 Next database is SCOP. SCOP is slightly different, so the strength of CAD is that it's automatic, but that occasionally also means that the rules of physics or math in this case will place folds in places where it's not entirely logical, or at least logical in our sense because we tend to focus on the function of proteins and in particular how they have evolved evolutionary. SCOP is a database called structural classification of proteins. Structural classification of, and in many ways it's the opposite of CAD. This is a so-called manually curated database, so in particular Alexi Murtzin has spent an eternity of time looking at structures and deciding based not just on the look and the topology of the sequence, but what it does and how it's as likely evolved in what class it belongs best. So SCOP uses slightly different names for things, but of course the overall concept is the same. So let's have a look at SCOP the same way. There's a website for this too, and you can just find it by search for SCOP database. It's in Cambridge. So the SCOP website is less graphical but exceptionally useful, and this too is kept updated quite frequently, but it can occasionally take some time to classify things here, particularly if you use manual curation to get it better. Here we can start to browse by the structural class. What should we look at here? Should we look at alpha proteins because we look at beta for CAD. Here we have folds. So we started by looking at the class of all alpha, and under that now we have folds. This would kind of be architectures in CAD. So do you see here that you have left-handed parallel coils, SAM domain, Spectrum repeat like. There's a lot of information here. So we see if we can find something that we might recognize here. I'm going to be nasty and search for something. We should have a Globin fold here somewhere. The Globin like fold. Do you see here? It describes this as a core of six helices, a folded leaf partly opened. Under the fold you have what SCOP called super families, and then the super family we started in this particular case it happens to have the same name. But now we're on the level where things are evolutionary related, and within the super family we now have the families that are even closer related here. So now we're down on the family of Globins, and these are quite closely related. They're all heme binding proteins, and here I should be able to click on something. Do you see that there's a PDB linked to these two? Because we wouldn't have the structures if we didn't have a PDB structure. Now I can find the four domains in this particular protein and check what each of them looks like. And here I see the sequence, this domain compared to the unit protein, and then I can even look at this. So in this particular case this is the domain while this is the entire PDB file. So SCOP is just as useful as cats. They have slightly different strength and weaknesses. SCOP is manually curated. Occasionally that is exceptionally useful because it tells us more than just the architecture. But the drawback is that occasionally it's lower and even humans make errors. Cat for better or worse is mostly automatic.