 But now I want to get really big and very practical. How do you go from a nature paper to 500 million liters of fuel? These high volume challenges are also global challenges. Access to food, platform chemicals for industry, energy security. Products need to be cost competitive. But in synthetic biology, this is something that is often left unstated. In fact, sometimes we don't even acknowledge it at all. And that's a challenge that interests me. So today I want to talk to you about scale-up. How do you go from something that works once in a test tube to large populations of engineered cells in an industrial bioreactor, working day in and day out, making products in some of the largest factories on the planet, and making them economically? Because discovery isn't the same as delivery. You might invest 10 or even 100 million dollars in a synthetic biology application, only to find that you need 10 or 100 times that in order to take it to market. So it would seem that synthetic biology has a problem. But I think synthetic biology can also offer a solution. One of the challenges with scale-up today is that it's a really slow process. We rely on trial and error, and there's a big difference between looking after a single cell and looking after an entire population. Just as there's a huge difference between running a single store and running a global franchise. Each store needs to make the right product to the right standard. But each store is also an individual, and it's the same with cells. You might think all of the cells in a bioreactor are the same. They're supposed to be the same, but they aren't. And that's because they adapt and they change as they grow to fill the bioreactor. And like a franchise gone rogue, some of those changes can eventually interfere with their ability to make the product. One of the reasons that cells might start misbehaving as you move to a larger scale is that keeping their environment the same is pretty hard. You need to constantly mix in fresh oxygen and nutrients and spread things out evenly. But dead zones inevitably arise. Spaces where cells don't have enough oxygen or food. And the cells that get trapped there can get pretty unhappy and they may start to revolt, and they may even eventually die. So in my lab at Imperial College, London, we're trying to develop ways to detect when cells are starting to become unhappy. So we can take corrective action and make them happy again. And we use a scientific approach to try to eliminate the kind of trial and error that slows translation. But cells can't talk. So how do you figure out when a cell is unhappy? We use a biosensor to give each cell a voice. And the reason we do this is because we could instead try to engineer cells to do something differently, but that's a long, slow and difficult process. And instead what we'd like to do is be able to detect when the cells are becoming unhappy and figure out what we can do to their environment, which is much faster. The biosensors we use are a set of instructions written in DNA that causes a cell to change color in poor conditions. Take for example lactate. This is a poison that accumulates when cells don't have enough oxygen. And it's actually the same thing that will make your muscles hurt when you work out. We've created a biosensor for lactate so that when lactate concentration increases, the cells turn color and they tell us that we need to increase their air supply. So equipped with a biosensor, every cell in a bioreactor can tell us how it's doing. And we can use that information to act if we need to, to fix local problems and manage systems efficiently and turn unhappy cells into happy cells. But we do face some challenges. We don't always know what will make the best biosensor and there are thousands of possibilities. And also it's important that our choice does not interfere with the cell's ability to make the product. I think that this is a challenge for biology in general and that's why delivery is also a research challenge. It's also an applied research challenge because testing our sensors in a laboratory setting can be artificial. So having access to industrial facilities to validate them is key. And I think this is one place where industry and academia can work together so that speeding up delivery becomes a shared priority. A scientific approach to pipeline delivery means that we can bring products to market faster and we can start to design useful features like biosensors into cells from the start. And it should mean that the cost of bringing a new synthetic biology product to market need no longer be something eye-watering. Scale-up is a one-time investment so it's worth getting it right and it's needed for competitive products. It will benefit from the kind of know-how that research can generate. But since we're talking about a self-replicating technology once that know-how has been baked into a product how do you prevent others from stealing it? This is a question I'd like to discover with you today. Thank you.