 Dear students, now we will go in depth into the seven steps that are to be done during homology modeling. So just to warm you up to the background, so there were three different strategies to predict the structure of proteins. The first one was homology modeling of course, while there was full recognition or threading as well as ab initio full prediction. So to start with, let's take a look at the seven very important steps or the seven salient steps that any homology modeling pipeline has. So these steps are crucial in their order and any software that is written for homology modeling employs each one of these steps in different shapes and forms. Before we get into the details of the seven steps, you need to understand two terms. So the first one is the template and the second one is the target. So the template is your protein which has the sequence and the structure both known. While your target is the protein which has got only the sequence known and the structure is unknown and therefore we want to predict its structure. So template versus target. Now let's move to the seven steps. So we'll briefly go over them. So any homology modeling software in bioinformatics begins by identifying which template is to be used and then performing an alignment. So remember the template has both the sequence and the structure they are known and if the alignment is performed then you may need a slight bit of correction in the assignment by optimizing the gap penalties and mismatches. So once you have formalized the alignment, next you create the backbone of the protein. So you need to create the backbone by looking at the alignment. So after forming the backbone, after generating the backbone, next you need to look at the various loops that are going to join the backbone like that. And of course this is followed by, as you know already, that every protein is formed of amino acids and every amino acid has a side chain. So you have to add side chains accordingly and then you follow it up by optimizing the model that is by optimizing the structure of this protein. So once you have optimized it, all you need to do then is to validate it versus the known protein structure. So let's take a look at these seven steps in a flow chart. So you begin with the amino acid sequence of the target structure and then you identify suitable 3D templates that is the template proteins. Remember the template proteins have their sequence and the structure known, both of them, while the target does not have its structure known and you're trying to predict it. So this is followed up by listing down of the templates. So there can be two cases, one in which you only find one template here and the other case in which you find multiple templates. So in case you have a single template from the PDB that is coming to you, then you align the template and the target sequence and then you go for 3D structural alignment of the templates. But in case you had more than one template, then you have to align the templates at the amino acid level for all of these templates and then perform the 3D structure alignment. So you have to keep repeating this loop multiple times for each structural prediction. So it depends on how many templates you had. So once this is finalized, then you perform the alignment of the template and target sequence. So the comparative model generation takes place here, which is followed by validation of the model. However, in case the template and the target are not aligned nicely, then you would want to do secondary structure prediction such as the fold recognition or threading. So homology modeling essentially takes place here. But the overall homology prediction employs these steps. Please remember this portion of the flow chart. So to conclude, homology modeling is a seven step process and at each step you are trying to improve the quality of your prediction. More so it is a repetitive process that is you have to repeat it for multiple templates, multiple times and therefore arrive at the best match. Next we will look at each step within the homology modeling pipeline in detail.