2. Academic Validation
  3. Discovery of a structural class of antibiotics with explainable deep learning

Discovery of a structural class of antibiotics with explainable deep learning

  • Nature. 2024 Feb;626(7997):177-185. doi: 10.1038/s41586-023-06887-8.
Felix Wong # 1 2 3,  Erica J Zheng # 1 4 5,  Jacqueline A Valeri 1 2 5,  Nina M Donghia 5,  Melis N Anahtar 1,  Satotaka Omori 1 3,  Alicia Li 3,  Andres Cubillos-Ruiz 1 2 5,  Aarti Krishnan 1 2,  Wengong Jin 6,  Abigail L Manson 1,  Jens Friedrichs 7,  Ralf Helbig 7,  Behnoush Hajian 8,  Dawid K Fiejtek 8,  Florence F Wagner 8,  Holly H Soutter 8,  Ashlee M Earl 1,  Jonathan M Stokes 1 2 9,  Lars D Renner 7,  James J Collins 10 11 12
Affiliations

Affiliations

  • 1 Infectious Disease and Microbiome Program, Broad Institute of MIT and Harvard, Cambridge, MA, USA.
  • 2 Institute for Medical Engineering and Science and Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA.
  • 3 Integrated Biosciences, San Carlos, CA, USA.
  • 4 Program in Chemical Biology, Harvard University, Cambridge, MA, USA.
  • 5 Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, USA.
  • 6 Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, MA, USA.
  • 7 Leibniz Institute of Polymer Research and the Max Bergmann Center of Biomaterials, Dresden, Germany.
  • 8 Center for the Development of Therapeutics, Broad Institute of MIT and Harvard, Cambridge, MA, USA.
  • 9 Department of Biochemistry and Biomedical Sciences, Michael G. DeGroote Institute for Infectious Disease Research and David Braley Centre for Antibiotic Discovery, McMaster University, Hamilton, Ontario, Canada.
  • 10 Infectious Disease and Microbiome Program, Broad Institute of MIT and Harvard, Cambridge, MA, USA. jimjc@mit.edu.
  • 11 Institute for Medical Engineering and Science and Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA. jimjc@mit.edu.
  • 12 Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, USA. jimjc@mit.edu.
  • # Contributed equally.
PMID: 38123686 DOI: 10.1038/s41586-023-06887-8
Abstract

The discovery of novel structural classes of Antibiotics is urgently needed to address the ongoing Antibiotic resistance crisis1-9. Deep learning approaches have aided in exploring chemical spaces1,10-15; these typically use black box models and do not provide chemical insights. Here we reasoned that the chemical substructures associated with Antibiotic activity learned by neural network models can be identified and used to predict structural classes of Antibiotics. We tested this hypothesis by developing an explainable, substructure-based approach for the efficient, deep learning-guided exploration of chemical spaces. We determined the Antibiotic activities and human cell cytotoxicity profiles of 39,312 compounds and applied ensembles of graph neural networks to predict Antibiotic activity and cytotoxicity for 12,076,365 compounds. Using explainable graph algorithms, we identified substructure-based rationales for compounds with high predicted Antibiotic activity and low predicted cytotoxicity. We empirically tested 283 compounds and found that compounds exhibiting Antibiotic activity against Staphylococcus aureus were enriched in putative structural classes arising from rationales. Of these structural classes of compounds, one is selective against methicillin-resistant S. aureus (MRSA) and vancomycin-resistant enterococci, evades substantial resistance, and reduces Bacterial titres in mouse models of MRSA skin and systemic thigh Infection. Our approach enables the deep learning-guided discovery of structural classes of Antibiotics and demonstrates that machine learning models in drug discovery can be explainable, providing insights into the chemical substructures that underlie selective Antibiotic activity.

Figures
Products