data is on [indiscernible] and being able to collect all of that. And I've just been really impressed and it's sometimes also overwhelming in the last several years about how much data are out there that you can collect and try and process and interpret and being knowledged from. And so when you think about all of the data from the plants itself, it's metabolome, it's genes that are expressing its DNA, the proteins that are all in there on top of that, everything else that's going on in the environment where we have plant sensors and satellite imagery, all of this information coming in really helps us, I think, address the whole plants itself and also in the context of the environment that it's growing in or is expected to be grown in.

So how does this flood of data that is always -- and that always seems to be increasing? How does that shape your research? And how does that guide the new -- or sort of reshape the horizons as possible for what you're doing at Bayer?

That's a great question. And it's -- this is something that's not just a Bayer issue. It's true for, I would say, other companies, other industries and academic research as well as being able to make sense of all of the information that's out there and know what information and what data are relevant for the question that you're trying to ask and ensuring that you have that. And so there's this balance of collect as much information as you can about everything, but also needing to be able to understand that data, have high-quality data and make decisions based on it. So from my team's standpoint, I mean it's great. It's our job to be able to collect ingests, process, and make insightful decisions based on it, but also being able to store it and keep it and understand it for a long period of time. There's a lot of, I would say, stewardship things that come up, which is a totally different topic than today, but that's something we have to be mindful of.

Corey, how does your team or your group identify the problems that need to be solved? How do you sort of direct your energy and do research?

How do we direct the research? Yes. This is an area where predominantly been synthetic chemistry in the past. And then now, obviously, 2 collaborations with Gingko and with others, how do we expand our portfolio of technologies that we think might be beneficial to farmers and picking up off of technology and the data piece of this is absolutely critical. I think we have over 200 million acres annually that we collect data from all the way at the farm level globally. And that insight and that data can tie back into the type of data that Ruth just mentioned at the very scientific and research level. And what's really the -- as you get more data, you get more intelligence. And what we're finding is we can produce and build and identify better -- and discover better products, more effective. And one of the elements that we really spend a lot of time on and thinking about sustainability and what's the regeneration -- regenerative approach of that product, how could it work and then have all the data to support it. And another element that we're looking deeper into is soil science. Soil has been around longer than life itself and the amount of the information and knowledge we actually know about the science of soil and the microbiome and the interaction that microbiome is still a relatively infancy in terms of the scientific understanding of it. So understanding of this data, uncovering it and creating and driving towards new discovery and using different modes of action, we use a lot. So we're really looking at a lot more natural places like what Emily just shared in terms of microbials. We'll also look at fine chemicals and even biofermentation for pheromone. So there's a lot of really new interesting technologies that are being unlocked by synthetic biology that are really helping us in treating plants more effectively.

Right. I think that one of the things that agriculture and life on earth really is facing is the long-term effects of chemical fertilizers. Farming practices are overly reliant on chemical fertilizers and pesticides and thus, kind of leach the soil or deplete the soil. So how do the new approaches thinking about plant well-being and plant futures kind of deal with or address these kind of legacy problems of chemical fertilizers?

I would say there's a wealth of research going on right now in carbon sequestration and understanding how we can leverage plants for this. I even saw an article yesterday and it was on sheep's wool and how you can process things differently and do this in a more green way. So it's not just in the plants, but also in animal and other efforts, like you said, synthetic. Just some examples I mentioned earlier, there are covered crops out there. We have an example of covered crops that is a gene-edited pennycress product that's we're in pilot launches now, and this is one that people can -- farmers can rotate between their soy and their crop fields and it will grow. It will give them additional money. It is expected to be a great source alternative biodiesel and on top of it can sequester carbon. So there are some things like that, specific to the carbon capture market.

Yes. I might be a little more dramatic. There's 800 million people in this world that wake up hungry every day are affected by hunger. That's up 150 million from pre-pandemic and we're not going to have more land. We're not going to bring more land into production. So degraded soils are one of the most important things that we need to deal with if we're going to feed a growing population that's going to be headed towards 9 billion very quickly. So I don't think we have a choice. We must do something about overcoming degradation of soil and how we make plants more resilient to all that. And I think, fortunately, technology is generally an innovation, is usually the answer to that, and you can see that starting to emerge. But I really see it in the nutrient use efficiency type of technologies where everything from endophytes that fix the old nitrogen in the plant all the way to how do you utilize the existing nutrients, whether it be macro or micronutrients in the soil to make them more available for plants. So there's -- science is unlocking this and we're learning more about how it's all coming together. But the clock is ticking against us. And we need more science. We need more innovation and adoption. And soil degradation, I can tell you where we're at here in Boston, and we're here in the U.S. It's certainly there's some room to improve. But globally, it is really a super epidemic in some of the areas that a lot of work to do.

So this all sounds great. So what is -- what's standing in the way? Why aren't more farmers planting covered crops, as you mentioned, Ruth? Or why aren't more regions, more countries looking into solutions like the ones that you've described? And what are the roadblocks to scientific questions that we have yet to answer and social and critical lines?

I would say, I don't think anybody cares more about soil and soil health than the farmers do. This is their livelihood. It's what they're passing on to the next generation. And so I don't think that there has ever been some idea if I want to destroy the soil or ruin things. And so as technologies become available, it's important that they do want to adopt them and they do. I think there's a couple of limiting factors. One would be financial. I'm not going to get into that too much except to say a farm is also a business. And so some of these programs where there's incentivization to do this is really helpful where they can do work that protects the soil better, and it brings more money into their pockets and makes their business more successful is really important. But the other piece is the technology availability. A lot of farmers -- farming is done in rural areas, and there isn't the same kind of infrastructure that you see here in a city for equipment and technology to get there and reliable high-speed Internet to pass information along. I think that a lot of these technologies that are out there would be better used if they had better infrastructure and access to it.

Yes. Not much to add. I entirely agree with everything that Ruth said. There's -- one of the biggest challenges is education, getting the innovation on the farm. There's over 680 million farmers globally. And to try to get your technology or get a new technology that might affect soil health or affect plant resilience to that many farmers, it's not like going to a hospital system where you can just push it out quickly. It is a very heavy infrastructure-oriented business that we're in.

Right. How do you make this appealing and viable for smallholder farmers, especially in -- on the African continent, in South Asia. What are the -- how do we make this actually a viable option for people who are farming -- who are doing most of the farming in a lot of the world actually?

I can start. 3 weeks ago, I had the chance to be in India and met with our local team, and we have a team that are out talking to farmers every day. Not enough of them, but we're talking to farmers. What really got me, as I said, I spoke to a lot of these growers and since you're starting at such a low production base, when you were to speak to them, you would ask, well, what impact did our products have on your farm? And the minimum amount was 30% increase in output and yield and up to twice as much. So I said, well, what does that mean to you? Is that be like, that's the difference between my son having to stay home on the farm and not get an education versus having enough production that I can sell enough often to send my son off to be educated. So it's absolutely fundamental. And I think that's really the key of how do we get those technologies because once the adoption happens, once they see the impact that it can have, it helps the socioeconomic system as well. So reaching them is really, really, really difficult.

Maybe I agree with all that. And one thing I would add on is, and some of this is within some of our control and me, personally, appeal sometimes a little bit out of my control. But I would say things like the regulatory landscape, it's very complex. And so for a company to spend 10 units and hundreds of millions of dollars developing a product to get it out there in a way that is affordable for all farmers. It doesn't have to be expensive, but they have to be able to access it. And so even if we can develop a technology, if because of regulatory or infrastructure, it can't get to that region, we have a problem there. And so in a dream world, the simplified regulatory landscape would be helpful for getting technologies to everybody who wants to use them.

So Ruth, changing tax a little bit. As a historian of science, one of the things that you see is -- over history is when disparate areas of knowledge converge, you often see these kinds of explosions of research and generativity in new scientific fields as somebody who was trained in plant biology and now finds yourself in data science, can you speak a little bit about how these 2 fields are kind of converging? And what that makes possible?

Yes. That's a good question. And earlier when Nadia had told me about this one, I was excited because I told her the way she wrote it, kind of read like my career of talking about all these kinds of data and how do you put them together and how do you do something with it and then connecting that with modern analytics today. And so for me, it's been really fun watching organizations like mine and like other data science groups move from being seen as a service to actually driving innovation and product opportunities. So there's one way we look at it at Bayer is the spectrum of digital transformation. And on one end of it, you have basic process automation. So how do you take an existing process and automate it. And that's great, but it requires somebody to tell you, this is how we do it and then you automate it. And as you get more mature on that spectrum, you can start to do things like come up with new product concepts that can help. Related to precision agriculture, some things we can think about. The teams earlier were talking about AlphaFold and predicting protein structures and the new waves of that with OpenFold really pushing that technology but also things like -- when you think about promoters and optimizing gene expression, getting a plant to express the protein you want to see in the exact tissue, at the exact time is something that can't be done without these kinds of sophisticated analytics and nominations into our pipeline. And so I get really excited when we can actually look at a product in our pipeline and say we couldn't have done that without the data science, we couldn't have done that without the AI, and that's something I look forward to continuing to see in the future.

So it seems like the scale of data collected properly used can sort of contribute to both the speed of innovation, shape and guide research and work and also lead to a kind of increased precision of results. Corey, to kind of finish this up, can you think -- can you speak to the ways in which the increased precision that's available now is shrinking the dynamics of this field?

Yes, it's certainly having a huge impact. And I think Ruth really picked up some important pieces of bringing digital through the entire system all the way from discovery to the use on farm and making better decisions, which will drive better products. So I'll use a couple of examples of where we see precision having a big impact. We launched a product last year in Indonesia called [Indiscernible] and what it does is it's a biofermented pheromone. So instead of using a lot of applications in synthetic insecticide to control rice stem borer, we put out the pheromone that is disrupting the mating habits basically of the rice stem borer. So they go elsewhere and procreate, therefore, reducing over by half the amount of synthetic chemical that's needed. So the precision of that is really understanding how do you apply it and what way do you apply it, put the data analytics over top of that and assure that you have the right product at the right place, at the right time. And then in microbials, for example, there's a lot of opportunity. I think we're just starting to understand and I keep going back to the soil because I really think the soil can uncover a lot of the capabilities in a lot of the plant resilient issues that we currently have today. And climate change is only going to bring more upon us. So I think the idea is how do we actually use data to create better products. We can start to see early signs of it, but there's still a lot of ways to go.

Any thoughts on the sort of precision that's allowed by -- that's being made possible by data, Ruth?

We have only a minute left though. I think I covered a lot of it. Maybe just saying the ability -- I talked before about the insights and tracking these decisions. I think that's the most important piece. A lot of people here who work with data or any pipeline know the phrase garbage in, garbage out. And we really need to be able to avoid that situation and not get overwhelmed with the amount of information. And this is where subject matter experts are always needed. And some people worry about, can the computers just do it all and the algorithms do it all? And really, you always need people to look at over? This is true right now with ChatGPT being out there. It can give you wrong answers, and you have to have people who can decipher and distinguish a good answer from a bad answer. So we still have a long way to go, but it's nice to still have that connection to the data and information.

Fantastic. Well, I think on that optimistic note, we'll end this discussion about the future of plants and our own futures. And thank you all for listening, and thank you both for participating.

All right. Thanks so much. Thanks so much for the panelists. I am here to introduce another round of lightning application talks. But first, I need to ramp a little bit as we change over the stage. I have a scientific background. I'm a scientist by training. Now at Ginkgo, a lot of my work is marketing. I work in marketing, and I'm very good at it. I'm very good at it. And I'm going to tell you why. I'm going to share my secret. The secret for marketing, foundry biotechnology. It's one simple trick. All right, you ready? You're ready for this? What I bring to this job is jealousy. Jealousy -- deep authentic jealousy for our customers, for the people who get to build the cool technologies on top of this platform. That's the secret. I'd say, you thought -- what did you -- you thought I was going to say, "Oh, I talk to customers and I listen to their needs. I do that. I do that, right? That's day 1 s*** though. Everybody does that, right? Envy. When I talk to these customers, when I talk to our developers, I'm hearing about the amazing projects that they are bringing -- that they are bringing to life on this foundry. I burn with jealousy and envy. I think, I want to be you. I want to do that. That is amazing. This is a biological wonderland, and you are bringing it to life. And they can see that. They can see that. I get the crazy eyes at these meetings, and they can feel it. Now, sometimes it creeps them out, that's true. Sometimes you got to dial it back. You've got to dial back. But sometimes, that really connects. They can see through that jealousy, the passion and the whimsy and the just pure nerd energy, the pure nerd joy that drives this company. All right? That's my secret. That's my secret to marketing. So now we're going to hear from four foundry developers in this next session. We've got Lygos. Lygos uses advanced biological and chemical tools to convert low-cost feedstocks to high-value products. We'll hear from Eric Steen, CEO and Co-Founder. SaponiQx is working to deliver a secure supply of vaccines and adjuvants to improve vaccine performance. We've got Chandresh Harjivan, President and Chief Operating Officer. Ayana Bio produces and discovers micronutrients called bioactives for a healthier and more sustainable world. We've got Frank Jaksch, CEO. And Prokarium is leveraging the evolutionary advantages of proprietary strain of salmonella as a microbial immunotherapy platform. So we welcome Kristen Albright, CEO. So please welcome our foundry developers.

All right. Thank you. Thanks, Jake. Consumer pressure, government mandates and environmental degradation are fueling a massive need for innovation. And some of the biggest brands and companies are reinventing their product suites for sustainability and decarbonized supply chains. But these consumers, us and these companies will not settle for underperforming products. They need performance. And that's where we come in. I'm Eric Steen, the CEO and Co-founder of Lygos, and we're on a mission to accelerate the world's transition to high-performing products. This year, we're launching our commercial product, [indiscernible] in 3 key markets: home and personal care, agronomy and clean water. And I want to give an example of what this means in home and personal care. First, our product is biodegradable. Second, it enables less cost to the end consumer. And third, it's outcompeting or outperforming the #1 brand in the U.S., that generates over $2 billion per year in revenue. So we're really excited about formally launching this -- our product suite over the coming year or this year. How we do this? A little bit about our process at Lygos. So we address customer problems. We start at the end, right? We're head on with solving customer solutions. We do this in a vertically integrated 3-phase approach: design, develop and deliver. That results in sustainable solutions that perform better than the incumbent processes. So we think this is game changing. We're well resourced in the downstream parts of these buckets for both process application and product development to bring those solutions to market. And then finally, a core strategy of ours is to collaborate to accelerate. And what I mean by that is that we're partnering with some of the best companies in the world across this value chain. And so we're excited about our relationship with Ginkgo and other partners that help us do 2 things: one, get to market faster for our customers; and two, increase capital efficiency for our investors. We don't want to reinvent the wheel where lots of capital has gone into different parts of this value chain. We want to collaborate to accelerate. We're delivering sustainable, high-performing solutions to blue chip partners and helping realize the future where there's less toxic processes and safer ingredients for the world. We've got an awesome team. We have the tech. We have the strategy. We have a great set of investors, come find me, reach out to learn more. We're Lygos, the go-to innovation partner for sustainable solutions. Thank you.

