Predicting clinical outcome of Escherichia coli O157:H7 infections using explainable machine learning
Microbial Genomics · 2025
DOI: 10.1099/mgen.0.001591 ↗Kurz gefasst
Fachartikel in der Zeitschrift Microbial Genomics, 2025. Verfasst von Julian A Paganini und 7 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Veterinary Medicine, Institute for Risk Assessment Sciences (IRAS), Utrecht University, Utrecht, 3584 CL, the Netherlands. Hauptthemen laut MEDLINE: Escherichia coli O157, Escherichia coli Infections, Machine Learning. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 188 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Health~Holland, Biotechnology and Biological Sciences Research Council.
Abstract (Originalsprache)
Shiga toxin-producing Escherichia coli (STEC) O157:H7 is a globally dispersed zoonotic pathogen capable of causing severe disease outcomes, including bloody diarrhoea and haemolytic uraemic syndrome. While variations in Shiga toxin subtype are well-recognized drivers of disease severity, many unexplained differences remain among strains carrying the same toxin profile. We applied explainable machine learning (ML) approaches – Random Forest and Extreme Gradient Boosting – to whole-genome sequencing data from 1,030 STEC O157:H7 isolates to predict patient clinical outcomes, using data collected over 2 years of routine surveillance in England. A phylogeny-informed cross-validation strategy was implemented to account for population structure and avoid data leakage, ensuring robust model generalizability. Extreme Gradient Boosting outperformed Random Forest in predicting minority classes and correctly predicted high-risk isolates in traditionally low-risk lineages, illustrating its utility for capturing complex genomic signatures beyond known virulence genes. Feature importance analyses highlighted phage-encoded elements, including potentially novel intergenic regulators, alongside established virulence factors. Moreover, key genomic regions linked to small RNAs and stress-response pathways were enriched in isolates causing severe disease. These findings underscore the capacity of explainable ML to refine risk assessments, offering a valuable tool for early detection of high-risk STEC O157:H7 and guiding targeted public health interventions.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Faculty of Veterinary Medicine, Institute for Risk Assessment Sciences (IRAS), Utrecht University, Utrecht, 3584 CL, the Netherlands.
- Division of Biosciences, Institute of Structural and Molecular Biology, University College London, London, WC1E 6BT, UK.
- Division of Bacteriology, The Roslin Institute and Royal (Dick) School of Veterinary Studies, University of Edinburgh, Edinburgh, UK.
- University of Bath, Biology and Biochemistry, Bath, BA2 7AY, UK.
- United Kingdom Health Security Agency, 61 Colindale Avenue, London, NW9 5EQ, UK.
Förderung
- Health~Holland
- Biotechnology and Biological Sciences Research Council
Zitieren
APA
Paganini, J. A., Khatun, S., McAteer, S., Cowley, L., Greig, D. R., Gally, D. L., Jenkins, C., & Dallman, T. J. (2025). Predicting clinical outcome of Escherichia coli O157:H7 infections using explainable machine learning. Microbial Genomics. https://doi.org/10.1099/mgen.0.001591
Vancouver
Paganini JA, Khatun S, McAteer S, Cowley L, Greig DR, Gally DL, et al. Predicting clinical outcome of Escherichia coli O157:H7 infections using explainable machine learning. Microbial Genomics. 2025. doi:10.1099/mgen.0.001591
BibTeX
@article{Paganini2025,
author = {Paganini, Julian A and Khatun, Suniya and McAteer, Sean and Cowley, Lauren and Greig, David R and Gally, David L and Jenkins, Claire and Dallman, Timothy J},
title = {Predicting clinical outcome of Escherichia coli O157:H7 infections using explainable machine learning},
journal = {Microbial Genomics},
year = {2025},
doi = {10.1099/mgen.0.001591}
}
EUVIMED