 Dear students in this module, we are going to look at the workflow for structural modelling or structural prediction of proteins. As you already know, homology modelling can be used to predict the structure of those proteins whose structure is unknown but their sequence is known. The process works by comparing the sequence of this protein which has an unknown structure to other proteins whose sequence and structure are known. By comparing the sequences, we can talk about the possibilities of similarity in the structure of these proteins. So this is essentially the theme of homology modelling. So let's consider how the workflow or the flowchart of homology modelling works. So first of all, you need to start with finding the DNA sequence. So once you have the DNA sequence, then you can find the ORFs. As you know, there are six possible ORFs and then you find the longest ORF which is then used to translate the protein. So by looking at the ORFs and then the translation process, you can come up with the amino acid sequence or the primary sequence. As you know, the proteins have the primary, that is one prime, secondary, two prime, tertiary, three prime and quaternary, four prime structures. Next, if you don't have the DNA sequence here, then what do you do? So you cannot translate the DNA sequence into the amino acid sequence. In such a situation, you have no choice but to look at the protein that you have and try to sequence it using admin degradation or mass spectrometry. So in case you don't know the DNA sequence, you can always go after mass spec. Now, since you have the sequence with you for your protein, you can search for other proteins in the databases such as Uniprot. So Uniprot is a sequence database or SwissProt, which is also a sequence database. So you can search for your protein within these databases to find other proteins and those proteins should have a similar sequence. So towards this goal, after you search the databases, the sequence databases, you need to perform a blast, a sequence blast. So the blast algorithm will give you the best or the closest sequences from the database when compared with your protein. Okay, next. Now you have your protein sequence along with similar proteins from the database, yes. And all you have to do now is find out which proteins have a structure which is known as well with them. For instance, if you found four different proteins from the database search, you need to go and find these sequences from the protein database or the PDB. So let's say all four of these sequences had a protein structure for them in the PDB. So in that case, you are in a good position to perform homology modeling because the 3D structure of all these proteins is known, right? But it can be the case that one of these proteins does not have its structure which is known. So in that case, you can leave that sequence out and only perform homology modeling for those sequences whose 3D structure is also known. So you move on to homology modeling. Remember that the only requirement for homology modeling is that the structure of the proteins is known. So the proteins that came from the blast should have their 3D structures also with them. In case they do not have the 3D structures with them, then there are other methods such as motif recognition in which you search the secondary structural databases. Sometimes it's also called pole recognition. You can also do ab initio modeling which is a complex method towards determining the protein structure. So overall, there are three different strategies for predicting the structure of proteins. The first one is the homology modeling which is the case in which you have other proteins whose sequence and structure is known and then you come back to your unknown protein and you try to predict its structure. If that is not the case, you go for the motif or pole recognition and if that is also not possible then you go for ab initio modeling. So next we will proceed to specifics of homology modeling that is the seven important steps that are there to perform homology modeling towards predicting the structure of proteins. This we will see in the next module.