So I've got 3 minutes and I have got so much to talk about, but I'll go quick. I'm going to talk about adjuvant. I'm going to talk about what we're doing with adjuvants. And then I'm going to say why it's important and why it's very important because we're at Ginkgo here and everything has to be very, very important. So for -- so adjuvants, what are adjuvants? So adjuvants are added to vaccines to make them more effective. It increases the immune response. There's not that many adjuvants around, but 4 or so. There's aluminum [indiscernible] cartilage. And then there's QS-21, it's a saponin, which we're focused on. Now QS-21, the source of it is Quillaja tree bark. The only source of it is trees in Chile. And this causes a lot of supply chain issues. And in fact, when I was working at operation work speed, I was at Health and Human Services and the secretaries are -- and the DoD talked about securing the supply of these trees in Chile and sent people down to guard the forest in Chile. So it was a bit of a problem. Well, we at SaponiQx can now make QS-21 in plant culture. So no trees needed. It reduces the supply issue. It's consistent supply, consistent quality. So we've kind of solved that problem, and we're doing deals with pharma now to give them QS-21 to put in their vaccine candidates. So that's what we've done right now. Now with Gingko, what we're doing is we're looking at cell engineering to make the adjuvant, QS-21, in cells and working with [indiscernible], it's a great team to be able to scale up and get even more yield out of QS-21 and then reduce the cost tremendously as well. We're also working with them on novel adjuvants, and that means really creating a toolkit right now. So we can develop new adjuvants that can be tailored towards different novel diseases or existing ones. So why is this important? Well, it's important because vaccines are so critical to health. If you think about somebody who is healthy, they have a lot of wishes, but if you're sick, you only have 1 wish to be healthy. And so what we're able to do is provide our adjuvants to different vaccines to make them more effective. Vaccination is considered to be the most cost-effective medical invention ever introduced. In fact, the CDC said that -- we're sure they're going to read this between 1994 and 2013, 20 years, vaccines prevented 0.75 million deaths exceeding $300 billion in direct medical costs and $1 trillion in societal costs. So we just see the value. So vaccines impact not just how long we live, but how we live, our productivity, our happiness. So that's important. But now why is it really important. So when we think about the disease that we've seen, we looked at COVID, it put the world at a standstill, right? And we developed the vaccine to COVID in a remarkable time. It was 1 of the greatest achievements of the U.S. government and for science. But you know what, the vaccine for COVID is not perfect. It's expensive. It's a complicated cold chain and the duration of protection is not that long. And the costs are very high, especially for the developing world. So by using QS-21, we could make these vaccines for COVID cheaper and more accessible. If you think about other endemic diseases, I was born in South Africa and when I moved to Canada, I was diagnosed with malaria and tuberculosis. There wasn't good vaccines for tuberculosis and malaria back then, and there's not good vaccines now for those diseases. There are vaccines, but they're not that great. So while vaccines are phenomenal, there's still a whole lot more we can do with those vaccines. Now if we think about the future. So what we're seeing more and more is outbreaks. We've seen Zika. We've seen monkeypox, SARS. This is going to keep happening. It's going be happening with more and greater frequency just because of the nature of human interface, et cetera, and climate change. And so what we're doing now is we're actually pairing up what [indiscernible] is doing in the biosurveillance aspect, we've got this radar. But if you don't have a way to respond to it, it doesn't really matter. So what we're doing is working through generative molecular design, and it's very much like the ChatGPT, the G is generative. So through novel data sets, the novel questions, just like ChatGPT, we come up with novel adjuvants paired up with these emerging pathogens very quickly. So pairing up the threats, but then building countermeasures to them very, very quickly. So I'm working with our colleagues at [indiscernible] and the team, [ Dan Helmer ], et cetera, to really think about how we can accelerate not just the identification of these diseases, but the response to those diseases as well. And then also, we're addressing an equity. So as I mentioned malaria, there's no vaccine for HIV. So globally, the worst inequity in health is health. So 5 million children die every year of infectious diseases. That's down from 10 million, 20 years ago, but that's still 5 million children every year that die of infectious diseases. So I don't think there's anything better we can do than using AI in generative molecular design to develop new adjuvants to make better vaccines, to make the world a healthier and happier place. Thank you.

All right. I'm so excited to share with you guys the cool things we were doing at Ayana Bio. We grow plants with outgrowing plants in the ground. And by that, I mean we use plant cell cultivation instead of agriculture to deliver or create plant bioactives. The picture that you're looking at here is plant cell callus, and this is the first step in creating plant cell lines. Also, I'm thrilled to report that today, actually this morning, we announced the launch of our first 2 plant cell ingredients, and I'm going to talk a little bit more about that at the end as well. So we exist to address 4 main problems. One is a global nutrition problem. And that might not be what you think it is. Second is to break the limitations created by agriculture. Third is to address the climate crisis. And then last is to solve supply constraints related to ingredient quality problems. All these factors have negatively impacted people's access to proper nutrition. Recent statistics published in Lancet show that in 2023, more people are going to die from poor nutrition and from smoking. That narrative was a significant feature of the White House Conference on nutrition that had happened last September. So why? The reality is that most people can't afford to eat fresh food. And as much as we might not like it, processed foods are a necessary evil. I happen to like Kraft macaroni and cheese. My kids like Kraft macaroni and cheese. I couldn't survive college without Kraft macaroni and cheese. But what if we could take Kraft macaroni and cheese and we could be formulated with ingredients that provide health benefits. If we want to fix the nutrient efficiency problem, we need to find a way to integrate nutrient dense ingredients into processed foods. The climate crisis has already started impacting our food supply. There's been a lot of recent media attention around crops like coffee and wine grapes. I'm not sure why coffee and wine had to be the canary in the coal mine here, but I would have chosen something else, but nonetheless, we have to deal with it. So it's estimated that half the land suitable to grow coffee will be gone by 2050. And I don't know by you guys, but the idea of coffee and wine being impacted by climate change, scares the hell out of me. So I need coffee and need wine. There's been a lot of recent news about heavy metal contamination in cacao and that's going to be a hard problem to solve with agriculture. So agriculture is going to be difficult to fix. But that is a problem that Ayana Bio can solve. At Ayana Bio, we use plant cell cultivation to improve access to health beneficial bioactives. So we're not trying to replace agriculture. These are some of the most nutrient dense plants, the ones that you're looking at here, but they're also some of the most expensive. We can bring these ingredients to market in a more sustainable way and at a more reasonable cost. Earlier, I mentioned the launch of our first 2 ingredients. We're getting quite a bit of media pickup these days on that or actually this morning. So at the reception tonight, you're going to be the first to try Ayana Bio based plant cell ingredients and [ Jason Kakoyiannis ], they're not going to be nefarious powder, bags of powder, so don't worry about it. So you'll have a choice of 2 different drinks tonight. One is going to be -- the first is a Manhattan spice featuring plant cell echinacea and the second is zesty lime aid featuring plant cell lemon balm. So try one of them. Hopefully, you'll try both of them. I hope you guys enjoy them. Members of the Ayana Bio team are going to be here all day. So if you want to talk more about the cool things we're doing at Ayana Bio, feel free to find me, some of the other team members around, and we'll be glad to talk to you about that. Thanks.

Hello, everyone. Today, let's take a journey back to the late 19th century when William Coley made an incredible discovery. So Coley had a patient that developed a severe bacterial infection, but also had an inoperable tumor and the tumor went away. He went on to treat over 1,000 cancer patients with bacteria with astonishing results. But what had happened during the time is chemo and radiation we're taking off in the background and the world focused on development of chemo and radio. It wasn't again until the 1930s it was a noted post autopsy that patients who had active tuberculosis infections had lower frequency of cancer. So let's quickly fast forward 50 years or half a century later to when mycobacterium bovis or the BCG vaccine became FDA approved as the first cancer immunotherapy. Now despite BCG's success and Coley's success, DNA recombination was still in infancy during this time and pharma began to focus on antibody development. Bacteria, again, was left behind in the dust. Today, Prokarium aims to change this trend. With our lead program, we have a live salmonella bacteria that's entering clinical development this year, and we're very excited to shift a decade-long treatment paradigm in bladder cancer that has been untouched for the past 40 years. But our ambitions do not stop there. So Prokarium's vision is to revolutionize the immunooncology field with living cures. Our mission at Prokarium is to reprogram bacteria just like you would a computer to create a bacterial biofoundry in the human body to deliver therapies in codes to tissues and cells with specific information and specific tasks. We plan on delivering this with our partnership with Ginkgo, who is helping us create a vast library of synthetic bespoke circuits for us to unlock the potential of our platform to build, test and grow the next generation of immunooncology therapeutics. Now Prokarium is challenging the status quo of medicine today just as Coley did, reminding us that what may not be possible today can actually be made possible tomorrow with synthetic biology. So please join Prokarium and our team as we jump on the next wave of immunotherapies.

Thank you, everyone, for your time today.

All right. Let's have one very warm round of applause for Lygos, SaponiQx, Ayana Bio and Prokarium. Very cool, very cool. All right, and that's lunch, people. That's lunch, people. We will meet here again at 1:30. Enjoy your lunch.

Hi, everyone. Welcome back. My name is Shawdee Eshghi. I'm excited to tell you a little bit about our mammalian platform that we spent the last 4 years building here at Ginkgo. Okay. So the advances in synthetic biology are really revolutionizing medicine today. And this is really one of the reasons why about 4 years ago, the leadership at Ginkgo decided to move into therapeutics. And these examples here are what -- some of the examples from different new modalities that are changing the way we treat disease. So in cell therapy, we have a CAR-T revolution, where Emily Whitehead here was just a toddler when she had failed multiple lines of therapy and her parents very bravely enrolled her in a clinical trial where her own cells were taken out of her body and reprogrammed to attack her cancer and put back in. And now 11 years later, she is thriving as a teenager and living the life like a -- that all parents want for their kid. We see in AAV therapy, the ability to correct a broken gene. So here, [ Jack Hogan ] had a hereditary form of blindness and he was 13 when he received an AAV therapy to deliver the corrected form of the gene. And this story just really is inspirational to me. He had been losing his vision. He couldn't see in dim light. In just 3 weeks, after the single AAV treatment, he was able to ride his bike at night, which was something that he wasn't able to do at all before. Within CRISPR, we see the ability to correct sickle cell disease, which is a debilitating disease. [ Victoria Gray ], here was unable to have a full-time job or to care for her children and now she can do that. And then in stem cells, we see a lifelong diabetic, [ Brian Shelton ], who now has his blood sugar controlled normally with a pancreatic cell transplant. So these are pancreatic cells that are made from iPS cells or stem cells that now produce insulin. And so he is free from all the finger pricking and insulin injections that a diabetic has to deal with. So these are remarkable demonstrations of how across different new modalities, synthetic biology can transform lives one at a time. And you may have heard, some of you that the mission of Ginkgo is to make biology easier to engineer. And for me, personally, what that means is so that we can have more of these success stories. So to extend CAR-T beyond blood cancer to more challenging tumor situations; to make AAV work not just in the eye, which is a privileged immune environment but to all tissues in the body. And what that really means is improving the efficacy and the safety of these medicines as well as the access critically. So I'm happy to sort of unveil today the 3 therapeutic service areas that we are actively working on in our mammalian team. I often tell people that our mammalian group, my experience has been very much like a startup within the bigger Ginkgo ecosystem. And so we're sort of coming out of cells with RNA therapeutic services, cell therapy services and AAV services. And these are all very different therapeutic modalities but what they all have in common with each other and with the rest of Ginkgo, all the plant stuff and the burgers and the fragrances is the common infrastructure and approach to library design, to high-throughput screening and to integrated data analytics and model building. And so again, what's common is high-throughput exploration of design spaces, genetic design spaces. And what I want to show you now is some vignettes from each of these 3 fields that demonstrate how we're leveraging our capabilities in, again, high throughput acceleration of genetic design spaces to address the critical challenges. Okay. So starting with RNA therapeutics. RNA is a really attractive therapeutic modality because of it's easy programmability. You've all seen that with your multiple COVID vaccines, right? But that comes with trade-offs for durability, stability and in other properties. And one way that we're approaching these trade-offs at Ginkgo was to explore circular RNA, which just by the nature of its circular structure has -- addresses some of the stability and durability problems. And just as a little teaser, you can see the plot in the bottom right that shows the extended expression time of the -- the lines on the top are all the circular RNAs. These are different designs made at Ginkgo, relative to the traditional linear capped mRNA, which is the red dotted line that dives off pretty quickly. So -- but this is a relatively new therapeutic modality. So there's lots of sequence space left to explore. So how to extend that time line longer? How is it different in different cell types? How is it different with different payloads? How can we optimize the translation? So RNA is just step 1. You have to turn that into a protein and that's driven by different sequence elements and different cells. So together, we've built a really world-class team to look at these questions and as part of that, we welcomed Circular's biotechnology last fall, who brought decades of RNA experience to add to our code base and our methods for circularization. And I also want to point out that our work in therapeutics doesn't end at discovery. Critically, we also work in manufacturing process optimization because that's really what's going to turn what is a technology into a drug, right? And in RNA, for example, we just signed a -- announced a new partner in a novel mRNA manufacturing technology with Sensible Biotechnologies. And I'm hearing in-house with our own circular RNA platform, we're getting over 80% purity, which is really a high bar for these advanced therapeutics. Okay. So continuing with the theme of exploring untapped biological diversity, I want to do a little bit of a deep dive on our CAR-T platform. So CAR stands for chimeric antigen receptor. And what you really need to know about that is that it has an external part that engages with the tumor and an internal part that drives signaling in a killer T cell. And there's 5 of these that are approved today. And what's just fascinating to me is that these 5, all have 1 of 2 different signaling domains. They're all the same within 2 choices. And furthermore, this is the same design that was the first one that worked 20 years ago in the academic publication. And but since then, our ability to read and write DNA has progressed dramatically. So if you were to start the question today of how would you design a CAR, you might do something differently. And here at Ginkgo, this is exactly what we've done, is harness our ability to make big DNA libraries and to use sequencing as a cheap and efficient readout. And so here, we made not 1 or 5 or 10, but 10,000 unique CAR designs, they all differ on their inside signaling parts. And we put these into T cells and we do an assay that mimics what happens in a solid tumor. And each one of these CAR designs has a DNA barcode. And we can use our sequencing capability to now read out the effect of each one of these 10,000 different designs. And then after we get that, we take them and we put them in head-to-head tests against sort of the standard therapeutic FDA-approved CAR design. And in our first time of doing this, we found a number of designs that are meaningfully better than the sort of first prototype designs from 20 years ago, which are actually approved therapeutics today. And so now we're able to continue and iterate on this library design, using some of the methods that you heard Emily Wrenbeck and the Protein Engineering team, talk about this morning. We can build ML model. And at this platform that we've built, there's different ways that partners can access this now. We can -- we've reduced the screening to practice. We can just do this again and give you new results in specific context within a short period of time or you can just take the designs that we've already validated and use them in a new CAR-T context. And then we've built this platform for this problem specifically but it also can be extended to similar problems. So we can make binders -- libraries of extracellular binders or do this in specific different cell types, let's say, natural killer cells. So we see this as a very flexible platform that we've built within our team. And just to kind of double-click on what diversity and scale really means to us. I just wanted to walk you through a schematic of what a typical Ginkgo experiment looks like for us in CAR-T. So I talked about the 10,000 different signaling domains. But these are now put together with different CAR binders of different affinities and then we test them in T cells from different donors. That's a really important thing to do because my T-cells are not going to be the same as somebody else's. And then we test this against multiple tumor cell lines with a high replication factor. So for us, one experiment is 720,000 unique tests. And so this is how we can really explore the corners of biological diversity to improve upon functions. Okay. And Ginkgo is nothing if not adaptable. And I think for these kinds of advanced technologies and advanced therapeutics, that's especially critical where some new breakthrough from an academic lab or some really important clinical trial results can shift the focus in these fields. And so for CAR-T, we've built a suite of services that address not just CAR-T as we know it today, which is called autologous CAR-T, taking a patient's own cells and reprogramming them but the future modalities of what CAR-T may look like in the future, including allogeneic, which means taking a healthy donor's T cells and reprogramming them, starting from stem cells or actually doing all the reprogramming in the body. And so there's a number of services on here. I won't talk about all of them but just to highlight some of what I think are the most exciting things coming up for us, are mining for a novel gene editor from our proprietary metagenomic databases and actually screening these in relevant cell types. And taking a new approach to what's called immune cloaking, being able to hide transplanted cells from the immune system to prolong their function over, hopefully, what can be decades in some cases. Okay. I'll end by introducing our AAV platform, which is, in many ways, the most mature expression of our vision for end-to-end capabilities across discovery and manufacturing. Today, through our internal work and also our recent acquisition of StrideBio, we really have all the pieces that you need to put together to make an effective therapeutic. So this includes the capsids that will determine the cell type specificity. The promoter can also add another layer of specificity and importantly, control the strength of the transgene that you're expressing. We can -- we have a number of technologies in manufacturing as well. And this is really how our mammalian team got started several years ago with just one sad person here on stage heading all day. And now we're proud to say that we've automated most of this work to tens of hundreds of thousands of automated AAV experiments and measurements across multiple modalities. And we're still scaling this with the opening of our newest Bioworks called Bioworks7. And if some of you come on the tour, you'll be able to see that and some of the new work cells that -- or robots that we have put in place there. And I'll end by introducing Kennon Smith, who comes to us from StrideBio to explain their unique evolutionary approach to capsid discovery and design.

