A pattern-based search engine was developed and implemented to efficiently find the helical peptides containing the pattern in our TP-DB database without relying on sequence homology. To allow further evaluation of the selected peptides, two criteria such as helical propensity (HP) and contact number were used to assess the likelihood of a peptide staying in helical form when expressed or synthesized alone. 

Using MD simulations, we found a physicochemical property of antimicrobial peptides (AMP) such as the presence of equally spaced and positively-charged amino acids (e.g., tryptophan/W) could be a factor in its membrane insertion ability and low cytotoxicity. 

We then treated this physicochemical property as the template pattern (e.g., W - - W - - W; where “-” means the other 20 amino acids) and demonstrated our search engine’s functionality in identifying another helical peptide, W3_db5, having improved antibacterial (E. coli) and antifungal (C. albicans) activities as compared to their template peptides having the same pattern. In considering the peptide’s cytotoxicity (ability to lyse/damage cells other than the microorganism), we found that it is 20-fold less cytotoxic as compared to its template peptides. These results could help in developing a more effective treatment for such kinds of infections without implicating any harmful effects on humans.  

It has been a routine to use alanine for point mutation(s) having a reason that this could still maintain the peptide’s helicity. Here, we have demonstrated that this can also be achieved using our pattern-based search engine by introducing point mutation(s) besides alanine in developing protein-protein interaction blocker peptides as an antitumor therapy.