Pre-trained molecular representations enable antimicrobial discovery
Top Cited Papers
Open Access
- 10 April 2025
- journal article
- research article
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-15
Abstract
The rise in antimicrobial resistance poses a worldwide threat, reducing the efficacy of common antibiotics. Determining the antimicrobial activity of new chemical compounds through experimental methods remains time-consuming and costly. While compound-centric deep learning models promise to accelerate this search and prioritization process, current strategies require large amounts of custom training data. Here, we introduce a lightweight computational strategy for antimicrobial discovery that builds on MolE (Molecular representation through redundancy reduced Embedding), a self-supervised deep learning framework that leverages unlabeled chemical structures to learn task-independent molecular representations. By combining MolE representation learning with available, experimentally validated compound-bacteria activity data, we design a general predictive model that enables assessing compounds with respect to their antimicrobial potential. Our model correctly identifies recent growth-inhibitory compounds that are structurally distinct from current antibiotics. Using this approach, we discover de novo, and experimentally confirm, three human-targeted drugs as growth inhibitors of Staphylococcus aureus. This framework offers a viable, cost-effective strategy to accelerate antibiotic discovery.Keywords
All Related Versions
Funding Information
- This work was funded by a grant for the "StressRegNet" consortium within the Bavarian research network bayresq.net funded through the Bavarian State Ministry of Science and Arts, Germany
This publication has 45 references indexed in Scilit:
- Extensive impact of non-antibiotic drugs on human gut bacteriaNature, 2018
- MoleculeNet: a benchmark for molecular machine learningChemical Science, 2017
- In Vitro Screening of an FDA-Approved Library Against ESKAPE PathogensCurrent Pharmaceutical Design, 2017
- Mitochondria-targeted antioxidants as highly effective antibioticsScientific Reports, 2017
- Predicting antimicrobial peptides with improved accuracy by incorporating the compositional, physico-chemical and structural features into Chou’s general PseAACScientific Reports, 2017
- Molecular graph convolutions: moving beyond fingerprintsJournal of Computer-Aided Molecular Design, 2016
- CAMPR3: a database on sequences, structures and signatures of antimicrobial peptides: Table 1.Nucleic Acids Research, 2015
- ESKAPEing the labyrinth of antibacterial discoveryNature Reviews Drug Discovery, 2015
- Extended-Connectivity FingerprintsJournal of Chemical Information and Modeling, 2010
- The Properties of Known Drugs. 1. Molecular FrameworksJournal of Medicinal Chemistry, 1996