Hi, everyone. My name is Kennon Smith, Senior Engineer II at Ginkgo Bioworks and formerly Head of Capsid Engineering at StrideBio. While at Stride, my colleagues and I brought together our combined structural and functional understanding of the AAV capsid in order to try and tackle preexisting immunity, tissue tropism, manufacturability and other challenges in the AAV gene therapy space. And I want to talk to you guys today about the engine that results from that and some of the capsids that came out of it. The Stride engine utilizes the structural knowledge to identify areas of highest impact across the surface of the AAV capsid with roots in modifying the AAV antigenic surface in order to generate synthetic libraries of AAV variants. These libraries have been evolved through multiple animal species. They're across species method in order to enrich for cross-species compatibility in order to try and derisk clinical development. The resulting platform enabled the discovery of a portfolio of promising capsids, many of which directly address the aforementioned challenges as well as modulating other functional areas, such as liberty targeting, potency and cell type specificity. In addition to these select candidates, the platform also resulted in a wealth of untapped library data across AAV capsid variants, across different tissues and different species. And these StrideBio capsids are currently divided into 3 tiers of development. The first tier includes fully developed capsids with extensive data that includes large animal models such as NHP and pig. And these capsids are ready to go for partners to match their desired indication or application. Tier 2 capsids have entered the first stages of testing and have demonstrated profiles differentiated from their parental backgrounds with initial in vivo or ex vivo data. These 2 tiers, Tier 1 and Tier 2 exhibit a wide range of diversity in their tissue tropism, for capsids targeted to the CNS, to the heart, to the muscle, to the liver and even to ex vivo targets such as T cells. In addition, some of these capsids have exhibited potency higher than their parental serotypes such as the Tier 1 capsid [ Stride47 ], discussed at ASGCT last year. This capsid has been shown to strongly transduce the kidney and at higher protein expression, 2 vector copy number ratios than the parental AAV, indicating a more potent capsid. This last tier of capsids, Tier 3, is made up of capsid libraries and preexisting evolutionary tissues from StrideBio's programs. This tier of capsids represents the largest well of untapped potential and contains variants ready for discovery and for development. So with the acquisition of these tiers, we at Ginkgo Bioworks have access to an unprecedented depth of data around the structure function relationship of the AAV capsid. We're uniquely positioned to take immediate advantage of both the capsids themselves in order to enable the ambitions of our partners across the gene therapy space but also to use the data to continue improving the capsids for our partners. Ginkgo's institutional capacity for machine learning and automated high throughput screening will allow us to pursue and build upon the assets from Stride, utilizing our pre-existing work in the gene therapy space to expand upon the capabilities of the STRIVE platform. For instance, Ginkgo's ability to engineer regulatory elements may enable more sophisticated and refined applications of Stride's capsids with respect to tissue tropism. So I want to take a second and take an example of one of the capsids designed using this platform, the Tier 1 capsid STRIVE 5, and I want to discuss its unique characteristics. So STRIVE 5, in particular, was selected for its cell-type specificity, liberty targeting character and manufacturability. And this capsid has been extensively vetted for its biodistribution and expression in mice, pigs and NHPs with multiple transgenes. So the data you can see here, which is discussed at ASGCT 2022 shows STRIVE ability -- STRIVE 5 ability to broadly target the CNS after ICM delivery and NHPs, but specifically to target neurons over astrocytes. -- as shown here by Newen and [indiscernible] costaining. When we characterize this transaction by [indiscernible], we also see that the STRIVE 5 capsid produces up to a tenfold more protein than the wild-type AAV9 capsid in areas such as the premotor cortex and cerebellum, despite a lower vector copy number in those areas, again, indicating a more potent capsid. In addition, STRIVE 5 has also exhibited a 1,000-fold lower transduction by vector copy number and protein ELISA in the liver after both IV and ICM administration data represented here again by IHC. So higher potency of transduction in the CNS after ICM administration coupled with lower transduction in the liver is exactly the type of safety improvement that Ginkgo looking to unlock through the refinement of StrideBio's capsids. And then last and certainly not least, is manufacturability. So STRIVE 5 capsid has exhibited scalable manufacturing with over 10 transgenes and has produced a high-purity vector from R&D grade up to GMP Phase I/II. So the selection of high-yielding capsids is actually a built-in component of the STRIVE platform, as we mentioned previously. And these efforts really neatly complement Ginkgo's successes in engineering the other components of AAV production and its manufacturing systems. So in total, with this acquisition and continuation of the Spirit of StrideBio's work, Ginkgo has now realized its ambition of a full stack shop for AAV gene therapy from capsid and transgene through manufacturing. Thank you guys very much, and I hope you enjoy the rest of the talks.

Hi, everyone. I thought we would be introduced, but we're not. So I'm Michael Specter, I write about synthetic biology and the merger of information technology in the biological world for the New Yorker. And this is why you're here. Renee Wegrzyn is the Head of ARPA-H, a new and amazing government agency that wants to disperse money to make health problems disappear...

We're going to down regulate to a handheld, which is probably where I should have started. So before I even asked the first question, I just want to make a comment based on a conversation that Renee and I had last week when trying, -- I was asking like what is the scope of ARPA-H. And she said, "Well, I don't want people to give me these nickel and dime proposals." And I said, what do you mean? And she replied $100 million proposals are acute. But I'm looking for something substantial. So when a federal health official says that you better listen. So my first question is, how would you actually describe what you're doing? What's the mission of this new organization? And a lot of us have heard of DARPA. We know you were there. Is it a health version? Is it different?

Yes. So our mission at ARPA-H is to accelerate better health outcomes for everyone period. And so health outcomes is how we're going to measure our progress. This is really what investment can ARPA-H make. We're only an investment agency. We're not bricks-and-mortar labs, but what can our funding accelerate to actually improve people's lives and standard of care? So you should think of ARPA-H as a place where there'll be a transaction as part of the health ecosystem to create a little spark. And so DARPA was launched in response to Sputnik. We don't have a specific Sputnik moment for ARPA-H, but it's really this energy around innovation like you're hearing about today, but also this urgency around crisis and health, whether it's a pandemic or climate crisis that are making us more and more frequent. And so that's the urgency by which we're trying to achieve these goals, and I know we'll talk a little bit about the model and how we're going to pursue that.

How does it actually work like your approach to risk taking, is it like DARPA? Is it completely different? Do you approach problems in the way that we're used to funders approaching them?

Yes. I think when you think about ARPA-H, it's important when you talk about risk that we're not taking on risk to the patient or safety risk. It's risk in terms of technological risk. And so our approach is to make sure that we're funding projects at the level where we know you're well resourced, you have all the capital equipment that you need to do the job, you have the team in place and the funds will help you buy down execution risk and so that you can really only focus on that technical risk. Some of the ways that we can approach that, I really -- I've only been there for 6 months, by the way. So we're still standing up the organization. But part of that was working with the Congress to make sure that we had authorities in place that we could be uncompromising on so we could take risks. Some of those include things like exempting us from some NIH policies, not things like ethics or safety, but study section. We don't make decisions through a study section. Our program managers are the sole decision-maker. So by allowing a person to be the decision maker, you can take on some of that risk. Our budget is small. It's only $2.5 billion when you compare it to HHS, which is more than $1.7 trillion, it's a very special place for us to be focused on these high-risk projects that NIH won't take on, HHS want to take on. And also, frankly, industry can't take on.

I was interested to hear the salary situation is different in ARPA-H than it is at other agencies, too, right?

That's right. So maybe I can walk you through some of the authorities that we locked in. So for our program managers and salaries, I can administratively determine physicians. And what that means is that we can go outside of the GS scale. I can hire somebody directly. So if we love somebody on a Friday, we can bring them on board on a Monday. We have our Chief Operating Officer, Joe is here today. This is day 8 on the job for Joe. And once we found Joe, and we knew we were going to be a great fit as an organization, as a person, we got them on board very quickly. And we can also be flexible in our salary. So we can hire up to the President's salary with my favorite trivia questions, anybody know what the President's salary is? No. Well you should find out. We can hire up to that level. And so it's $400,000. And so this is really to attract people to become program managers because we want to hire these people because they are amazing CVs. They can think and act like a CEO, very actively manage their program, but we're asking them to bring a big problem in health as they want to solve. It's a chance to change the course of your field. And so come in for 3 years, there's this urgency to get that done, and then you can go back to academia or to industry or to government what you were doing before.

Let's roll back one step and maybe you can explain the actual -- it's sort of a bottom-up managing system, which is unusual, and how does that work? And what are program managers, how are they found, hired and what do they do? And how much autonomy do they have. Okay, I'll stop.

Absolutely. So ARPA-H is a very unique funding model for government. So a lot of people ask me, are you going to fund this? Are you going to fund that? That's up to you all, right? Our ideas come from you responding to our open broad agency announcement, which is live now. So this is a place if you go to our website, arpa-h.gov, you'll land and find this open broad agency announcement where we want to hear your ideas and your projects that you want us to fund. So that's one way that we get ideas. The other is through our program managers. We are building this organization and hiring these people with a big idea and health that they want to solve. They come in-house once we select them. So the application process is literally you frame your problem in a set of questions that we've adopted from DARPA, the DARPA model, the Heilmeier questions. the George Heilmeier was one of the early directors of DARPA. So it's what are you trying to do? What problem and health are you trying to solve? How is it done today? What's new in your approach, that's where you bring your technical expertise forward. What are the risks? How much is it going to cost? How long is it going to take? And then what are your milestones for progress. We've added 2 additional questions at ARPA-H, and this is really leaning into the health side of things, which is how are you going to address cost, accessibility and user experience. And so there are a lot of product people in the room or customer and user experience is going to be key at ARPA-H that we're going to develop devices, therapeutics that people actually enthusiastically want to adopt. We don't want to assist it on a shelf. So program managers, I will not give them a bank account unless they can very clearly articulate how they're going to address that. So supply chain, think about costs as well. And then the last is, how might this program be misunderstood or how might misinformation be created around these like wild new technologies that we want to pursue. And we want to make sure that, that isn't what ends up being a blocker for us. We want to make sure we're communicating that. So it's a simple application process, but an intensive for a program manager has to articulate their problem that they want to solve through the Heilmeier questions. The main vitamin for job talk. And if that goes well, they come on board and they hit the ground running. They're going to be there for 3 years. So they have like 3 months to launch this program announcement where they look for groups of solvers to solve their program. And then they put a portfolio of solvers together and then very actively manage it. They can add more funds to teams, they can take funds away if somebody is not being successful. So that's sort of how it works. And so if you have a great idea in engineered gut microbiomes, for that to be a program at ARPA-H, you need to have a program manager on that concept. So there's 800 people here today. I ask each of you to think of 2 people in your network who could be a great program manager that is a champion for your topic, and we want to hear from them and their ideas, and that's how we're going to create this flywheel of programs people and projects that ARPA-H can catalyze going forward.

Great. So you're -- let's say I'm a program manager and I'm managing a microbiome project. And it's going well, but I'm running out of resources. I can just come to you and say, "give me more"?

Well, it's not that you're just running out of money. If you see -- this happens all the time, you project this 4-year program that you're working on and 2 years in there's a new technology that emerges from somewhere else and you're like, wow, if I have that, I know that this can accelerate better health outcomes if I have this the program manager will work with you to try to make that happen. If there's a group that hasn't reached their milestones, it's not that it's not great work, they're not successful. But if they're not reaching those program goals, they may not proceed to the next step. So very, very active in that way.

So what you're describing is the thing we always refer to as a sort of moonshot program? Do you take these giant bets and maybe some of them will come true. Can you describe particularly in the areas of synthetic biology, some of the ideas that might be good moonshot type approaches?

Yes, I really want to hear from everybody here. We're here all day. I want to hear your big moonshot ideas, but what we did at ARPA-H was we've established 4 technical focus areas. These aren't meant to be prescriptive, -- but there are kind of areas and helps that we think with further investment or really advance the state of the art. So I'll share those with you and then you kind of think about how synthetic biology can align with those. So one is health science futures. These are the tools, the technologies, the platforms of the future, whether it's mammalian cell engineering or the next mRNA platform that can address a number of diseases. So we are a disease-agnostic organization if the program manager has a new platform program, they may bring on a team that's working on Alzheimer's. They may have a team that's working on cancer but we really want to advance the state of the art for all of those projects. So that's in the health science futures. A lot more fundamental research, I would imagine, happens in that office as well. Scalable solutions. So this is whether you're getting the scale of 10 doses to 1 billion doses or something, lots of Synbio opportunities there, obviously. And then the scale of meeting the American people where they're at. So how are we getting things to people's homes that you can actually start to deliver care that could also be, let's say, through engineered microbiomes. But there's more and more rural health care centers that are closing and so we have to start solving this at the home going forward. Proactive health just simply keeping people from becoming patients in the first place. So diagnostic detection behavior, social science projects that are going to be important. Those are all kind of like products and key types innovation-focused offices. The last office resilient systems is more of a system-level integration office. And so there's many, many technologies that are represented in this room today. A lot of them are in silos. What if we integrate them to create a greater hole, and we added a data layer that's machine-readable that we can then advance the state of the art. There are types of investments and focus areas that we've laid out.

Well, along those lines, every presentation today has had at least something to do with artificial intelligence. And I'm wondering how you see AI emerging with synthetic biology in health care? Is that going to be an important part of what you fund?

