Skip to contents

PERCEPTIONx 0.1.0

  • Initial release of the R implementation of the PERCEPTION approach (predicting personalized drug response from single-cell transcriptomics).
  • Train drug-response models on large-scale cell-line screens (train_models) or load pre-trained models (load_model).
  • Predict response at clone level (predict_drugs) and patient level (predict_patients) with multiple aggregation strategies.
  • Interactive visualizations with hover tooltips (ggiraph) and publication-quality export (600 dpi PNG, PDF, SVG).
  • Shiny web application (run_perception_app) covering the whole pipeline.
  • Vignette “The PERCEPTIONx Pipeline” documents the complete workflow.