Machine learning algorithm to characterize antimicrobial resistance associated with the International Space Station surface microbiome

This research paper, led by Kasthuri Venkateswaran and Afshin Beheshti from NASA's European Molecular Biology Laboratory (EMBL-EBI), focuses on the growing problem of antibiotic resistance. Over the past thirty years, many people have used antibiotics to fight infections, but this has led bacteria to become resistant to these drugs at an alarming rate. The study aims to understand how quickly and where major types of antibiotics are disappearing from our environment due to resistance. To do so, the researchers collected samples containing millions of tiny DNA fragments (metagenomic reads) which were then carefully cleaned up using specialized software tools like Trimmomatic.

The research paper by Kasthuri Venkateswaran et al., published in an Open Access journal, investigates antibiotic resistance through a metagenomic approach using paired-end reads from the NCBI Short Read Archive (SRA). The study's methodology involves processing these raw sequencing data with Trimmomatic to remove adapter sequences and low-quality ends based on Phred scores. Only high-quality, 100 bp long metagenomic reads exceeding a minimum quality score of 20 are retained for further analysis. The key findings from the study reveal that antibiotic resistance genes (ARGs) have been widely disseminated in various environments due to extensive use and slow discovery rates over recent decades, as highlighted by the World Health Organization. The researchers employed a comprehensive approach using metagenomic sequencing data from diverse samples collected across different locations worldwide. The study's scientific implications are significant for understanding antibiotic resistance patterns in various environments and developing strategies to combat this global health threat.