Supplemental Information and Analysis

This page will house information on supplemental notebooks in the repository

1. Constructing the Cancer Subnetwork
   • In this notebook, we demonstrate how to use various collections of cancer-related genes to construct a cancer-specific subnetwork that can be used in the pyNBS algorithm to stratify patients with sparse mutational profiles.
   • This notebook uses the gene_conversion_tools module to convert gene names from these various collections to a unified gene naming system.
   • Data used in this notebook can be found in ~/Supplementary_Notebooks/Supplementary_Notebook_Data/CancerSubnetwork_Data/
2. pyNBS algorithm analysis
   • In this notebook, we use the TCGA Bladder Cancer (BLCA) cohort as the example to explore how various decisions to use different underlying data or choosing not to execute certain parts of the pyNBS algorithm will affect the final clustering results.
   • Data used in this notebook can be found in Supplementary_Notebook_Data. It is the same data that is found in the ~/Examples/Example_Data/ folder for BLCA.
3. pyNBS Stochasticity and Parameter Benchmarking
   • In this notebook, we explore the the stochasticity of pyNBS and the effect of changing alpha and k on the survival log-rank p-value.
   • The clustering results used to construct figures in this notebook can be found in Supplementary_Notebook_Results/Benchmark_Results.
   • This notebook can be easily adapted to study the effect of changing these parameters on different cancer datasets.