Absolutely. So there's a really amazing opportunity that we have at ARPA-H where I'd like to think of us right now as a minimum viable agency. We just have the people in place to do contracting and HR. And we have a chance to build a data system that is really going to be exemplary for health care. It can be best-in-class in government. And that's something that we want to do really well. And think about if every program that we had, had machine-readable data that's associated with it, that's accessible, that's open access, where it's appropriate. This is something that doesn't really exist at scale. And we can build that in from the beginning. Some other interesting innovations, I think, are opportunities to take negative data in some ways. There's a lot of data that's failed around drug trials, et cetera. that kind of remains behind closed doors, right? So what if we can mine some of that data in a way that there might be learnings to advance the state-of-the-art for patients. Again, I'm not going to start programs in this area, the ideals will come from this group. But I think we're wide open on that. And I want to like just to alert you to another opportunity that's open at ARPA-H right now. We have this kind of unusual situation where Congress has told us, you're going to have 3 geographic sites. And by the way, only 210 employees. So we're very lean management. And so how do you use those 3 geographic sites? I mean we thought about it, there's no 3 geographies that will really represent all Americans, those are our customers. So we leaned into what we decided to do was a hub-and-spoke model where we really use those sites strategically to make sure we graduate programs out of ARPA-H and really access the customers that we need to reach. And so we'll have in the Washington, D.C. region, address to be determined, a stakeholder and operations hub. So we know FDA CMS have to be at the table, to be reimbursed, we need to be regulated. So we'll establish one hub there. But the others are being completed right now. And so what we ask the folks is to -- we have a site, a hub site that's devoted to customer experience, so we want to be able to get to our patients. If we have a program manager that has a pediatric prosthetics program, we want to go to the hub site and make sure that we get a population that's representative of those that were there to serve. And if we have an in-home geriatric aging-in-place program, the hub site needs to go out and pull in spokes that will allow us to work in all these different geographies. Layered on top of this, coming back to AI, I was going to get back to it, was we're going to have a data layer there that we don't have to start from scratch every time you start a program. We can lean into this network. Jason mentioned this earlier. You don't need to start from scratch every time you start a new biological program. We don't want to start from scratch every time we fund a clinical trial for every time we start a program. We can -- the system can learn and be a smart innovation hub. And I think I'm excited to see what AI applications will come out of that.

Not that I'd never run a company or anyone would let me, but there are a lot of people out here kind of scrappy startups. And I think it's fair to say they often have trouble dealing with the government. But you kind of have to work with the government in these fields. Are you going to make that possible?

Yes. So we have unique to ARPA-H, we have a whole office that's devoted to transition. So if you think about Department of Defense, you have a missile program like the DoD is going to make your missiles for you at the end of the day, right? This doesn't exist in health care. There's no government organization that's going to make your product for you. So we rely on all of you to be the advanced developers of our platforms that we fund and develop, but how do we make it easier to work with us. And so this office we set up. One of the ways we're doing it is through a partnership intermediary agreement. This is government speak for having a not-for-profit third party that knows how to work with small businesses and makes it simple. If you think of a university that has like a grants office, so Professor whoever doesn't need to learn everything about the government, the grants office has got it covered. We want this entity to be the grants office for businesses so that we can just make it super simple, very low friction to work with the public sector. This helps our program managers, too. So before they launch a program, they can work with this transition office and say, is there a market for pediatric prosthetics. And so they can learn a little bit about that, who are the investors, who are the players so that we're pitching the right program. But learning from those investors, what do we need to derisk in order for it to transition and survive, survive in the wild is what our Head of our transition office likes to say. And so we are like fiercely committed to things surviving in the wild through that transition office. And so we want to hear what the investor risks are because the ARPA-H can help derisk that.

Can you and maybe slightly more detail, explain the decision-making process that would go into one program? I mean, are you the ultimate decider or is it a panel?

It's very lean. There's basically 2 approvers. So the program manager will first develop their concept and pitch it to the Mission Office Director and those 4 focus areas that I mentioned. That's the first green light, and then it will come up to the director's office. There's a -- we have an S&T Board that provide their perspective and review and kind of give their suggestions, but ultimately, I'll decide if the programs go forward or not. So with that really lean decision-making structure also lets us take those risks. I should say, and we talked about this in the green room a little bit. We're established within NIH, but I report the Secretary of Becerra. And so this is -- and he's the head of HHS. And that was a really important piece of this so that I could take those big risks. I don't risk -- we work -- we have a wonderful partnership with NIH, like world experts in so many places, but they won't be a gateway to decisions that ARPA-H makes.

I'm wondering if you're worried at all about -- so you have these truly great ambitions, but I look at some of the moonshot industrial proposal. It's like Google ads -- it's all moonshots. Some of them have sort of worked [indiscernible], but it's a private company. So no congressmen is going to be running in and screaming at them, at least it won't matter if they do. You're in a different situation. So how do you -- like what does success look like for you?

Yes. So we have engaged a lot with Congress, and I ask every 40 members of Congress have engaged now. I feel like explain what this is and set expectations, and I ask them what success is and to them. And what's really interesting is we have unanimous bipartisan support because everybody has an elder parent that has Alzheimer's disease or a child that has a rare disease. It helps, touches everybody. And so that's -- I was very pleased to find out that everybody wants us to be successful. But what success looks like is different to folks. I think for us, in the near term, success is getting our authorities so we can make decisions. And so like we've done that, really excited hiring the right people, we need to have a portfolio this year. The 20 program managers that I hope to hire this year should represent 20 totally different projects that really start to unlock the imagination of what ARPA-H can catalyze. And 3 years from now, it should look totally different. That will be a success for us. If we have 400 cancer projects, we failed. If we have some cancer projects and some technology focus, that's a success. And then this is going to sound maybe a little funny, but success will also be if we fail. We have to have room to fail. That means that we're taking those moonshots. We want to be very transparent in those failures and learn from them as we go forward. So yes, keep me -- hold me to that. Let's talk in a year or 2 years from now and see how it's looking if we're hitting the right note.

I'll hold you to it. I don't worry. I'm curious speaking of that, what happens -- like I know that it's an established agency, but it's established by a political individual called the President. What happens if there's a different kind of President?

Yes. So starting with that unanimous support, the Congress has given us $2.5 billion appropriation, the President just asked for another $2.5 million for us. I think that will continue because the new solutions in health are needed. Whether I'm at the helm or not, in our authorities, I have a 4-year term to start renewable for another up to 4 years, but ultimately, the President is a decision maker on that. But I'm confident that the agency will remain. It's just how does that leader change over time.

What are you looking for in a program manager. I mean you're talking about -- I mean, is there something new focuses on cancer biology for one thing or a broader -- I'm not really sure.

Yes, it could be. So we're looking for folks who are an expert in their field, whatever that field is or that willingness to totally reinvent themselves and jump to another field, right, like [ Time Knight ] is a great example of someone that is just like, "All right, I'm doing this," because it really passionate about like what can happen next? These people are not afraid to take risks. They're decisive, you're only there for 3 years for the first term. So that means every 2 weeks that goes by 1% of your time has elapsed at ARPA-H. So there's an urgency, so people willing to move quick. And then I think people that are good listeners. So you're never going to really -- you're not going to be a patient in everything, so you need to be able to listen to those patients, those health care providers, industry about what needs to go forward in the programs. And then these people can be early career. They can be mid-stage. They're tired of the status quo. They want to come in and change the course of their fields and then go back and continue their work. or their late career stages that have just seen the problem so clearly, they're not afraid to take a career risk because this might be the last step. So any of those are all possible. But we're looking for diversity of like demographic geography, we're not making anybody move. We're looking for a variety of projects, as I've mentioned, but also experience. Have they been in government? Have they been an industry, have they been in academia or all 3? We really want to have a good mix group in that regard.

Are you the person who ultimately hires the program managers? Do you have a team that votes or...

We have a team, we're building a pipeline, and this is something we're doing a little bit different than DARPA. We will be hiring, -- today, we'll be hiring a decade from now because of this flywheel. And so we really want to give ourselves the best chance of recruiting. And so we're setting up a good pipeline. The mission office directors guide a lot of that, but we're going to have a team to help do some of that recruiting in an hour, especially we're just starting out minimum viable agency. That's the hardest part is bringing these folks together, getting their $1 million idea to $100 million idea takes a lot of time.

I want to ask how different you are from DARPA, but I think a lot of people, when they hear about you are going to say it's DARPA for health. How long is that approach?

We're adopting the business model, but it's not a copy pace. So I talked a little bit already about our transition office, which is definitely different. For this group, in particular, I do want to flag one of our authorities is I think it's going to be magical. And I can't wait to see what we're going to do with it, which is we can reimburse FDA directly. And so NIH does not have this authority, and that means that we can incentivize industry in nonfinancial ways to work with us. And so I've been working closely with the commissioner's office on like how do we define that? Is this what does this look like in terms of funding platform technologies, having a different approach to regulation there? We want to hear from you about what would be most empowering for your organization. So on Monday, we released a request for information, where we've listed a few different types of ideas that we have that might be things, tools that we could use, but we would really love to hear from industry. Like what -- if your technology is amazing and the regulatory pieces is a challenge, to be clear, we're not going to reform the FDA. That's just like off the table. It's not what ARPA-H is going to do, but what are the creative incentives that you'd be excited about.

Before they come out with a hook, is there anything specific left that you want to make sure everyone knows.

Just that we're here. I think that the health sector into bio maybe doesn't know what ARPA-H is about, and I hope we've helped demystify that a little bit today. We're really excited to work with the SynBio community. I mean one of the reasons I went back to government is because I'm a big believer. This really is a century of bioengineering, biotechnology. And let's harness these tools for -- to better all of our health, and we're really excited to do that in partnership with all of you here.

Well, we're very excited to have you doing it. Thank you very much for the conversation.

Okay. I have a funny story. Funny thing happened to me the other day. I think it can help put a lot of what's going on today into perspective. I look around this audience, and I think about what Jason said this morning about this is a very carefully curated community. This is a carefully curated gathering. Everybody here gets it. Everybody here gets it, right? So this is a place where you come and everybody gets that automation is transformative, right? We're not pipetting my hand if we can avoid it, we're no pipe heads. Everybody here gets that machine learning is transformative, if you can support it with good quality and large data sets. Everybody here gets that foundries are transformative. We need to make biology easier to engineer. We need to take on more of that technical risk. That's going to be what unlocks a lot of the power of this ecosystem to build new products with biology, right? Everybody here gets that. And so we come here, we come here to this community, and we start from that baseline, and we talk about the details. We talked about the details. We say, okay, well, what's the right data set to use for AI? What's the right way to set up your automated workflow? What's the right way to build a company on top of Ginkgo's foundry. And that's what we talk about. And it's easy to lose perspective about what a lot of people out there are still doing and how they're still thinking about these problems. Okay. So here's my story. Talking to my friend, does R&D for big, well-known well-financed pharmaceutical company does R&D, lavishly funded research group. And I -- they do biologics, right? It's protein drugs. So I'm a straight hustler. So I'm like, "Oh, I'm going to do some market research." I'm going to use my network here. How do you do R&D? What's your process? What does your process look like, right? These are lavishly funded, absolutely next-generation biologics research group. What do we do? Oh, well, my boss designs 2 or 3 constructs gives it to an intern to clone. Yes. Yes, that's right. That's right. I see your faces. Actually, I was stunned. I was floored. I was flummoxed. I was flabbergasted. This is true. This is a lot of this is still going on out there. You think there aren't DNA sequences being shared on word documents, there is a lot of that going on out there, okay? So it is important at this gathering to not lose perspective on that. And I say it not to -- so that we can just pat ourselves on the back, right? Although we should, that's okay. we can, we're smart. We deserve it. We're good. We should pat ourselves on the back. But I think more importantly, we have a responsibility to help this ecosystem grow by spreading the word about how to do it right, how to do it right. And I think one way that we can do that is by lifting up and celebrating examples of companies that are succeeding because they do it right. With that in mind, it's time for another round of application lightning talks. We've got Persephone Biosciences. Persephone are pioneers in synthetic biology, reimagining patient and infant health. We've got Stephanie Culler, Co-Founder and CEO. Microba Life Sciences are advancing health with new solutions developed from the human microbiome. We'll be hearing from Trent Munro, Senior Vice President of Therapeutics. Optimvia is a biopharmaceutical company specializing in engineering enzymes and their co-factors to synthesize complex therapeutic molecules. We've got Keith Kleeman, CEO. And Arcaea is a beauty revolution for biology, where nature is the lead technology. We're hearing from Jasmina Aganovic, CEO.

For 5 years, Persephone has focused on unlocking the microbiome and fields, including oncology, infectious disease and infant health. Our oncology clinical studies have covered the gut microbiomes impact in patients' response to the latest immunotherapies, enabling us to develop the next generation of engineered living medicines. The gut microbiome affects vaccine response and virus clearance speed. Our COVID Infectious Disease Research aims to develop living medicines that enhance vaccine response and reduce virus infection with a focus on fighting even preventing future pandemics. Our current microbiome research illuminates what's happening inside people today. We believe in exploring the connection between the microbiome of the past and the present by starting from the beginning. We created our infant clinical research program called My Baby Biome to map the gut microbiomes of infants nationwide to understand how the microbiome at first affects the rest of their lives. The infant microbiome crucially impacts immune cognitive and overall development and its status correlates highly with disease risk later in life. By understanding the infant microbiome, we can create and develop interventions that address the following questions: Can we prevent immune disease? Can we prevent food allergies? Can we prevent chronic illnesses like cancer? According to our research, over half of U.S. infants lack necessary gut microbes for proper development primarily due to antibiotic C-section births, formula use and poor maternal diets. In the place of those missing microbes, we found potential pathogens many of which we have observed in our studies of advanced stage cancer patients. Armed with this groundbreaking data sets, we're developing an infant probiotic for optimal gut health, planning to make it commercially available by next year to promote healthy lives for all generations. Understanding the microbiome from day 1 creates opportunities for new and improved therapeutics which we, at Persephone, are passionate about. Over the next 5 years, we plan to touch the lives of millions of babies. And from the infant data, we can begin to answer critical questions and raise new questions, that we aren't even aware of today. At Persephone, our dedicated scientific team will continue to grow and pursue the answers to these questions. not only for infant health, but for oncology and many other diseases. Our continued partnership will enable us to scale our living medicines and help us bring these much needed interventions to market. I am so hopeful for the future. If we can raise healthier babies today, it's a paradigm shift in health for future generations. Thank you.

Good afternoon. It's a great pleasure to be here and really excited to share with you the journey we've been on at Microba. My name is Trent Munro. I lead therapeutics development at Microba. We're an Australian-based company publicly listed founded in 2017, we've raised over $50 million, and we're focused on developing both testing solutions for the microbiome but more importantly, the development of next-generation microbiome-based therapeutics. So I've spent some time really thinking about drug discovery over the last little while. And what I'm really fascinated about is the tools that we have to think about the modalities that we've deployed. But what's really surprising when we look at the data and if we take the 350 or so products that have been approved by the FDA, what we see is the vast majority of this is built on reductionist biology. The idea that we can come up with a natural compound or design by chemistry or biology something that's going to impact meaningful biology. But in reality, I think that's really not the way we need to be thinking about going about the future of creating next-generation therapeutics. At Microba, we're taking a different approach and really, our mission is to use our deep understanding of the microbiome built on data and harness that to deliver next-generation therapeutics. When we think about -- and you've just heard the importance of the microbiome that really, today, we look out 6 and 10 of people in this audience are suffering from chronic disease. But the vast majority is not well controlled by current therapeutics. We see evidence that it's clear the link between the microbiome and disease is becoming more and more important. And I think more importantly than that, when we think about cell biology, we typically only think about ourselves, not the way we can think about the complete ecosystem and what we need to actually do to develop those new solutions. The microbiome produces and metabolizes a range of potent compounds, which control not only functions within your gut, but at distal sites and control things like mood in the brain. So really in terms of a drug discovery pipeline, this is essentially untapped. What might be surprising to many of you is that the vast majority of the microbiome remains uncharacterized. Yet there are more than 4,800 known species of the microbiome. In any one of us, we have 200 at any one time. They're controlling functions, thinking about this is almost like another organ. But the other thing that's really happened in the microbiome space is that people have really only scratched the surface from the way they characterize the microbiome. We were founded at microbiome thinking about the problems in characterizing the microbiome, the deficiency and analysis, the high error rates and the issues. We've developed a novel technology to overcome that with more accuracy and more coverage, and we've built a data set that now lets us interrogate this in different ways, using advanced algorithms to go after the root of the microbiome, the key organisms that are driving the key biology. We've built a pipeline across inflammatory bowel disease, cancer and a program we're working on with Ginkgo in the autoimmune space to create a differentiated set of therapeutics. We've been very fortunate to work with the talented team here at Ginkgo and the amazing infrastructure to take a new approach to doing drug discovery for autoimmune disease. We're excited about these results, and we're looking forward to sharing more very soon. So with that, I'm going to finish. Thank you for listening, and enjoy the rest of the day, and I look forward to connecting with folks throughout the rest of the afternoon.

