Antimicrobial peptides (AMPs) have been identified as a potentially new class of antibiotics to combat bacterial resistance to conventional drugs. The design of de novo AMPs with high therapeutic indexes, low cost of synthesis, high resistance to proteases and high bioavailability remains a challenge. Such design requires computational modeling of antimicrobial properties. Currently, most computational methods cannot accurately calculate antimicrobial potency against particular strains of bacterial pathogens. We developed a tool for AMP prediction (Special Prediction (SP) tool) and made it available on our Web site (https://dbaasp.org/prediction). Based on this tool, a simple algorithm for the design of de novo AMPs (DSP) was created. We used DSP to design short peptides with high therapeutic indexes against gram-negative bacteria. The predicted peptides have been synthesized and tested in vitro against a panel of gram-negative bacteria, including drug resistant ones. Predicted activity against Escherichia coli ATCC 25922 was experimentally confirmed for 14 out of 15 peptides. Further improvements for designed peptides included the synthesis of D-enantiomers, which are traditionally used to increase resistance against proteases. One synthetic D-peptide (SP15D) possesses one of the lowest values of minimum inhibitory concentration (MIC) among all DBAASP database short peptides at the time of the submission of this article, while being highly stable against proteases and having a high therapeutic index. The mode of anti-bacterial action, assessed by fluorescence microscopy, shows that SP15D acts similarly to cell penetrating peptides. SP15D can be considered a promising candidate for the development of peptide antibiotics. We plan further exploratory studies with the SP tool, aiming at finding peptides which are active against other pathogenic organisms.

Antimicrobial peptides (AMPs) have been identified as a potentially new class of antibiotics. There is a lot of computational methods of AMP prediction. Although most of them can predict antimicrobial potency against any microbe (microbe is not identified) with rather high accuracy, prediction quality of these tools against particular bacterial strains is low (Bioinformatics, 2018, in press; Journal of Chemical Information and Modeling 58, 1141-1151). Special prediction is a tool for the prediction of antimicrobial potency of peptides against particular target species with high accuracy. This tool is included into the Database of Antimicrobial Activity and Structure of Peptides (DBAASP, https://dbaasp.org, Nucleic acids research 44 (D1), D1104-D1112). In this presentation we describe this tool and predictive models for some Gram+ bacterial strains (Staphylococcus aureus ATCC 25923 and Bacillus subtilis) and a model for the prediction of hemolytic activity. Predictive model for Gram- Escherichia coli ATCC 25922 was presented earlier (Journal of Chemical Information and Modeling 58, 1141-1151, In 2nd Int. Electron. Conf. Med. Chem. 01-30 November 2016; Sciforum, 2016, A031). Special prediction tool can be used for the design of peptides being active against particular strain. To demonstrate the capability of the tool, peptides predicted as active against E. coli ATCC 25922 and S. aureus ATCC 25923 have been synthesized, and tested in vitro. The results have shown the justification of using special prediction tool for the design of new AMPs.

Emerging bacterial resistance to the existing antibiotics makes the development of new types of antibiotics an increasingly important challenge. Antimicrobial peptides (AMPs) can be considered as novel and efficient type of antibiotics that are hard to acquire resistance against. We have developed an algorithm to design peptides that are active against certain species. The prediction is based on clusterization of peptides with known biological activities by physicochemical properties. The Database of Antimicrobial Activity and Structure of Peptides (DBAASP, https://dbaasp.org) now includes Special Prediction (SP) tool, which allows to apply this algorithm to any amino acid sequence to predict whether this peptide is active against particular microbes. To verify the efficiency of the algorithm, we designed several variants of active peptides and tested them in vitro against two strains Escherichia coli ATCC 25922 and Staphylococcus aureus 25923. Prediction precision for the designed peptides against Escherichia coli ATCC was 95% and against Staphylococcus aureus was 68%. To improve prediction precision against Staphylococcus aureus we applied the linear regression analysis based on binary classification. This approach allows us to improve the prediction precision of the peptides designed for Staphylococcus aureus 25923 up to 92%.

Antimicrobial peptides (AMPs) are anti-infective agents that may represent a novel and untapped class of biotherapeutics. In the lab of bioinformatics of IBCEB, the Database of Antimicrobial Activity and Structure of Peptides (DBAASPv.2 - accessible at http://dbaasp.org) has been developed. DBAASP provides information and analytical resources to the scientific community in order to develop antimicrobial compounds with high therapeutic index. Quantitative structure-activity relationship (QSAR) studies for the development of predictive model for AMPs are generally based on discriminative analysis and especially machine learning methods. These methods, as a positive training set, have used a full set of antimicrobial peptide sequences, without taking into account variation in mechanisms of action, structure, mode of interaction with membrane and other differences. Contrary to available approaches, we think that strategy of prediction should be based on the fact that there are at least four kinds of AMPs for which four independent algorithms of prediction have to be developed in order to get high efficacy. We can distinguish linear cationic antimicrobial peptides (LCAP), cationic peptides stabilizing structure by intra-chain covalent bonds, proline and arginine-rich peptides, and anionic antimicrobial peptides. Simple predictive model which can discriminate AMPs from non-AMPs has been developed for LCAP. Sequences have been taken from DBAASP. As descriptors the sequence-based physical-chemical characteristics responsible for capability of the peptide to interact with an anionic membrane were considered. On the basis of these characteristics, a new simple algorithm of prediction is developed and in silico evaluation of the efficacy of characteristics is done. The algorithm was based on the clusterization of AMPs by their physicochemical properties. The results show that descriptors relied mainly on hydrophobic and hydrophilic features allow us to predict AMPs with the high accuracy. The developed predictive model was used to create new amino acid sequences. On the basis of designed sequences the peptides have been synthesized. Antimicrobial potency of the new peptides has been evaluated by in vitro testing of peptides activity against different pathogenic bacteria (including drug resistant strains). In vitro estimation shows that the accuracy of th developed predictive model is higher than 90%.