Optimizing peptide design


I would like to design an optimized peptide to inhibit a protein-protein interaction. I have limited bioinformatics experience and am currently using the haddock 2.4 webserver to dock my starting peptide to my protein of interest. The webserver gives me the docking results and I will take the models in the top HADDOCKscoring clusters into pymol and look at how the peptide binds to the protein and the protein-peptide interactions and then I make point-mutations to residues in the peptide that I see don’t interact with the protein/don’t fit well in the binding pocket of the protein. I run the mutated peptide through the haddock webserver with the protein again and analyze the new results and repeat this process. This is an iterative process so I feel like there should be a way to automate this however I am not sure how and I am wondering if there is anyone that knows an easier way to do this.

Additionally, I am not sure that my approach is the best since I know that HADDOCK score doesn’t represent binding affinity (I know that prodigy does but through reading the forms I have also learned prodigy is not validated for peptides). I am not sure if there is any way to compare binding of several different peptides to the same protein in one run, so I am open to new ways of optimizing my peptide if you have any ideas.

Thank you in advance!