All right. Thank you very much. I'm excited to be here. I don't have a clicker. That's because I don't have any slides. And we were going through sound check yesterday and somebody commented on it and said, that's really bold. I don't know if it's bold or incredibly stupid, but we're going to find out here in a minute. I'm Keith Kleeman, I'm the CEO of Optimvia. Optim, optimal via pathway, and that's what we do. We create and develop the optimal pathway to manufacture globally critical medications. As an example, Heparin. It's an anticoagulant blood thinner. It's used in everything from open heart surgery, kidney dialysis and everything in between. It's actually estimated that Heparin saves up to 100 million lives each year. It's an astonishing number, right? Even more fascinating is it's also estimated that 50% of the human population, half of us will require Heparin at some point in our lives. That's fascinating. But there's a big problem. The problem is that this globally critical medication is currently made by scraping the intestines of slaughtered pigs, and that's disgusting. But disgusting is not the problem. The problem is availability of medication and supply chain. The problem is you have a globally critical medication problem #1 coming from a livestock prone epidemic. Problem #2, coming from a singular livestock pig. Problem #3, 70% of those pigs are raised in a single country. And so when an epidemic or an outbreak hit, as it did in 2019 with African swine fever, 1/4 of the hog population gone. 100 million hogs gone, which results in an instantaneous shortage of this most critical medication. So clearly, this is not the optimal pathway to manufacture it. So at Optimvia, we looked at it and using synthetic biology, we created a manufacturing process to produce synthetic heparin that's equivalent to porcine-derived Heparin. And here's a bold statement. I actually believe in the next several years that the vast majority of all Heparin manufactured around the world is going to be manufactured using this biosynthetic technique and no longer using pig gut. Maybe it's founder to speak, maybe not. But if we look to a very well-known historical precedent, the path to synthetic Heparin becomes abundantly clear. I'm talking about insulin. Life-saving medication like Heparin, about 100 years old, like Heparin. Previously produced using cow and pig byproduct just like Heparin. For 70 years, these 2 stories were intimately intertwined until about the mid-1990s and then they diverge because synthetic insulin is produced using synthetic biology recombinant DNA. From that point forward, 100% of all insulin sold around the world has been synthetic. Heparin is on the very same path and same trajectory. In the very near future, in the next several years, the $10 billion a year annual market, of course, in derived Heparin is going to be replaced entirely by synthetic Heparin. And in doing so, we're going to diversify that supply chain. We're going to ensure access to this most critical medication for half of the patients around the world who need it regardless of the geopolitical situation regardless of the global hog population and all because of what's been done with synthetic biology. Thank you so much.

Hi, everyone. I am really excited to be back at Ferment this year. The last time I was here, we actually announced the formation of our company, Arcaea. So it's really awesome to be back here talking about a product that we just launched just 18 months after I was last year. That product is called ScentARC. So for those of you who don't know us, we are a biology-first beauty company. And what that means is that we are working with a whole slate of biology centric technologies and we are pointing them at large product categories like deodorants. And you might think this is not really a sexy category. But guess what? The $25 billion category that has been relying on the same underlying science since its inception over a century ago. Basically, masking smells using fragrances, blocking or absorbing sweat and then killing things using antimicrobials. But more importantly, I don't really know that many people that love their deodorant. It is a low NPS score category. And so the reality is, today, we know a lot more about deodorant than we did back then, thanks to biology. In fact, we know that body odor is caused by certain microbes in the armpit and not others. And so we wondered because microbes are living things, just like us humans. They're highly influenced by their environment and what they eat. And so we thought could we design food for them that made them act in a specific way. And in this case, it was to make them not produce these smelling compounds. And that is what we did with ScentARC, and I'm going to tell you how. So first, we collected lots of microbes from lots of arm pits. We then sequenced all of those microbes. And specifically, we were looking to understand which of those microbes have the capacity to produce those smelling compounds we don't like and which ones do not. Then we work together with Ginkgo Bioworks, basically accessing their high-throughput automation. What we were doing was screening tens of thousands of different nutrients against all of these microbes to understand how they would behave differently and how they might produce odorous compounds depending on the nutrients that they're fed. Then we set this into a machine learning model, and we started training that model. We were basically asking this model to predict which combinations of these nutrients would result in that behavior that we were looking for, basically the lowered production of smelly compounds. We iterated a lot to improve the accuracy of this model. And then we finally tested it on humans. 87% of participants in this third-party study said that ScentARC reduced the production of odor throughout the day. So prevented the formation of odor throughout the day, and that is without ingredients that you're used to seeing. That's without baking soda, without essential oils, without things like magnesium hydroxide or fragrances. And so today, just a few weeks after launch, ScentARC is already being sampled by deodorant formulators, hundreds of them. And so that means that deodorants with ScentArc are coming to shelves near you probably within the next 12 months, which is really exciting for our team. And in reality, we look at biology and ScentARC really just being the beginning, we see biology being an entirely new ingredient talent for us, and we call it expressive biology. Upstairs on the mezzanine, we have a couple of samples of deodorants formulated with ScentARC as well as some other product categories that we're pursuing, like Sun Care. So I hope you can go upstairs and check it out, and say hi to our team. Thank you so much.

Right. All right. Let's give it up one more time for Persephone Biosciences, Microba Biosciences, Optimvia and Arcaea. Can we? Yes, Yes. And that is afternoon break time. We will return at 3:15.

Thanks everybody for coming today. Before we dive into what's going to be an awesome panel, I just wanted to take a few minutes and share a couple of things about biosecurity at Ginkgo. So I want to make sure we do 2 things. One, I'm going to give you a little bit of footing like why we are investing so heavily in biosecurity. And then second, I want to actually go into a little bit more detail than we have in lots of these environments about what we're doing on the ground, like what are we inventing, what are we building. So if we get those 2 things done in 9 minutes and 30 seconds, we will have dramatic success. So Ginkgo believes, as you've heard throughout the day, as you've learned about Ginkgo, we believe deeply in this like amazing future made by biology. The simple reality is for that to be true, you've got to do 2 things. One, people have to love and trust products. That's like in any industry. And then the second thing that needs to be true is that may be very fresh from the last 3 years, people have to be willing to leave their homes to engage with a beautiful future invented by biology, right? And so when you think about biosecurity as a footing that is what we're trying to make sure exists, that those 2 things are true. I love the analogy here of how companies like Google have spent so much money over the years and their peers on cybersecurity. Why is that true? It's true because for Google's products to be amazing and to be loved, people have to trust the Internet. The Internet has to work. It has to be an amazing platform. And so cybersecurity is critical for the future of -- for the digital ecosystem to exist. Ginkgo -- we have not lost the lesson of the Internet here. And it's very clear to us that for this like amazing future of biology to exist, we have to invest in biosecurity. And then the last piece of footing is to continually remember, and it would be forgiven to forget this again after the last 3 years, that biosecurity is not actually just synonymous with health security. Biology is everything around us. It is everywhere at all times, right? So it is plants being healthy. It is the climate being healthy. Biosecurity permeates lots of different parts of the world we live in. And I think that for us, like when we think about the suite of technologies over time, like over arch of time as you build biosecurity, you've got to really imagine every place that biology interacts with the economy. Where are those inflection points where biology interacts with the economy and those are the ones that we need to be focused on securing. So with all that kind of big picture stuff, I will say that the COVID-19 pandemic did give us a chance to take philosophy and move it into very specific practical tools. And that started for us with the opportunity to operationalize kind of -- at massive, massive scale, a response to SARS-CoV-2. Over the last 3 years, kind of in detail what the core -- the core that what Ginkgo built is essentially what you can think of as like a radar system for places like schools and nursing homes. To be able to go to a school and give somebody who would never usually have these tools of data generation around biology, a principal, for example, give them the ability to monitor a classroom for COVID outbreaks. That is something that we were able to build and deploy at scale across the country. We ultimately worked in 37 states. Many of these programs are still going. We collected almost 12 million samples to date and supported 5,500 organizations. This also for kind of framing in our biosecurity business gave us a new way to think about essentially a product road map, like what is the architecture for what things do we need to be investing in and innovating in. And I really love this way of thinking about the definition of biosecurity in the context of how we think about it at Ginkgo. It's the application of modern tools of biotechnology at scale. I think nationwide to counter harmful biology in all its forms and origins, right? And as we move through the pandemic, the public health emergency is over. Thankfully, many of our school programs are moving on over the last year, schools returning to normal. Now modular, you would love to make sure you had outbreak detection if a new variant popped up, that was terrible for children, but thankfully, things are moving on. We started to ask ourselves, where does this mindset, where are these capabilities best applied. And with a lot of thinking about a lot of places, it became pretty obvious when you think about how global travel works that we should probably be doing this in places like airports, and we are super lucky to find a partnership with CDC and XpresCheck like super innovative parts of the organizations that say, let's try to figure it out how we do we take this like radar station mindset, this outbreak monitoring mindset and monitor airports. And so what we're doing today across 7 airports in partnership with CDC is running a program where when airplanes land in the U.S., we're both taking anonymous swabs from volunteers just like we did with the K-12 students in their classrooms and now wastewater off of airplanes to quickly detect variants of SARS-CoV-2. We're also doing that with flu. So now it becomes multipathogen. This is a platform to monitor for lots of different types of infectious disease. What's cool about this program as it starts to move us to the slight next layer of what people talk about as like the global immune system, right? So when you get a new variant of SARS-CoV-2, you're able to start in partnership with CDC and others thinking about, well, is this going to be immune evasive, is this is going to be worse or is this kind of a normal threat of a variant. With the flu program, the aspiration there over time is to be able to detect and immediately put something like a border monitoring program into a vaccine manufacturing program. That would be how do you tie your detection into response. This is 7 airports today. This is something that we are growing, and it feels like it's an amazing program, but what is most important about it to me personally, and as we talk to lots of people around the country is that it starts to feel like a new type of infrastructure. It's a platform for biosecurity, just like you have platforms for other type of security. It starts to feel like a network that can be used for making sure that we are better coming out of this pandemic than we were going in from a capability standpoint. The United States is not the only place that has issues with biological risk, right? Biology does not respect borders. We've taken this idea of radar stations now, and for the last 18 months have been -- our teams have been going around the world, doing partnerships with countries just like those countries deployed cybersecurity technologies as digital technologies became pervasive. Many of them are looking and saying, "Yes, I get that I need to think about innovation in my biosecurity platform." We now have active programs, pilots or MOUs in 10 countries that we've announced, and we continue to push really hard to expand those. Here's a really cool picture of some of the work that's being done on the ground. I would just point out these capabilities exist, we're partnering with countries to deploy systems and bring things that we're innovating on here in the U.S. So if you look at the little device that connects the airplane -- to the airplane, that's the device that our team invented to make it really fast and easy to collect wastewater out of the wastewater tanks in a marginally less disgusting way than you can imagine otherwise. I think then the third phase of this, if you go from schools and outbreak monitoring being the use case for radar stations to airports, to get early warning for pathogens and then move into characterization and vaccine development. The third one is once you have radar mindset, where else would you want to be monitoring that have -- that are areas of high biological risk. Like with cybersecurity, we monitor nodes all the time. We monitor every phone, every network, constant persistent pervasive monitoring to make sure that you can quickly respond to threats. So we see a world, and this is kind of -- this is very much where we're going with all of our country partners where there's a number of areas that you'd want to drop a radar station to get very, very consistent monitoring of pathogens. One of the next major areas are areas that are affected by conflict or natural disasters. So this -- I was just in Kyiv 2 weeks ago. We're launching a partnership with the Ukrainian Public Health Center to help do wastewater monitoring in areas that have been affected by the conflict with Russia. It's a rebuild of public health infrastructure, but a rebuild with a mindset with the capabilities that have come out over the last 3 years that we've been able to build across the world. So I think I would just leave this group with as we go into this panel. I think there's 2 big things that are on our minds. One, you can think about Ginkgo Biosecurity, our laser focuses, focusing on building the technologies that you need to enable the global immune system. This is not one point technology. It's systems of systems. It's new capabilities that need to be innovated over and over again. And then the second piece is what Jason mentioned right at the beginning. This is not just a public health challenge. This is also a national security challenge and the unification of national security thinking, public health thinking with a mindset around how you build technologies into the future for -- to prevent any future events like we just had for the last 3 years is really kind of where we're thinking. So I'm super excited about this panel for that exact last reason, which is this is a group of people that have been very involved across the national security to public health conversations for many years around biotechnology generally and biosecurity biodefense specifically. So Megan Frisk, who is the Senior Advisor and Biotechnology Policy Coordinator in the US Department of State, really thinking about what is the leadership the U.S. government has in setting regulatory regimes and leading on what biosecurity should look like. Michelle Rozo, who has recently joined In-Q-Tel, the Intelligence Community's venture arm, but over the last many years has really been one of the key people defining our biotechnology generally and biosecurity policies in the U.S. and Dr. Richard Hatchett, who in this audience doesn't need much introduction in our community, doesn't need much introduction. He is -- he leads CEPI, which has done just some amazing work in vaccines across the world. I should mention the other 2 are also doctors, I apologize for it. We have many doctors and surrounded in this world by doctors and it's amazing. And we have a general. So General Tom Bostick, who is the former Chief of the U.S. Army Corps of Engineers, but also a reform biotechnology executive, and who has just been an amazing mentor to so many of us at Ginkgo and is a really amazing human being, who will moderate the panel. So thank you so much. I appreciate you all being here. We're excited about what we're building in biosecurity at Ginkgo.

Well, good afternoon, everyone, and I want to thank Matt, for the kind introduction. It's great to be amongst some wonderful leaders in both national security and biosecurity and have a great conversation on the panel today. What I thought I'd do is take off of what Matt was talking about. He talked a bit about cybersecurity. We know AI is in the news. During 9/11, I had nuclear codes around my neck and I was in the National Military Command Center, where we rehearsed all the time nuclear attention -- nuclear attack. And we've got these screens that we monitor. And we know if North Korea is lighting up a missile, China, we can see that all the time.

In cyber, we're now at a pretty good place where we can see cyber attacks happening to us, and we're able to react. We're not there in biosecurity. But it's an interesting comparison when you look at those 3 and I wonder if our panel, starting with you, Richard, could give your thoughts on lessons learned, comparisons of those. What do you think?

