Identification of key drivers of antimicrobial resistance in Enterococcus using machine learning feature ranking (submitted to Canadian Journal of Microbiology)
Files include:
- rast_pangenome_ml_feature.ipynb
- machine learning and feature selection (Block HSIC Lasso & model-dependent feature importance ranking) analysis with pangenome annotated via RASTtk
- the other pangenome features follow the same script, except the pangenome binary matrix that was resolved via respective annotation software (e.g., Prokka or Bakta)
- plasmid_cluster_ML.ipynb
- machine learning with predicted plasmid clusters (from MOB-suite software) encoded as presence and absence binary matrices