 Dear students, in this module, I'll be going into building the rationale behind Abinishow modeling. The term Abinishow means from scratch. So if you have a protein sequence that you want to predict the structure of, you don't have a protein database at your disposal and you have to predict its structure from scratch. So Abinishow methods, they primarily rely on the energy of the predicted structure. So they use the overall energy that is contained within a structure and then try to minimize it. So if you have a sequence, you can predict multiple structures that can be formed from this sequence and from the resulting structures, you have to select the one with the lowest energy. This is the rationale behind Abinishow modeling. Moreover, there can be a case where you have predicted a very nice structure with a very low energy but it does not have a biological plausibility. By biological plausibility, I mean the structure that you have predicted may be misfolded. It may have a problem in folding its hydrophobic core or there could be some problems with the membrane. So why Abinishow? If you compare Abinishow methods with homology modeling or for recognition, you will realize that in these two methods, we were trying to see, we were trying to predict the structure of the sequence at hand by looking at the structures that are already known. This will obviously bias the entire structure problem towards the structures that are already known. What if you have a sequence with you which does not have any homologue and therefore may have a unique structure? So to predict such unique structures, you do not want to use homology modeling or for recognition. Abinishow methods suit such a scenario. The novel structures that continue to be reported every day can be easily computed by using Abinishow modeling. Okay, if you were using homology modeling or for recognition, you could never predict a structure that is very different from the structure in the PDB. So all of your search for structures will be biased and therefore the novelty in the structure prediction will be finished. Homology methods help restore the physical and chemical novelty that may exist within a structure thereby creating novel structures. So what do I mean by physical and chemical properties of structures? So in case of homology modeling, we are not considering the different bonds that are created between the amino acids. We are also not considering the electrostatic interactions that may be existing within the structure. All we are trying to say is the sequence matches the structure. But in Abinishow methods, we try to consider the bond energies, the electrostatic interactions, the van der Waals forces and we combine all of them and try to minimize it. These physical interactions between the different atoms or amino acids is going to help us lay a foundation, a physics and chemistry foundation for each structure. Homology modeling or for recognition do not do that. So in conclusion, the Abinishow methods, they rely on physics and chemistry of various atoms and amino acids and then predict the structure. In case of homology modeling, this is not so. So in both the cases that is Abinishow versus homology modeling or for recognition, there are some advantages and disadvantages and we need to carefully weigh which strategy we want to use for a specific sequence.