A deep learning approach to antibiotic discovery

By Jonathan M. Stokes, Kevin Yang, Kyle Swanson, Wengong Jin, Andres Cubillos-Ruiz, Nina M. Donghia, C. MacNair, S. French, L. Carfrae, Zohar Bloom-Ackermann and 10 more. Cell, 2020

1,050 citations recorded by the source corpus.

SUMMARY Due to the rapid emergence of antibiotic-resistant bacteria, there is a growing need to discover new antibiotics. To address this challenge, we trained a deep neural network capable of predicting molecules with antibacterial activity. We performed predictions on multiple chemical libraries and discovered a molecule from the Drug Repurposing Hub – halicin – that is structurally divergent from conventional antibiotics and displays bactericidal activity against a wide phylogenetic spectrum of pathogens, including Mycobacterium tuberculosis and carbapenem-resistant Enterobacteriaceae. Halicin also effectively treated Clostridioides difficile and pan-resistant Acinetobacter baumannii infections in murine models. Additionally, from a discrete set of 23 empirically tested predictions from >107 million molecules curated from the ZINC15 database, our model identified eight antibacterial compounds that are structurally distant from known antibiotics. This work highlights the utility of deep learning approaches to expand our antibiotic arsenal through the discovery of structurally distinct antibacterial molecules.

Subfields: Clinical Biochemistry, Computational Theory and Mathematics, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/6ab228a541cfc77aedf3ebbc9dc9283b4acb1c82

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

Citations in the corpus

Empowering Biomedical Discovery with AI Agents (2024, neutral): “These models have advanced modeling of proteins [22–26], genes [27], phenotypes [28], clinical outcomes [29–31], and chemical compounds [32, 33] through mining of biomedical data.”
Discovery of a structural class of antibiotics with explainable deep learning (2023, methodology): “No statistical method was used to predetermine sample size for all mouse experiments in this study, but our sample sizes are similar to those reported in previous publications 1–4 , 6– 8 , 14 .”
Open Graph Benchmark: Datasets for Machine Learning on Graphs (2020, background): “; Welling, 2017; Ying et al., 2018b; Veliˇckovi c´ et al., 2019; Xu et al., 2019), which have produced promising results in applications from diverse domains (Ying et al., 2018a; Zitnik et al., 2018; Stokes et al., 2020). How can we further advance research in graph ML? Historically, high-quality an”
ADMETlab 3.0: an updated comprehensive online ADMET prediction platform enhanced with broader coverage, improved performance, API functionality and decision support (2024, positive): “Presently, DMPNN methods have been successfully applied in various fields of drug discovery, including antibiotic discovery ( 28–30 ), reaction property prediction ( 31 ), infrared spectra prediction ( 32 ), solute parameters prediction ( 14 ), and molecular optical peak prediction ( 15 ).”
Structure-Based Virtual Screening: From Classical to Artificial Intelligence (2020, background): “However, some researchers applied machine learning techniques to discover a new antibiotic capable of inhibiting the growth of E. coli bacteria (Stokes et al., 2020).”
Identification of antimicrobial peptides from the human gut microbiome using deep learning (2022, background): “The recent surge in applying artificial intelligence approaches in bio-medical research has brought up many new opportunities, including image-processing-related tools for clinical diagnostics of eye 39 , lung 40 , cancer tissue pathologies 41 and antibiotics 19,42 .”
Artificial intelligence in drug development (2025, neutral): “Phenotype-based virtual screening is thus crucial for diseases with undefined targets (for example, rare diseases) and broadly phenotypic diseases (for example, aging) 66–72 .”
Graph Representation Learning in Biomedicine and Healthcare (2021, neutral): “Simplistically, GNNs aggregate information from neighbouring atoms and bonds to learn the local chemistry of each atom 153 .”
Machine Learning Methods in Drug Discovery (2020, background): “This antibacterial candidate was also discovered to be a potent growth inhibitor of Escherichia coli , and had shown e ﬃ cacy against Clostridioides di ﬃ cile and Acinetobacter baumannii infections in murine models [95].”
How Machine Learning will Transform Biomedicine (2020, neutral): “…and treatment have shown promise to diagnosis breast cancer from x-rays (McKinney et al., 2020; Wu et al., 2019), discover new antibiotics (Stokes et al., 2020), predict onset of gestational diabetes from electronic health records (Artzi et al., 2020), and identify clusters of patients…”
Artificial intelligence to deep learning: machine intelligence approach for drug discovery (2021, result): “Moreover, in comparison with AI and ML algorithms, DL provides a much more flexible architecture to create a neural network for a specific problem [477–480].”
The rise of artificial intelligence in healthcare applications (2020, background): “…for antibiotic discovery, where the researchers devised/ trained a neural network that actively “ learned ” the properties of a vast number of molecules in order to identify those that inhibit the growth of E. coli , a Gram negative bacterial species that is notoriously hard to kill [66].”
The Medical Segmentation Decathlon (2021, background): “mance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the”