Sure, thanks. The -- I do think the other domains where we have to develop a comprehensive security solutions are really important to look at and then we can draw lessons from them. I think -- I mean, it's incredibly encouraging to hear coming from the private sector, Matt, for example, describing his concept of systems and systems in global pervasive surveillance. Areas like nuclear security, cybersecurity, I would even say other areas you might not think of like earthquake security or fire security. These are problems that for the most part, we have figured out how to distribute costs across society. These are not areas where we are exclusively relying on public sector funding to address the security problem. But the problems, the security problems that we're trying to address are each structured in unique ways and the sense of solutions that we can find are based on how they're structured. I mean, cybersecurity, right, we're all carrying around the phone. I mean we're all at immediate risk and we're all willing to undertake some of the cost of paying for cyber protections. Nuclear security, the example that you mentioned is very remote for most people that's about controlling material and change of custody, but there are analogies from security problem to security problem. And my concern looking at bio is that -- because it does stretch across health and national security and other domains in this novel, most people haven't spent a lot of time; thinking about it. We haven't really figured out how to conceptualize the problem in a way that helps us come up with sustainable ways of regulating or distributing costs across the economy, which is what we're going to need to do if we're going to address these problems in the way they need to be addressed.

Thank you for that comprehensive answer. And I wonder if our other 2 panelists would like to offer anything in this area.

Thank you. And thanks, Richard. Those are great points. Look, I love the question. I think it's important to look at technologies that have come before and as we apply them to biotechnology. When I was at the National Security Council it was helpful to make references back to other technologies, the policy community may be more familiar with, right, like cyber, like AI, like semiconductors. And so I think it's important to look for lessons learned, but also lessons tried, right? What was tried and then didn't work, so that we can start from a different perspective. Richard talked a lot about some great analogies. Maybe I'll bring in the importance of kind of collective norms generation and building. This is really formative in the early days of the cybersecurity community. It was community-driven codes of conduct, right? And we're seeing this in real time in the AI community. Discussions along the advances and generative AI. The community is pondering in real time, where should appropriate restrictions be? What should the role be of industry, of governments, right? And how do governments think about regulating a technology alongside very rapid innovation in that area. So I think a lesson that we can take to biotech is since we haven't yet potentially reached that leap ahead moment or we're still in the formative stage of a lot of these technologies is can we start these conversations now. Can we start to think about what that intersection is between the public and the private sector and some of those points that Richard touched on and do so in international context, right? Biology doesn't respect borders, right? It is self-sustaining. So we need to have these discussions now. We need to have them with private sector, with public sector, with international colleagues and build a multi-lateral system, not just of norms and governance, but really moving into biosystems like we heard Matt talk about earlier. And in doing so, we can build in reinforcement. Mutual reinforcement takes the barrier or the burden off of one individual country, one individual industry and we need to do this with a large, diverse group of people that share our values.

Those are excellent points. Thank you. And thanks for raising the public and the private coming together to work on this challenge together. In many cases, you're seeing the private sector lead in biosecurity and the more that we can do in a public private fashion would be great. But we'll come back to that. Megan?

Thanks. Actually, I would like to build off of Michelle's point too on AI and putting together the international community. I've seen throughout the course of the day how prominent AI has been in these conversations. And so it's interesting right now to take a look at what's happening in AI and draw the parallels where we need to for biosecurity. And what we're witnessing with ChatGPT is this whole scale scramble that's been generated by a signature public breakthrough, and it's disrupting or promising to disrupt business models. It's moving governments all over the world and citizens alike. It's capturing the imagination of everyone. And I would pause it that we haven't had the ChatGPT moment for biotechnology yet. We here in the room, we see it. We see the potential. We know it's coming. And so why does this matter for biosecurity. We are at this inflection point where we don't have the height machine forcing us to take action. So we have actually have this time right now where we can think, we can resource, we can plan, we can act methodically to build our bio future. So at the State Department, we're thinking about this, as Michelle pointed out, the need to move internationally, so I can just build off of that. We are asking with our foreign partners and that includes industries, we're asking with our foreign partners, governments, academia, industry leaders, what does the global architecture look like for managing biotech risk, but also promoting the best potential of biotechnology. What value systems are we aligning with. This values conversation is very important and one that we really need to come together around to understand what our values are. What guardrails are we putting in place to manage bio risk and strengthen biosafety and biosecurity all over the world. And we really need to work together in that anticipation of that moment that hasn't yet arrived and ensure we unlock the power of biology while at the same time, inspiring creativity, equity and managing risk. But you asked about biosecurity specifically. So I just wanted to also take a moment to talk about from the State Department's perspective, what we need to do around biosecurity. COVID-19 obviously really heightened our awareness of biological risk. I think that comes as no surprise. The bioeconomy executive order, our National Biodefense Strategy, these really drive home the need to incentivize and strengthen biosecurity, both at home and around the world. There's more and more researchers experiment with biology. So from our perspective in terms of biosecurity, we will absolutely draw parallels where necessary with AI, with cyber security, with nuclear security. We're seeing a lot of good examples out there. But we are also looking at this bigger framework. We need to strengthen the political will all over the world. We need the governments to come in and say this is a priority. Two, we need to build capacity. So the capacity building point is very important that we all are a part of. And three, we need to strengthen the connective tissue amongst all of these players, government, academia, nongovernment, all the practitioner community industry, we need to strengthen that connected tissue and share these best practices. So I think we're going to be learning a lot from AI, but we're also going to be learning a lot with AI, as we look at this interface between biology and artificial intelligence.

That was excellent. And I want to pick up on one of the points you made about we haven't had our ChatGPT moment. And it's really amazing how that has really accelerated where we are at, where the country is on AI today. But we've had our COVID moment. And you would think that, that would be enough to drive the world towards a more secure biosecurity framework. And I wonder why it has not. I spent many years in Washington. I was the Chief of Engineers, and we dealt with crises all the time. In lot of these natural disasters, we knew the impact of what would happen. We didn't invest ahead of time and then we invested a lot more, much later after the crisis occurred. And then we went on to the next priority. So the question I'd like to pose is we're post the COVID pandemic, it seems logical that we would create a biosecurity framework. And the question is who should lead that? How do we keep the priority in the country a priority in fact, in the world and I thought maybe Michelle, you could help lead this.

Sure. I'm happy to kick us off with the caveat , right, as you mentioned, it's a challenging question, right? They don't call it the panic and neglect cycle for nothing. We've been here before. I hope we're not here again, but we had similar conversations after H1N1, after Ebola. So your point of how do we sustain this momentum, it's critical right now. And unfortunately, we might already be in that neglect stage of the cycle, right? So I think in terms of who should lead on the U.S. side, I don't know that there is necessarily one right leader. But what I think we need our champions, right? We need Champions spanning Congress, the executive branch, private sector, international partners, right? And we need to make collectively the argument to do 2 things. The first is to sustain the advancements that we've made over the last 3 years, right. We invested a lot of money at the onset of the pandemic, it would be not only a shame, but it would damage our economic security or national security to let those advances fall away right, and have to start over again and reinvest those money the next time around. So anything from vaccine manufacturing, capacity, clinical trial infrastructure, diagnostics and telemedicine at the edge, bio surveillance, right? We made a real leap ahead advances in this capacity. Of course, more needs to be done. But I think collectively, we can make the argument for what is the type of investment that could be made to sustain what we already have in progress, right, whether it's keeping capacity warm, maybe it's not functioning at the maximum pandemic speed or maybe better yet, can we integrate this into our health system, right? And can we look for investments that strengthen our ability to just deliver better health care and doing to make ourselves a more resilient population for the next pandemic. And any of those examples of buckets of technology I mentioned, by continuing to invest and strengthen those tech areas and those facets of pandemic response, it will make us better be able to deliver health care to United States and to the globe.

Great. Megan, do you think those investments will come through so that we can sustain the momentum that we've got?

Yes. I mean -- sorry. The investments other than monetary investments, we've made so much investment over the course of this administration and our policies that really set that strong framework to keep going the sustaining portion. We've had the bioeconomy Executive Order, the National Biodefense Strategy and its implementation plan. The 100 days mission that maybe Richard will comment on. And hopefully, through these policies and frameworks, we are keeping warm, we are doing exactly that is preventing that neglect. You've asked about leadership and who would lead in this effort. And I know a lot of you might be thinking domestically, who would lead, is that the White Houses, is the agencies and everybody has a role to play. There's probably not one leader amongst all these. But what we think about it at the State Department is U.S. leadership as a whole globally. And how do we project that? So what is the role of diplomacy and what is the work of the U.S. government alongside a private sector, so many of you represented here today are academia, our nongovernment institutions and organizations, and our leadership along with our allies and partners internationally. It's absolutely critical right now that the U.S., our allies, our companies, our innovators remain world leaders in this vital technology to our economic and national security. We have to -- we believe in the role of the private sector in setting the standards, consensus-based standards to norms, being able to harness the power of biotechnology for good and solve global challenges together. And as my secretary -- Secretary Blinken said, our task is to put forth and carry out a compelling vision for how to use tech in a way that serves our people, protects our interests and upholds our values. So this is absolutely vital in terms of leadership globally. It is not necessarily one entity. It is not just the U.S. government and it's incumbent upon all of us to project those things internationally. So it's another part of not neglecting the future and not falling into complacency as we emerge from COVID. We have also put together an international engagement plan as part of the executive order to drive forward the positives of this technology that promotes of this paradigm, while also mitigating the risks that we know are emerging or exist and doing that alongside our allies and partners.

I think it's really important point you're raising, is that, it's going to take global leadership and everyone's got to be involved. And I think if we go back to the parallel with cyber and you think that when there was a cyber issue in my organization, it was the IT job, IT person's job. Until we really got serious about it. We put funding, we put structure, we created a cyber command. And now it's everybody's job and everybody is ingrained in it. And I wonder, Richard, you've been involved in the government, in the structure concerned about these biosecurity issues. And as you wrestled with the structure as it was and envision what the structure might be in the future, do you see a structural component to how we should move forward?

I do. One group that I think should not be in charge of our biosecurity is the Department of Health and Human Services. This is not just a health problem. Matt made this point in his opening remarks. This is a problem that is pervasive. It needs to be across the public and private sectors. But it certainly needs -- within government, it certainly needs to be hold of government. And I think just as a thought experiment, imagine that infectious diseases didn't have names and we just had this [ passive threat ] infectious disease. And they would come and go in every winter, we would have epidemics and periodically, we have gigantic sort of world chattering pandemics, but they didn't have names and they didn't have particularity and the countermeasures were not very specific for each specific threat. Everybody would recognize the importance of infectious disease as a category of threat, and we would figure out how to build a safer buildings and have better ventilation systems and have building codes and distribute that risk in the way that I was talking about. I think we are globally moving towards kind of understanding that this is not just a health problem. And I do think I just want to reiterate the point about the importance of engaging internationally. This is not a problem that the United States, whether the public sector in the U.S. or the private sector in the U.S. can solve for itself and by itself and be safe. It has to be a global solution because of the nature of the threat. I would charge the people in this room as being at the leading edge of the bioeconomy. Your responsibilities in helping the government and governments develop solutions are something that you should take very seriously. Biosecurity should not be just a topic at the end of the day. It should be a frame through which we think about everything that we're doing.

Excellent. Excellent. One of the things that I wanted to follow up on is, this international component, and you travel the world, Megan. You're seeing these other countries. And I wonder if you're seeing anyone that's got a comprehensive biosecurity strategy or they're moving forward with a program that's really starting to stick.

Yes. We are seeing a lot of good progress around the world and thinking about this issue, thinking about biosecurity more holistically. And I think our executive order had a lot to do with that and showing that biosecurity is not an after thought. It's part and parcel of the bioeconomy and growing a prosperous bioeconomy. You cannot promote the positive of the technology and not to think about the security and not just from a health lens, but from across sectoral lens. And so we've seen a lot of good progress. I'll probably disappoint you by not naming -- calling out any countries here. But I would say we've seen a massive uptick in the number of national strategies and plans and consortia really trying to wrap their minds around how to grow bioeconomy and it's in synthetic biology. It's in workforce development, growing global talent. It's in biomanufacturing infrastructure and capabilities. It's not just developed countries, it's also emerging economies. And this is a really important voice to include in our discussions going forward. They have seen how the United States upon decades of investment in biotechnology R&D. They have seen how quickly we could respond to the COVID pandemic, and they realize the value proposition that biotechnology is for their economies and their national securities. And they want to get in on the ground floor of harnessing this technology. Being part of the discussion around values that I mentioned earlier, being part of the discussion around shaping a meaningful biosecurity framework and what it means to have equity and innovation. So we've seen it across the board. It's a really exciting time. But at the same time, we should be mindful of the fact that biosecurity policy regulation practice is patch work around the world. And that is something that we really do need to address if we want to have a sustainable, safe and secure bioeconomy. So going back to the 3 things I mentioned in terms of how we are going to address this in the political will, the capacity building and the connective tissue. I just can't come back to those enough, it's really, really important parts of the channeling that energy that there is all over the world for growing a bioeconomy, but also making sure that we are doing it in that safe and secure way.

Okay. We're about ready to wrap up here. And I wanted to do some closing comments, but due to time I'm going to ask just that, one of us do the closing comments. And maybe one of the panelists would like to close and talk about maybe the public private partnership idea, do you think that, Michelle?

Yes, let me just close I mean like the last 5 seconds. One of the things that I think what Megan and I were chatting earlier and targeting back to the executive order, the importance of what we're talking about here in biosecurity and making sure that, that disseminates down into the next generation of talent. We know that we need to train more and more people to be a part of this future bioeconomy. There's tons of potential here, and we need more -- an influx of more talent in order to be able to serve all of the cool products and capabilities that folks in this room are working on. And so part of Richard's point of not having every asset at the end of the day is as we think about how we train the future workforce, making sure that these discussions of biosecurity are part of those conversations from the onset.

Excellent. Thank you for wrapping it up for us. Thank you.

Okay. So I'm super excited to introduce our fireside chat that's coming up next with Jennifer Wipf, who is our Head of Commercial here at Ginkgo, interviewing John Maraganore, who is the founding CEO of Alnylam. And for those of you that aren't from the pharma industry, it's an amazing thing to develop a new drug, a new candidate that gets to clinical trials and treats the disease. It is especially a magical thing to develop a new type of drug, a new modality, it's called. And you would see this, we have small molecules, then we had protein therapeutics, the new modality. Recently, you've heard about mRNA vaccines. And John, as the founding CEO of Alnylam, developed siRNA all the way from a concept to patients and a $25 billion company along the way. And so he's one of the few people in the world who has seen that path. At Ginkgo, we want to make it easier to see new modalities come to fruition in biopharma. So I'm super excited, along with all of you to learn from John. So Jenn and John, welcome to the stage.

John, thanks for being here with me today to chat about your success in Alnylam.

Thank you, Jenn. I do want to start by saying if Ginkgo can make it easier I would love that, because it wasn't easy with what we did. So it's an opportunity for sure.

I would say a lot of people in the audience today probably have aspirations to do exactly what you did, which is to take a new therapeutic modality and bring it to market in a wildly successful way.

So I'm curious to learn kind of, what were some of the big decisions that you made along the way to address the kind of compounding risk, right? In drug development, you're trying to manage between biology risk, a new modality risk, regulatory risk, clinical risk and so on. How did you manage that risk through the process?

