Accelerated antimicrobial discovery via deep generative models and molecular dynamics simulations

• Published in: Nature biomedical engineering Vol. 5; no. 6; pp. 613 - 623
• Main Authors: Das, Payel, Sercu, Tom, Wadhawan, Kahini, Padhi, Inkit, Gehrmann, Sebastian, Cipcigan, Flaviu, Chenthamarakshan, Vijil, Strobelt, Hendrik, dos Santos, Cicero, Chen, Pin-Yu, Yang, Yi Yan, Tan, Jeremy P. K., Hedrick, James, Crain, Jason, Mojsilovic, Aleksandra
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.06.2021; Nature Publishing Group
• Subjects: 14/19; 14/34; 631/114/1305; 631/326/22/1290; 631/45/611; 631/57/2266; 96/63; Acinetobacter baumannii - drug effects; Acinetobacter baumannii - growth & development; Acinetobacter baumannii - ultrastructure; Amino Acid Sequence; Animals; Anti-Bacterial Agents - chemical synthesis; Anti-Bacterial Agents - pharmacology; Antiinfectives and antibacterials; Antimicrobial agents; Antimicrobial Cationic Peptides - chemical synthesis; Antimicrobial Cationic Peptides - pharmacology; Antimicrobial peptides; Biomedical and Life Sciences; Biomedical Engineering/Biotechnology; Biomedicine; Chemical compounds; Classifiers; Computer applications; Deep Learning; Drug Design; Drug Discovery - methods; Drug Resistance, Bacterial - drug effects; E coli; Escherichia coli - drug effects; Escherichia coli - growth & development; Escherichia coli - ultrastructure; Female; Klebsiella; Klebsiella Infections - drug therapy; Klebsiella pneumoniae - drug effects; Klebsiella pneumoniae - growth & development; Klebsiella pneumoniae - ultrastructure; Mice; Mice, Inbred BALB C; Microbial Sensitivity Tests; Molecular dynamics; Molecular Dynamics Simulation; Molecular modelling; Multidrug resistance; Pathogens; Peptides; Pseudomonas aeruginosa - drug effects; Pseudomonas aeruginosa - growth & development; Pseudomonas aeruginosa - ultrastructure; Simulation; Staphylococcus aureus - drug effects; Staphylococcus aureus - growth & development; Staphylococcus aureus - ultrastructure; Structure-Activity Relationship; Toxicity
• ISSN: 2157-846X; 2157-846X
• DOI: 10.1038/s41551-021-00689-x

The de novo design of antimicrobial therapeutics involves the exploration of a vast chemical repertoire to find compounds with broad-spectrum potency and low toxicity. Here, we report an efficient computational method for the generation of antimicrobials with desired attributes. The method leverages guidance from classifiers trained on an informative latent space of molecules modelled using a deep generative autoencoder, and screens the generated molecules using deep-learning classifiers as well as physicochemical features derived from high-throughput molecular dynamics simulations. Within 48 days, we identified, synthesized and experimentally tested 20 candidate antimicrobial peptides, of which two displayed high potency against diverse Gram-positive and Gram-negative pathogens (including multidrug-resistant Klebsiella pneumoniae ) and a low propensity to induce drug resistance in Escherichia coli . Both peptides have low toxicity, as validated in vitro and in mice. We also show using live-cell confocal imaging that the bactericidal mode of action of the peptides involves the formation of membrane pores. The combination of deep learning and molecular dynamics may accelerate the discovery of potent and selective broad-spectrum antimicrobials. A computational method leveraging deep learning and molecular dynamics simulations enables the rapid discovery of antimicrobial peptides with low toxicity and with high potency against diverse Gram-positive and Gram-negative pathogens.