 To do this type of simulations we're going to need to introduce a few other concepts first because it sounds easy that just take a protein structure. It's not quite that easy in practice. I'm going to start with something easier than a protein. This was a very old system coming from my days as a postdoc. I was working in the Bay Area, Stanford actually, and this was colleagues working at UC Davis. They had a company that they're working together with that was very interested in this system. So this particular system, what you have in the middle, is a so-called tannine, and these gray molecules are ethanol. So in fact, together with water, this is actually a simulation of red wine. This particular company was Gallo that provides, at least provided, significant funding to their unological faculty in Davis. And I'm a big fan of red wines. It was great fun to try to simulate that system at least. Unfortunately, we didn't get very far at the time. But this won't work because you need that the other part, right? We need the water, too. So what you rather should think about a system in this way, it's easier to see here than for a large protein. But do you see here that we probably have 80 or 90% of the atoms in this system corresponds to the atoms in the water molecules? So yes, technically, I'm simulating a tannine and some ethanol, but in practice, I'm mostly simulating water. This is going to be equally true for large proteins. In some cases, it's even worse for large proteins that more than half of the atoms in the systems are typically water. And if you're unlucky, it can be 80%. And that means that we typically make do with highly simplified water models to make sure that they're very fast. For instance, I typically ignore the Lennard-Jones interactions and the hydrogens and only include that on the oxygen. It speeds up the simulation a bit. The other problem I'm going to get here, though, is that here I have a surface between water and air or vacuum in this case. This is going to lead to strange surface tensions. So to get away from those surface tensions, I'm going to need to have a very large system, and you probably haven't seen square things in the lab. I would need to have some sort of sphere here. So that's going to be a gigantic system that I don't want those surface tensions in the first place. But maybe we can find a way around that by learning from physics, in particular, simple system in physics that have already solved this. Let me show you.