 So armed with that, it turns out that there aren't that many protein topologies or whatever we call them. On a high level, I would argue that the simplest protein we could imagine is to just take one beta sheet, one layer. And yet you never see that. The reason for that is that that single layer, it's going to be exposed to water on both sides. And sure, technically we can fold it, but what does it do? Not really anything. Same thing as a single helix hairpin. I can't fold it, but it's not going to do much functionally. At least I can't come up with anything immediately. And what's the reason for having the protein? It's not going to be good for evolution. So one layer, so it's not so useful. Technically it can exist. It's not unstable, but evolution is not going to select for it. Two layers on the other hand. That's going to be awesome because now we have shielding. Think of this more beta sheet pockets I showed you, right? If I have two layers, sheet one, sheet two, I can have an inside between them and I can have an outside with different properties. In principle, it works just as well with alpha helices. Three layers. You've already seen that. That's the Rossman fold. The Rossman fold, that kind of has two pockets, right? One on each side of the beta sheet because we had helix sheet helix. So here you have two small pockets and they might both be hydrophobic, but that's fine. I can survive two pockets. All inside residues here are going to be hydrophobic. Works great. Four layers. Ooh, I'll write maybe. They are rare. Why are they rare? Well you see, if I now, let's draw some layers here. Here I have layer one. Here I have layer two. Here I have layer three. You can probably guess the number of that layer. Here I'm going to have water. Over here I'm going to have water. This layer is definitely not facing the water. So here I will say O for oil. So this part has to be hydrophobic. But wait, here on the inside, now I have a problem here. I could make that hydrophobic, but if I make it hydrophobic, there isn't really going to be any strong driving force why the residues here should face that layer. They could face that layer instead. Okay, let's make something better. I'll make that hydrophilic so that it likes water. But now I have the opposite problem. Now I'm burying something that would like to be water. This could very well unfold instead. So because I either am going to end up with something that's very unspecific in terms of packing or I need to bury water-liking residues, hydrophilic ones. There are exceptions where this happens when it's particularly important for function. I can't even come up with anything immediately. But if you start seeing them, be happy. They're not going to be common. Five layers. And now I exaggerate a bit. We don't really see them in practice. It gets too complicated. Again, I'm sure, as smart as you are, you might be able to design something that could potentially be stable there. But it's not going to be enough to be stable there. We need to be able to fold. We need to make sure that we don't misfold and everything. The whole fold has to be unique. And that turns out to be too complicated in practice. So the take-home message here is that most proteins tend to stick to two to four layers. They're going to be relatively small, which means that they can fold in finite time. And the reason for this limitation is primarily the hydrophobic versus hydrophilic matching. You can't be too small. Then you can't do anything. You can't be too large. Then it becomes too difficult to be stable.