 I'm going to start by suggesting three potential hypotheses that might explain why proteins look similar. One simple way is maybe all proteins just derived from a common ancestor and that would be what I call evolutionary diverges. In the sense that they start from one protein and then all proteins have branched out from that. Some of them have changed functions, some of them might even have changed shape a bit, but obviously if you're evolutionary related it makes sense that they have the same structure. Another potential possibility though is that it's kind of the twin brother of that. Maybe we rather have some sort of still evolution but functional convergence. So it's kind of the same but not quite. The idea here is that these proteins they might have evolved independently but if some folds are very stable or very good for function it's easier to create binding sites in some folds. It's very easy to insert some of them in membranes. Maybe that's why we have all the alpha helices. It could just be that evolution tends to converge to the folds that are functional and nice and stable. That also makes sense. While the third alternative is maybe there is only a limited number of stable folds. For now we don't know which one of these is true, but it's kind of a triangle pyramid. They might be related, right? Obviously if you want to insert something in a membrane it's going to be nice and easy to have units that correspond roughly to the membrane thickness. It makes sense that those are alpha helicals. In that case maybe we're over here in functional convergence. There are a few beta sheets I'm going to show you one later today but it's pretty rare to have beta sheets in membranes and the main reason for that is that you don't want to expose the edge of the beta sheet to the lipids and the only way to get around that is having one of those barrels that I've showed you. You could certainly imagine that there is a mix of evolution diverging different folds into different functions and occasionally having different functions converged into stable structures too. And in all of these that just as there are only two stable secondary structures it could of course also be related to the limitations of physics. What I'm going to try not just to argue but more or less prove today is that we are very much in this camp. It turns out that most ways that you can organize a sequence in three dimensions is not going to be particularly stable. So there are only a relatively small number of, call them topologies, ways of physically arranging the amino acid backbone that will lead to stable proteins. That's kind of the basic rules that nature has then used in terms of evolution to either diverge into different functionality or reuse stable building blocks.