So the story of managing that risk at Alnylam is actually quite interesting. So at one level, we had to figure out where we can achieve delivery of small interfering RNA, and then really focus on that cell type and that tissue for developing our human therapeutics. But then around 2010, we crafted a strategy that we called Alnylam 5x15 that really tried to optimize our overall likelihood of technical success with this new modality. And so what we did is we focused on targets expressed in the liver, where there was human genetic data supporting the target, either in loss of function or gain of function, human mutations. Then, we also required any target that we would pursue to have a biomarker that we can measure early in clinical development, so we can remove the traditional pharmacology risk inherent in doing drug development. And then, of course, we wanted to have targets and indications where there were definable endpoints from an approval standpoint and value definition perspective. So that amalgamation of criteria turned out to be a really smart thing to do because it reduced our biological and clinical risk as asymptotically close to zero as we could possibly do. And of course, we kept technology risk with our new modality that we had to optimize. But at least we removed the biology risk as much as possible. So that approach, that decision, and that strategy really turned out to be a very wise move, and I think a lesson that could be used with other modalities and platforms that are out there as well.

And by the time you got to the 5x15 strategy, you said you've been working on the modality for some time. So did you have some sort of assessment or thought process you went through to decide that the modality was ready?

Yes, we had to really spend about a decade trying to optimize the modality, trying to figure out how to chemically modify these RNAs, so that they weren't stimulating immune responses. We had to figure out informatic-based approaches to basically achieve selectivity of the molecules we were creating. But the biggest thing we had to do is figure out how to achieve delivery of these small interfering RNAs, which are really not that small from a drug standpoint. They're 14,000 daltons. And so they're not engineered to get inside the cell. And so we really had to figure out different approaches. And we ultimately ended up with 2 different strategies for delivery. One was to use lipid nanoparticles, like the ones that we've all used in our mRNA vaccines. And we sort of helped pioneer the use of those lipid nanoparticles for delivering RNA. And then the second technology that we use that, frankly, ended up being more robust for us, was using a chemical conjugate to the siRNA that enabled delivery to a cell type through a receptor ligand interaction. And a very rational approach in enabled subcutaneous delivery of these small interfering RNAs, but it also enabled a remarkable durability of pharmacology for these agents as well. And that was something which we didn't anticipate until we started generating animal and then human data.

And when you were tackling sort of the delivery challenges, you gave 2 specific examples that you ended up using, but I understand you are starting to look at many, many delivery methods, how did you sort of think about looking at the scale of those or the expansion of them? And how did you narrow that down over time?

Well, I think one of the ways you have to think about a platform technology of a new modality is a bit like an hourglass. So you start wide, then you figure out where you can build your pipeline so you go very narrow and focused. And then at the end, if you're successful with that, you can then go broad again. And so that's exactly what we did in our journey at Alnylam. We had at the beginning, a very wide-ranging set of technologies that we would evaluate to figure out how to achieve delivery, to figure out where we could ultimately build our pipeline. Once we figure that out, once we identify the liver and hepatocytes as being a tractable cell type and tissue, we focused. And we focused very -- very, very clearly on many pipeline opportunities that we can engineer in the liver. But then more recently, we were able to then expand our efforts again and go after things like CNS delivery of RNAi therapeutics or other extrahepatic tissue like muscle, for example. And so that hourglass-like approach that we took is something which I think is applicable for any modality that's out there, certainly, what's being done to some extent with gene editing technologies right now using ex vivo-based approaches followed by in vivo based approaches. It's a wise approach because you ultimately have to find the low-hanging fruit and really try to make medicines out of those low-hanging fruit opportunities as sort of your first mission.

Yes. So speaking of the low-hanging fruit, you had an interesting sort of story around this and I like the analogy of ringing the bell. So the sort of jingle bells you hear before Santa Claus comes as sort of helping people in the market, stay the course with you. And tell us a little bit about that story and how you helped show the low-hanging fruit to keep people on the bigger course?

Well, the story of RNA interference is very similar to other technology stories that we all read about in business journals and business textbooks. The period of excitement followed by a period of despair and lack of enthusiasm, followed by ultimately, if you're successful, value creation on the other side. And in the middle, when we were in our despair moment as a technology, it was extremely clear to me that the only way we could really convince investors to stay the course with us was to generate human clinical data. And I use the analogy, if those of you with children may know, the children's book, The Polar Express with a little jingle bell and the kid that can hear the jingle bell and he's able to believe in Santa Claus because he hears the bell, but his mother and father and older sister can no longer hear the bell. And I use that analogy with -- my kids were young at the time. And so I used that analogy with my team to basically emphasize that no matter what data we were showing early on, people weren't hearing the bells, people weren't believing what we were seeing in animals. They needed to really see it in humans to really believe that we can do it. And so it was about bell ringing during those days. I had a bell-scale slide where, on the left-hand side, it was a jingle bell; on the right-hand side, it was the Liberty Bell and a big one. And so we would track where we were in our progress in terms of generating human data. But ultimately, we got to the big bell.

That's amazing. Coming back to this hourglass picture you painted for us of going big and then narrowing, and then going out again for the platform, it's very similar to a chart that Jason showed in his keynote speech. Thinking about sort of the conundrum, let's say, of improving the platform while advancing the pipeline, right? At the beginning, you need a lot of R&D, then you kind of go into clinical and focus, and then you want to ramp that up. Did you experience that at Alnylam? And how would you advise people to approach that same sort of sinusoidal curve?

Yes. It's a fascinating challenge in building a company like an Alnylam where you -- again, you start needing a very broad research effort around your platform. And then when you finally figure out where you can build your development programs, you need to focus. But also as part of that focusing, invariably, you realize you've got to preserve your balance sheet to invest in the clinical development side of it. And so in 2010 and then a second time in early 2012, we had to do 2 restructurings of our workforce to basically change the complexion of our workforce to enable more space in our balance sheet for doing development. And horribly tough thing to do, it will let your colleagues go and let -- a good number of them go. But if you can find a way to variabilize components of research early on so that you're not having to go through that dynamic process, it would be terrific. We have ways of variabilizing some of our development workforce with CROs that are out there for clinical trials or manufacturing, but there isn't as much on the research side. And so if there were away to keep that variable with external technologies and capabilities, that would be very valuable, because you would never have to unfortunately let people go from your team.

Yes. So thinking about the use of CROs or external innovators, maybe we can kind of close out with -- like how should pharma companies think about open innovation, the use of platform companies, when to think about going external and when to do work in-house? And what would be your advice for them?

Well, I think there's an enormous need to embrace pre-competitive technologies in a very open way. And there are many, many capabilities, technologies that really should be shared very openly across the biopharmaceutical industry. And it ranges from genetic data to different technologies, whether it's delivery technologies. Where companies need to create the intellectual property to ultimately defend their asset comes down to where they have composition of matter on their drug candidate, and that's where they have to defend it. But everything up until then really can be shared in a much more open and pre-competitive manner. And I think there really is an opportunity for the pharmaceutical discovery process to have common elements that can be shared across companies and with a common technology provider in a very open way. So I would welcome that type of opportunity for companies to have access to those technologies.

Yes, we have some. Any last advice you'd give to people in the audience who are building new modalities?

Well, I'll close with a quote from George Bernard Shaw, which I think is applicable to anybody who is developing a novel platform technology. The quote goes along the lines of this: the reasonable man adapts themself to the world; the unreasonable one persists in trying to adapt the world to himself, therefore, all progress depends on the unreasonable man. And so I think to do a modality, a novel modality, you have to just be outright unreasonable in what you're doing. And I think that's the best advice I can give you from my experience. So thank you.

Wise words. Thank you so much. [Break]

Hi, everyone. I'm Quinn. I lead events and experiences at Ginkgo.

And I'm Grace, the Creative Director of Grow, Ginkgo's Magazine. And we could not be more thrilled to welcome you to the final hour of stage programming here at Ferment, a collaboration between Grow and Pop-Up magazine.

Today at Ferment, you've been hearing all about what we're building at Ginkgo, what people are engineering on our platform and what innovation in biotechnology means for health, manufacturing and the climate.

Grow is where we get to explore those possibilities of biology further through storytelling. It's a place where we bring together diverse voices to think about the values that drive synthetic biology and how those values can drive positive change for our futures. If you haven't grab a copy of our latest issue yet, you can do so in the library of upstairs and read some of those stories in print.

This year, we're taking it up a notch by bringing stories from Grow magazine to life, live on stage. We've teamed up with the best storytellers around, Pop-Up magazine. For over a decade, Pop-Up has been producing stories performed alongside animated illustration and accompanied by an original score. If you're not familiar with Pop-Up whose shows are routinely sold out, you're in for a treat.

Today, you'll be hearing 4 stories that have been published in Grow. To start, Nadia Berenstein will survey how to make lab-grown seafood delicious. Alexa Garcia inspired, by Beronda Montgomery's work will share with us what science can teach us about equity.

Next, Claire Evans will meditate on what it means to approach organisms with empathy and care. And finally, Sudeep Agarwala will ponder synthesizing the world's holiest fragrance and what that could mean for us all. You should all have an envelope for a final story. Please wait for our cue to open it.

We hope these stories inspire you, challenge you and encourage you to ask what if we could grow everything. And with that, we'll pass the mic to our first storyteller. As a reminder, this is a performance, so we would love if you could keep your voices down and phones off, so we can all enjoy the full experience. We hope you enjoy the show.

My friend, the journalist, Larissa Zimberoff, is what you might call an adventurous eater. [Presentation]

When Larissa has a bagel with a schmear, the cream cheese comes from protein derived from extreme biomicrobes that grow in geysers at Yellowstone. [Presentation]

Larissa has become one of the world's few connoisseurs of future food in its beta-testing stage, lab to table cuisine. And so last March, when Wildtype invited her to their tasting room to sample sushi-grade salmon grown from immortalized Coho cells, she was down. San Francisco-based Wildtype is a leader in the cell-cultured seafood space. Their salmon is not available to the public, not yet. No cell-based meat is. But the FDA has already given the thumbs up to 2 companies growing chicken from scratch, and Wildtype is one of more than a dozen companies racing to bring cell-cultured seafood to a dinner table near you. Now if you're imagining a slice of lox served fresh from the petri dish, let me set the scene. There's a chef, first of all. Larissa recognized her right away from the last season of Top Chef. And she's showcased Wildtype salmon in 3 elegant presentations: On a slice of toasted brioche with creme fraiche and herbs, served ceviche-style with a dash of citrus and minced and spicy, like the inside of a salmon roll. Wildtype's founders and staff looked on as she dug in. This was a supercharged version of the classic critic-visits-the-restaurant scene we know from movies like Big Night or Chef, except that in this case, the meal in question took years of research and millions of dollars to produce. [Presentation]

Before we got to what it tasted like, why would anyone want to grow salmon from cells? According to the U.N. Food and Agriculture Organization, more than 1/3 of the world's marine fisheries are in danger of collapse. Popular fish like Atlantic Cod, Swordfish, Chilean Sea Bass and Bluefin Tuna are particularly [ threatened. ] And rising levels of heavy metals and other contaminants like microplastics, make eating certain kinds of fish a risky prospect. But this hasn't stopped us from chowing down on fish sticks and tuna rolls. In fact, global seafood consumption more than doubled between 1990 and 2018. The hopeful pioneers of cell-cultured seafood promised a bounty of fish, shellfish and crustaceans, free of toxic pollutants with a fraction of the environmental impact and no animal suffering. But this plan only works if people want to eat the [ stuff. ] You might expect salmon cells grown in a bioreactor to taste a lot like, well, ordinary salmon. After all, the cells used in [ cell ag ] come from real living fish. Same cells, same genes, same flavor, right? But flavor isn't just an expression of DNA, it's also a reflection of the life lived by an animal before it became our food. Flavor tells the story of the journey from planet to plates. Wine people out there may know the schmancy word, terroir. That's French for taste of place, and it refers to the way the land and labor of a specific region shape our experience of the flavors in the glass. The terroir concept has spread from the wine world to encompass all kinds of things, from clams to cannabis. So where does that leave cell-cultured seafood, which has grown in sterile vats designed to be scalable, standardizable and deployable everywhere? Is there such a thing as a laboratory terroir? First things first. How do you build a fish from cells good enough to feed a planet of hungry human plastivores? [Presentation]

Dr. Kaplan and his students are developing the technologies and techniques to cultivate and grow seafood and shellfish, which still remains a couple of steps behind cell-culturing terrestrial animals. A big challenge for researchers working on culturing seafood is our wide-ranging voraciousness for all of the Frutti Di Mare, the manifold and motley fruits of the sea. Hundreds of species of vertebrate fish as well as invertebrates like oysters and octopus. Compare that with pork. Sure, there are different breeds out there, but bacon and chucks all come from the same kind of animal. Not so for salmon and cod or lobster and eel. [Presentation]

To recreate something, you must know it intimately. Establishing new cell lines often means getting on a boat and going fishing. [Presentation]

An epic bout of seasickness, it was still worth it, though. Kaplan and his colleagues caught a mackerel that day that they successfully immortalized. What if the tastiest mackerel that ever did swim in the cold, cold waters of Massachusetts Bay? Who knows? When I began researching the story, I dreamed about endless rivers of salmon cultured from the cells of prize-winning fish, about the possibility of mass cultivating Toro from the tastiest tuna sold at Tsukiji market. I imagine that the fish bequeathing their cells to our cell-cultured future would be chosen in part for their deliciousness. I seem learned that taste is not a primary concern when selecting specimens for cell lines. researchers like Kaplan are looking for cells that are good performers and good team players, not flavor bombs. Think about it this way, when we take a bite of fish, we're consuming different kinds of cells: muscle, fat, connective tissues, arrayed together in accordance with the vital requirements of the species. Think of the flakiness of a grilled sea bass, the way the filet separates into layers with the gentlest prod of a fork. The sea bass's body grows like this so that it can live, but it won't grow this way unassisted in a bioreactor. Producers of cell-culturing seafood must find ways to recreate or mimic the 3-dimensional arrangements of cells. This has to come first. The cellular architecture that produces texture, is the framework for flavor, which brings me back to Larissa and her epic taste test. [Presentation]

The look was right on, but the texture wasn't quite ready for prime time. [Presentation]

The seafood we eat now is often far removed from its place of origin, whether wild caught or farmed, and is almost certainly frozen at some point. Even fish bought [ right off ] the boat at a dockside market may have been captured weeks earlier. But if cell-cultured seafood becomes a reality, seafood could be produced anywhere. This could radically change the landscape of food production. Someday, a prairie oyster grown in bioreactors and landlocked Oklahoma may be as fresh as any Wellfleet [ slurped getting a cod plan base. ] So let's say companies like Wildtype get the texture and taste of cell-cultured seafood to be indistinguishable from the real thing? Well, what is the taste of the real thing? The oceans are changing because of climate change and other anthropogenic factors, so the fish of the near future may not taste quite like the ones we nosh on today. And there's another thing that can change: our appetite and our desires, our sense of what tastes good. Deliciousness isn't just a matter of flavor profiles, it's also social and cultural. What makes food good to us is shaped by what we know about the food, where it comes from, how it's made, by our values. So maybe we have to recalibrate our expectations and learn to savor the untraceable freshness of laboratory terroir and entirely new taste experiences too. [Presentation]

No limits. How to get there, a question for another day. And this is the promise and the pitfall of food futurism. The spoils of the distant future are sometimes much more enticing than the struggles required to attain them. Thank you.

