 Once the template search is finished, we land on this page. This view shows a list of templates that have been found for your target protein. The list is already sorted according to a template score, meaning that good templates are most likely to be found further up in the list. But let's see what we are presented with. First is a box to select and deselect templates. If we select them, they are shown in the 3D viewer. In addition, selected templates will be used for modeling when we click the build model button. The second row is the name of the template in the Swiss model template library. I come back to this one in a bit. The third is the coverage, meaning how much of our sequence is covered by the template. We are very lucky in this case, since our sequence is completely covered. But often we can learn a lot about a protein, even if we can only build part of it, if the important domains are included. Identity shows how many amino acids are exactly aligned. Identity is only one criterion for choosing a template. But obviously, higher identity tends to give better models. Resolute and resolution should be self-explanatory. If the resolution is too low, we might want to use a different template. The oligomeric state is of course very important. Sometimes you get a warning, like here for 1e90.1.b. Whenever the oligomeric state of the template cannot be used for the model, it is estimated to be different, a warning and explanation are shown. The last column shows the ligands in the template. Depending on what we want to do with the model, this might be very important. Out of experience, getting a ligand in the right place with the right confirmation can be very challenging. However, a word of warning. This model is very conservative, meaning it will remove ligands relatively quickly if it causes steric clashes with the model, or if the binding site is not conserved. This arrow here opens the Detail tab. Let's use it on the top two and the fourth template. As I mentioned, those are the most likely to be suited for modeling. Here we see an image of the template structure and some more information. GMQE stands for Global Model Quality Estimate, which is a number between 0 and 1. The closer to 1, the better. As the name suggests, this number estimates the expected quality of the model, given this alignment we have. Sequence similarity shows the similarity between the two sequences. Next, we have the Quartinary Structure Quality Estimate. This score, again between 0 and 1, reflects how well the inter-chain interactions are, given the alignment. Simply said, a score above 0.7 is a very strong indication that the Quartinary Structure of the template is a good representation of the Quartinary Structure of our model we want to build. Lastly, the Oligomeric State of the Bio Unit. Here you can override the Oligomeric State that was automatically assigned for the model. All in all, we have four scores that indicate how suitable a template might be. The GMQE and the QSQE scores, as well as sequence identity and similarity. From these templates, I would probably not pick the top one, since it has no copper. You might have noticed that the second and the fourth templates have the same name, but have slightly different scores. One template was found with Blast, and the other with HH Splits. As we see later, the sequence alignments are not exactly the same. And this has an effect on the scores, and more importantly, on the final model. In order to find the correct Quartinary State, we should go to the Quartinary Structure tab.