Machine Learning Identifies Antibiotic-Resistant Genes in Sepsis-Causing Strains

Strategic Insights -

Researchers from the JSS Academy of Higher Education and Research in India have employed machine learning algorithms to identify antibiotic-resistant genes and virulence factors in Escherichia coli strains causing sepsis. The study used whole-genome sequencing and bioinformatic tools to analyze 18 sepsis-causing strains and identified various global and emerging multidrug resistance (MDR) sequence types. The findings highlight the importance of prompt surveillance, robust infection control, optimized antibiotic stewardship, and integrated genomic and epidemiological analysis to control MDR bacteria transmission and improve patient outcomes.

Key Takeaways:

  • The study employed machine learning algorithms to identify antibiotic-resistant genes and virulence factors in Escherichia coli strains causing sepsis.
  • Whole-genome sequencing of 18 sepsis-causing strains was performed to identify MDR and virulence factor genes.
  • The researchers identified various global and emerging MDR sequence types using a supervised machine learning approach.
  • The study found that the combination of machine learning algorithms and genomic data can facilitate the discovery of nonlinear interactions and complex patterns within genomic data.
  • The researchers correlated known AMR genes with resistance phenotypes and identified several crucial and novel AMR genes.
  • The feature selection methodology involved processing the genome into overlapping 13 bp k-mer features using a two-step selection process.
  • Logistic regression with nested cross-validation and synthetic minority oversampling technique confirmed the robustness of the model.
  • The study highlights the importance of prompt surveillance, robust infection control, optimized antibiotic stewardship, and integrated genomic and epidemiological analysis to control MDR bacteria transmission and improve patient outcomes.

Statistics:

  • 18 sepsis-causing strains were analyzed in the study.
  • 13 bp k-mer features were used in the feature selection methodology.
  • 2-step selection process was employed for feature selection.
  • 17 antimicrobial classes were used in the correlation analysis.
  • 17 AMR genes were correlated with resistance phenotypes.

Sources:

  • "Whole-genome sequencing and bioinformatic tools powered by machine learning to identify antibiotic-resistant genes and virulence factors in Escherichia coli from sepsis." Microbial Genomics, 2025;11(8).
  • "Findings on Bioinformatics Reported by Researchers at JSS Academy of Higher Education and Research (Whole-genome sequencing and bioinformatic tools powered by machine learning to identify antibiotic-resistant genes and virulence factors in ...)." Biotech Week, August 27, 2025; p 203.