Back in 2020, I was getting a PhD, studying infect the guts and pioneering genetic engineering techniques that could someday save our world from plastic pollution. We've just learned, that the humble mealworm could digest biodegrade and even subsist on plastic, surprisingly potentially transformative capability in an insect whose most popular use today was [ as fishing date ] at least those are my high hopes. But the arrival of a global pandemic meant that I couldn't go into the lab. And my precious mealworms within while we all waited for return to normal but not a game. That's when I turned to gardening. I started with a single succulent. I figured it would be hard to kill. Within 4 months, I was a proud steward of 2 jalapeno plants, a lime tree, 6 marijuana plants, a wide assortment of herbs, countless cacti and a cute little strawberry bush. I spent every morning out of my tiny balcony admiring greenery. That strange summer, under the pervasive threats of respiratory illness, police brutality and all-consuming pessimism, I found solace in my gardening. It's easier than ever to feel alienated from our work and from each other. More and more remote work and productivity apps are muddying the lines between our personal and our professional lives. But gardening is the opposite of all of that. It's dirt and water, silence, seed, it's rooted in reality. There's a direct connection between what you put into it and what you get out of it. And it begins with close observation. You monitor the plants as they grow. And when things start to go awry like wilting leaves or an infestation of pests, you don't blame the plants, you intervene. Maybe they need a sunnier spot or more water. For a plant to thrive, it's the gardener's job to make sure the environment is suited to its success. Sadly when it comes to humans thriving, we often forget to consider the impacts of the environment. This is true in a lot of communities, but especially in STEM academia and industry. [Presentation]

That's Beronda L Montgomery, a biologist, author and avid gardener. She spent a lot of time thinking about what society can learn from the natural world. She wrote about it at Grow. [Presentation]

And the opposite of that? [Presentation]

Approximately 2/3 of the workers in STEM are white. Black, Indigenous and Latinx people are all severely underrepresented, and the workforce only gets more blindingly white the closer you get to the boardroom. [Presentation]

The cost of this monoculture has never been clearer. Take the case of pulse oximeters. For decades, white scientists have known and ignored the fact that these devices, which monitor the oxygen level in our blood don't work accurately on darker skin. COVID brought this fact to life and more. Throughout the pandemic, America's long history of biomedical racism and the long-standing underrepresentation of Black, Indigenous and Latinx people in STEM, has led to bad health, lack of access to quality healthcare and in some cases, to an understandable distrust of public science. Now a few black scientists in their labs are working to improve pulse oximeter technology, which, in turn, could ultimately improve health outcomes and survival rates for all melanated people. There's so much more we can do in the right environment. Throughout my time in academia, I've worked on diversity, equity and inclusion initiatives. At MIT, I was an undergraduate ambassador for recruiting underrepresented minority students. During graduate school at Stanford, it was much the same, recruiting, advocating to professors and administrators, speaking on panels. I was pursuing my PhD with plans of becoming a professor and mentoring other black students to success. But when I asked the one black woman of professor of engineering for career advice, she said, well, we have serious discrimination but that's always expected. By the end of 2020 after a rewarding and fruitful season in my garden, I no longer wanted to expect discrimination or isolation or work unrewarded. I wanted better. In STEM, a few minorities are introduced into a monoculture and expected to thrive. When they don't, well, there must be ripples. [Presentation]

What is as Beronda suggests, the approach to STEM monoculture with the same care and attention that we give to our struggling gardens. There are widespread benefits to breaking up monocultures. [Presentation]

A good plant caretaker knows that each member of the community has particular potentials and needs and fosters synergies between them. They take a growth-based perspective and refine their strategies. They talk to their community exchanging wisdom and references. We can't do this in STEM, it will require some changes. [Presentation]

Gardening taught me to expect more from the people and institutions around me. Whether I'm working in the dirt or the lab or the office, I am always seeking to build a supportive environment I desire, a polyculture that not only allows but encourages everyone, truly everyone, to thrive. Thank you.

It's 2 in the morning. And at the University of Toronto, a postdoctoral researcher named Deboleena Roy, is still awake. Her job involves culturing rat neurons, it's demanding work. The cells can be finicky and they need to be monitored closely. In fact, Deboleena has to check in on them every 6 hours, which is why she's still here in the lab, in the middle of the night. [Presentation]

See rat cells multiply quickly. And so to keep the lab from swimming in them, Deboleena has to regularly call and dispose of extra cells. That is to say, she has to kill some. It's a routine process called subculturing. But tonight, something not so routine happens. [Presentation]

A biologist is supposed to stand at some remove from her subjects in the lab. But Deboleena in that moment, realizes that her relationship with these cells is not one-sided. The cycles of their lives are regulating her own life too. She's acting on the cells, yes, but they're acting on her as well. And now, she's just about to kill them, just like that. [Presentation]

She hesitated holding the incubator door. It's just a pause, but a meaningful one, because after it passes Deboleena realized that she will never look at cells the same way again. Deboleena Roy is not the first scientist to feel so deeply connected to her lab subjects. In the 1920s, a young cytogeneticist named Barbara McClintock appeared on the scene. She spent 50 years studying corn, and was known for observing her specimens very, very closely. McClintock could pick up on even the most minute changes in each of her plants. She chastised other scientists for attempting to impose an answer on their subjects. She preferred to let the answers come to her through careful sustained observation. It was slow, but fruitful. Although marginalized for half a century, her landmark work ultimately proved the chromosomal basis of genetics, earning her a Nobel Prize in 1983. When McClintock understood about her role as a scientist was this, the closer and more willingly you look at the world, the more it allows you to see. McClintock called this approach developing a feeling for the organism. And for her, nurturing that feeling was an essential part of being a good scientist. McClintock's extraordinary sensitivity earned her reputation as a mystic. She believes that all life is interdependent and interconnected. This understanding has long been part of indigenous views of the world, but it was and continues to be neglected by many of the scientists. Deboleena Roy wants to change that. That early morning experience in the lab with those rat cells, it led her to propose an entirely different scientific philosophy, one that builds on the McClintock's feeling for the organism. It's one that opens up new avenues for research and could, in the process, make us more humane. In the lab, scientists for reciprocal arrangements from cells. They provide or withhold those conditions that allow life to thrive. In turn, the cells revealed their behavior under novel circumstances. This is our relationship, but is it a healthy one? Some feminist scholars like Deboleena Roy have argued that the Western scientific worldview is rooted in the idea that we can control and subjugate an unfeeling world. And indeed, foundational thinkers in the history of science like Francis Bacon, who wrote about enslaving nature to the service of man; and Rene Descartes, who once compared the cries of a wounded dog to the sound of an improperly functioning machine, they've casted a long shadow. But could we step away from this shadow and shine a light on a new, more equitable path? Deboleena Roy thinks so, but only if we can move away from that input to dominate. [Presentation]

She believes that this world view creates hierarchies where they don't exist in nature. Instead, Deboleena proposes that we place ourselves along a continuous plane, human and nonhuman alike. After all, we're different from cells, but we're also made of cells. We are live acting on itself. And shifting our perspective in this way, it seems like a small act but it could have an outsized effect on the advancements that happen in science labs. [Presentation]

Making an effort to feel for organisms in their fullness isn't only a philosophical position. Cozying up to the small thing has its practical advantages too. [Presentation]

Sarah Richardson is a molecular biologist. Her hope is that humankind and bacteria can work together to produce useful materials, breakdown waste and ultimately save the world. For example, she hopes that someday train wild bacteria to convert unused biomass into petrochemicals. And to achieve these goals, Sarah and her team approached the bacteria that they work with not as subjects, but as collaborators, and that takes humility. [Presentation]

Scientists like Sarah Richardson are showing us that there are ways of encountering the nonhuman world, even putting it to work for us that emerge from a fundamental respect for life. Instead of dominance, Sarah opts for a different kind of power negotiation. [Presentation]

Like Barbara McClintock, Sarah let the organism speak for themselves. She studies where bacteria come from, what motivates their behaviors and what they need in order to thrive. Only then does she offer them a deal. Help us, and we'll help you. If you want to succeed in bioengineering, it's easier to work with life than against it. [Presentation]

When we put our egos aside, it becomes easier to have a particular kind of empathy with the organisms that we study, not empathy that would render us unable to experiment with them at all, but one that helps us to brokering more sustainable arrangements with the nonhuman world. And that empathy serves us is what drives her curiosity about the organism she studies. [Presentation]

It's a process we've used for centuries, from taming wild dogs to transforming maize into corn. And it's still at work. You see it in McClintock's discovery of the transposon, a DNA sequence that can change its position within a genome long before the invention of genetic sequencing. Or in Roy's cell culturing method, which help you discover new forms of communication between estrogen receptors and brain molecules. This approach is mutually beneficial most of the time. [Presentation]

Synthetic biologists have built their disciplines around the premise that life is a text that can be edited, rewritten and translated by organisms. Once bacteria have performed the task we wanted them though, they're promptly discarded. But what if there was more to explore there? [Presentation]

Deboleena said it's a missed opportunity to deem our counterparts in the lab worthy of our care and attention only when they're productive. Because how we treat the microcosm is a reflection of our attitude in the macrocosm. [Presentation]

When we question hierarchies, even at the molecular level, we're engaging in a radical act of humility, and it's one that manifests in our human relationships too. It's a project of a lifetime for us to see ourselves as one of many. We are all carriers in our own way of the patterns of life. And by deliberately decentering the human perspective, we only become more humane. Thank you.

I am standing at a grove of trees in Northeast India. Everything around me is absurdly green, must be the greenest place I've ever been. My ancestors are from this land and though my parents raise me far away in America, I still identify with it. I recognize certain aspects from -- summers traveling in Kolkata, the vast paddies of rice and fields of jute, the fruits sold on the trees, the color of the soil, even the smell. I'm here on a work trip, inspecting trees used for perfume and incense. My mission seems preordained. Shortly after the partition of India in 1947, my Hindu grandfather changed our last name from Poddar to the more cosmopolitan Agarwala. Agarwala, literally a peddler, wala; of incense, agar-batti. Somehow, I have walked into a trap set by my family 2 generations ago. You see for the last 15 years, I've studied yeast biology and genetics. And now I engineer yeast to produce rare flavors and fragrances that are slowly disappearing from the world. This air is still and hot, and I am comically sweaty. It's hard, but I must focus. The trees in this grove have been planted and carefully manicured so that after a few years of cultivation, they can be treated with a controlled fungal infection, which rots the wood and creates its trademark scent. A guide takes me to a bandaged tree and gently unwraps it. He cuts off a piece of wood with his pocketknife and rubs it between his palms to release some of the fragrance. It's musty, earthy like compost, but there's a slight sharpness that's hauntingly familiar. As we wander the property, the guide tells me this tree is the only wood that was allowed to be taken from the Garden of Eden. Ashamed of their nakedness, Adam and Eve wrapped themselves in its bark during their flight from Paradise. In the Hebrew bible, the prophet Balaam says the tree was planted in Israel by God himself. The Psalms describe the coming Messiah as being anointed in the fragrance of the wood's oil. Later on, the prophet Mohammed fell in love with the same smell. To this day, its fragrance is mixed with water from the Zamzam well and is used to wash the Kaaba twice a year at the center of Al-Masjid Al-Haram, Mecca, the holiest place in all of Islam. If you haven't already, I encourage you to open up your envelopes. Later that afternoon in an office next to a processing facility, I'm presented with a distilled version of the smell. The facility director opens a vial of amber liquid, and since it's too precious to hand over, wafts it towards me. It's a dark, heavy musk. One that hangs in the air well after the vials have been taken away. Its scent is a conduit of revelation, the fragrance of history and myth of God and his prophets. The conspiracy birth by my grandfather 70 years ago and carried on through my research has brought me to this place, and I understand why: I'm here to recreate God's own scent in yeast. I am young, maybe 6, on vacation in India with my parents, my cousins and I sneak into a sparse tidy room that belongs to the house matriarch, the woman who raised everyone who grew up here. Some of whom, like my father, moved away to start new lives in the United States. I am not supposed to be here and my young heart is pounding. In that moment, a lasting association is made. A faint scent that turns out to be endlessly intoxicating. This is the smell we have loved for thousands of years. But will the scent I create in yeast be real? Here's what I mean. Maybe you know the story, the myth from ancient Greece. Theseus sailed to Crete with his army to slay the Minotaur, and then sailed back a hero. For generations in Athens, his ship was well preserved. Every time a plank rotted or broke, it was carefully renovated until the entire ship was replaced by newer, stronger materials. Which raises the question, is that ship the same one that Theseus sailed to Crete? I'd argue that there was no ship of Theseus to begin with. As he traveled to Crete, then back, the ship was constantly changing. Either this thing we call the ship of Theseus has always been the ship of Theseus, or it never was. So back in Boston, I think about how to explain my projects, not to my clients, but to God. I imagine what the elephant-headed deity Ganesh would think. See when he was a boy, he had a human head. But he was killed by the God Shiva in battle. His mother grief-stricken and enraged, threatened to burn down all of creation unless her beloved boy were brought back to life. She demanded that he be made immortal and worshipped as a god. Shiva was chastened, immediately sending out his servants to bring the head of the first creature they encountered: an elephant. Ganesh was revived and attained immortality only after the gods spliced the elephant head onto the body of the boy. I think of Ganesh because the groves of trees that produced this beautiful fragrance are dying for reasons beyond our control, and how it is my job to identify the exact molecules that produce this fragrance. I will design snippets of DNA with encoded proteins that are capable of making this molecule. I'll synthesize that DNA and introduce it into a fungal species, the yeast that we know how to manipulate. And I'll coax it to produce the same fragrance down to the molecule as the perfume that pleased God from the beginning. But instead of living in the plants that God himself planted, it will come from this new yeast organism that I have created, grown by the liter, thousands of liter, in gleaming steel fermentation vessels. I confess, I didn't sleep well in India, and it wasn't the jetlag. I worry that I may be taking away something important and foundational from a scent that has been dear to the world throughout its history. Even though it might all be fiction, it's impossible to say whether the plant mentioned in the Bible is the same one written in the Vedas or the same one beloved by the prophet Mohammed. Still, this weighs on me. All of it. Is it rude that I should tell Ganesh what a burden immortality is? But there's hope. If things go well and the pieces align, I could create a new thing, a new living organism that will create the same molecules that lend the trees in India their awesome fragrance. It will survive long after the groves in Northeast India are no longer there. It has the potential to live forever. This could mean that someday, we wouldn't need to mow down these splendid groves to attain their fragrance. We could divorce this oil from its various plant species, growth conditions and processing technologies. We could democratize it, so it descends from its stations among Prophets and Messiahs, becomes available for everyone to know and love. You could use synthetic biology to give divinity a new meaning and purpose, not for it to be hidden, but to greet and welcome us all where we stand. Thank you.

I'd like all of you to please join me in thanking our storytellers, Nadia, Alexa, Claire and Sudeep. And also the team from Pop-Up and Grow Magazine. All right. That brings us to the end of our programming today at Ferment. I want to thank all of you for the amazing conversations I got to have with you. I want to thank all of our speakers. I think we accomplished our goal today of learning a lot. And I have to say personally, I'm super inspired by all of this. It's really been amazing. I want to give a very special thank you, and please join me in this to the Ferment team at Ginkgo: [ Quinn, ] Alex, [ Malik, ] Joseph, Christina, [ Jake, ] Connie, happy birthday Connie, [ Nina, Danielle, Hannah ] and the rest of our marketing team and the wonderful team here at [indiscernible] who've been taking care of us today. Please join me in thanking them. Okay. And with that, all that's left is the party. That concludes Ginkgo Ferment. Thanks again, everyone. Thanks for joining us.