This page contains R vignettes showing use cases for netDx.

  1. Run 10-fold cross validation for binary classification: This vignette shows the building blocks for feature selection and sample classification. Two types of input data are used, gene expression and copy number variants, the latter being an extremely sparse form of genomic data.
  2. Perform 4-way classification of a tumour with gene-level networks: This vignette shows an example of classifying a medulloblastoma tumour as one of four known subtypes based on its gene signature profile.
  3. Run nested cross-validation for binary classification. This R notebook shows you how to run nested cross-validation with multiple datatypes and similarity metrics, with one simple function call.
  4. Plot predictor results. This R notebook shows you how to visualize the results of a predictor run, including Cytoscape-driven network visualizations of selected features, of patient similarity.

.R scripts to run the vignettes are available in the netDx github repo. These files are currently limited to private access and will be made public upon publication of the netDx methods paper.