Skip to main content

ORIGINAL RESEARCH article

Front. Microbiol.
Sec. Infectious Agents and Disease
Volume 15 - 2024 | doi: 10.3389/fmicb.2024.1450804
This article is part of the Research Topic Conference Research Topic: 9th Symposium on Antimicrobial Resistance in Animals and the Environment (ARAE 2023) View all 20 articles

Bioinformatic Analysis Reveals the Association between Bacterial Morphology and Antibiotic Resistance using Light Microscopy with Deep Learning

Provisionally accepted
  • 1 Division of Special Projects, Osaka University, Suita, Japan
  • 2 Tottori University, Tottori, Tottori, Japan
  • 3 Osaka University, Suita, Ōsaka, Japan
  • 4 The University of Tokyo, Bunkyo, Tōkyō, Japan

The final, formatted version of the article will be published soon.

    Although it is well known that the morphology of Gram-negative rods changes on exposure to antibiotics, the morphology of antibiotic-resistant bacteria in the absence of antibiotics has not been widely investigated. Here, we studied the morphologies of 10 antibiotic-resistant strains of Escherichia coli and used bioinformatics tools to classify the resistant cells under light microscopy in the absence of antibiotics. The antibioticresistant strains showed differences in morphology from the sensitive parental strain, and the differences were most prominent in the quinolone-and β-lactam-resistant bacteria. A cluster analysis revealed increased proportions of fatter or shorter cells in the antibiotic-resistant strains. A correlation analysis of morphological features and gene expression suggested that genes related to energy metabolism and antibiotic resistance were highly correlated with the morphological characteristics of the resistant strains. Our newly proposed deep learning method for single-cell classification achieved a high level of performance in classifying quinolone-and β-lactam-resistant strains.

    Keywords: antibiotic resistance, light microscopy, Bacterial morphology, deep learning, Bioinformatic analysis

    Received: 18 Jun 2024; Accepted: 19 Aug 2024.

    Copyright: © 2024 Ikebe, Aoki, Hayashi-Nishino, Furusawa and Nishino. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence:
    Kota Aoki, Tottori University, Tottori, 680-8550, Tottori, Japan
    Mitsuko Hayashi-Nishino, Osaka University, Suita, 565-0871, Ōsaka, Japan
    Kunihiko Nishino, Division of Special Projects, Osaka University, Suita, 567-0047, Japan

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.