1.0
Sam Ng and Joshua M. Stuart
- Install dependencies
- Download the paradigmshift repository to the desired location
- Run "make" in paradigmshift/ to generate source files; may also require downloading the paradigm-scripts and pathmark-scripts repository (see https://github.com/ucscCancer/paradigm-scripts and https://github.com/ucscCancer/pathmark-scripts)
- Source init files for paradigmshift, pathmark-scripts and paradigm-scripts (init.sh for bash and init.csh for csh)
- Run code on example data in examples/ with "make"
paradigmSHIFT.py [options] paradigm_directory analysis_file
paradigm_directory - directory or zip of the paradigm cohort to perform analysis on
analysis_file - file containing events to run paradigm-shift analysis on
-w work_directory - directory to perform work in
-c config_file - file with user defined arguments
-s include_samples - one column file of samples to analyze
-f include_features - one column file of features to analyze
-n null_size - number of null samples to estimate shift significance
-b batch_size - number of samples in each paradigm run per compute node
-p pathway_file - path to paradigm pathway file
-y - this flag must be enabled when using the publicly available version of paradigm
-z seed - fix random seed
- bin : executables
- examples : LUSC and LUAD inputs for demonstration purposes
Feature requests, comments and requests for clarification should all be sent to the author at [email protected]. I will try to respond quickly to all requests, so feel free to email me!