Related Experiment Video
Updated: Aug 31, 2025
08:58
Isolation and Identification of Waterborne Antibiotic-Resistant Bacteria and Molecular Characterization of their Antibiotic Resistance Genes
Published on: March 3, 2023
6.7K
Machine learning algorithm to characterize antimicrobial resistance associated with the International Space Station
Pedro Madrigal1,2, Nitin K Singh3, Jason M Wood3
1Jeffrey Cheah Biomedical Centre, Wellcome-MRC Cambridge Stem Cell Institute, University of Cambridge, Cambridge Biomedical Campus, Puddicombe Way, Cambridge, CB2 0AW, UK. pmadrigal@ebi.ac.uk.
Microbiome
|August 23, 2022
Summary
Antimicrobial resistance (AMR) genes were identified in space habitats using deep learning. Kalamiella piersonii showed AMR dominance, and Enterobacter bugandensis and Bacillus cereus were confirmed as resistant pathogens.
Area of Science:
- Microbiology
- Space Science
- Genomics
Background:
- Antimicrobial resistance (AMR) poses a significant threat to human health, both on Earth and in space environments like the International Space Station (ISS).
- Spaceflight conditions, including microgravity and radiation, can alter microbial virulence and resistance.
- Investigating microbial diversity and AMR in the ISS is crucial for astronaut health during long-duration missions.
Purpose of the Study:
- To identify AMR genes in viable microbes from ISS environmental samples.
- To analyze shotgun metagenomics data and assembled genomes from the Microbial Tracking-1 (MT-1) project.
- To explore the pathogenic potential of ISS microbes using advanced computational methods.
Main Methods:
- Utilized shotgun metagenomics data and metagenome-assembled genomes (MAGs) from ISS environmental samples.
- Applied a deep learning model to analyze AMR genes beyond traditional sequence similarity cut-offs.
- Experimentally validated computational predictions using antibiotic resistance profiling of bacterial isolates.
Main Results:
- Deep learning analysis revealed AMR dominance in Kalamiella piersonii in later ISS flights.
- Hundreds of antibiotic resistance genes were found in isolates, notably in Enterobacter bugandensis and Bacillus cereus.
- Experimental validation confirmed high resistance of E. bugandensis and B. cereus to beta-lactam antibiotics.
Conclusions:
- Machine learning effectively identifies AMR determinants in complex metagenomics datasets.
- This study expands the understanding of ISS microbial communities and their potential human health risks.
- Findings highlight the importance of monitoring AMR in closed environments like space habitats.
