Pre-trained molecular representations enable antimicrobial discovery

Roberto Olayo-Alarcón, Martin K. Amstalden, Annamaria Zannoni, Medina Bajramovic, Cynthia M. Sharma, Ana Rita Brochado, Mina Rezaei, Christian L. Müller

Abstract

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 is still a time-consuming and costly endeavor. Compound-centric deep learning models hold the promise to speed up this search and prioritization process. Here, we introduce a lightweight computational strategy for antimicrobial discovery that builds on MolE ( Mol ecular representation through redundancy reduced E mbedding), a deep learning framework that leverages unlabeled chemical structures to learn task-independent molecular representations. By combining MolE representation learning with experimentally validated compound-bacteria activity data, we design a general predictive model that enables assessing compounds with respect to their antimicrobial potential. The model correctly identified recent growth-inhibitory compounds that are structurally distinct from current antibiotics and discovered de novo three human-targeted drugs as Staphylococcus aureus growth inhibitors which we experimentally confirmed. Our framework offers a viable cost-effective strategy to accelerate antibiotics discovery.
BibTeX
@inproceedings{olayoalarcon2025pre,
  author = {Roberto Olayo-Alarcón and Martin K. Amstalden and Annamaria Zannoni and Medina Bajramovic and Cynthia M. Sharma and Ana Rita Brochado and Mina Rezaei and Christian L. Müller},
  title = {{Pre-trained molecular representations enable antimicrobial discovery}},
  booktitle = {Nature Communications},
  year = {2025},
  doi = {10.1038/s41467-025-58804-4},
}

Loading related content...

Figure