Cross-Molecular Active Learning for the Discovery of Antimicrobial Polyacrylamides
Williams, S. C.; Greenstein, G.; Liu, X.; Fragasso, A.; Eckman, N.; Prossnitz, A.; Dong, C.; Makar-Limanov, A.; Jacobs-Wagner, C.; Cegelski, L.; Lopez Hernandez, H.; Appel, E.
Show abstract
Antimicrobial resistance poses an urgent and increasing threat to global health. The development of new antimicrobials is crucial. Synthetic copolymers are attractive as a potential solution, because they can be produced at scale and designed to mimic antimicrobial peptides and act as broad-spectrum antimicrobials capable of evading resistance mechanisms. This work leverages a cross-molecular machine learning pipeline, trained on antimicrobial peptides, to develop potent antimicrobial polymers to combat Escherichia coli, which were then synthesized and validated experimentally. One candidate copolymer was further characterized and shown to permeabilize the bacterial membrane, which is associated with decreased resistance. Furthermore, this copolymer demonstrated remarkable synergy in eradicating biofilm-associated E. coli when combined with a first-line clinical drug regimen, reducing the amount needed to eradicate bacteria in biofilms by three orders of magnitude. These results demonstrate promise for potentiating antibacterial activity of currently available antibiotics, treating serious and complicated infections, and combatting antimicrobial resistance.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Genetically-Encoded Discovery of Perfluoroaryl-Macrocycles that Bind to Albumin and Exhibit Extended Circulation in-vivo 94%
- Click-Based Determination of Accumulation of Molecules in Escherichia coli 94%
- AGILE Platform: A Deep Learning-Powered Approach to Accelerate LNP Development for mRNA Delivery 93%
Similar papers in this journal
Similar papers in this journal
- Plug-and-play assembly of biodegradable ionizable lipids for potent mRNA delivery and gene editing in vivo 92%
- Gas modulating microcapsules for spatiotemporal control of hypoxia 91%
- Systematic identification of engineered methionines and oxaziridines for efficient, stable, and site-specific antibody bioconjugation 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.