Designing antimicrobial peptides using deep learning and molecular dynamic simulations

With the emergence of multidrug-resistant bacteria, antimicrobial peptides (AMPs) offer promising options for replacing traditional antibiotics to treat bacterial infections, but discovering and designing AMPs using traditional methods is a time-consuming and costly process. Deep learning has been applied to the de novo design of AMPs and address AMP classification with high efficiency. In this study, several natural language processing models were combined to design and identify AMPs, i.e. sequence generative adversarial nets, bidirectional encoder representations from transformers and multilayer perceptron. Then, six candidate AMPs were screened by AlphaFold2 structure prediction and molecular dynamic simulations. These peptides show low homology with known AMPs and belong to a novel class of AMPs. After initial bioactivity testing, one of the peptides, A-222, showed inhibition against gram-positive and gram-negative bacteria. The structural analysis of this novel peptide A-222 obtained by nuclear magnetic resonance confirmed the presence of an alpha-helix, which was consistent with the results predicted by AlphaFold2. We then performed a structure-activity relationship study to design a new series of peptide analogs and found that the activities of these analogs could be increased by 4-8-fold against Stenotrophomonas maltophilia WH 006 and Pseudomonas aeruginosa PAO1. Overall, deep learning shows great potential in accelerating the discovery of novel AMPs and holds promise as an important tool for developing novel AMPs.

Automated summary

With the rise of multidrug-resistant bacteria, antimicrobial peptides (AMPs) have emerged as potential alternatives to traditional antibiotics for treating bacterial infections. However, traditional methods of discovering and designing AMPs are time-consuming and costly. This study utilized deep learning techniques, including sequence generative adversarial nets, bidirectional encoder representations from transformers, and multilayer perceptron, to design and identify AMPs. Six candidate AMPs were then screened and one of them, A-222, showed inhibition against both gram-positive and gram-negative bacteria. Structural analysis and subsequent structure-activity relationship studies led to the design of peptide analogs with increased activity against specific bacteria. Overall, deep learning holds great promise in accelerating the discovery of novel AMPs and could have significant implications in developing new antimicrobial treatments.