This repository has been developed to reproduce the results in McCurdy (2018, http://arxiv.org/abs/1803.06010). We implement the deterministic ridge leverage sampling algorithm that comes with a
To download the TCGA lower-grade glioma (LGG) tumor multi-omic data collected by the TCGA Research Network, you will first need to install CNTools (Zhang, 2015) and TCGA2STAT (Wan et al., 2016) in R. To install the R packages, open an R environment and type the following:
## try http:// if https:// URLs are not supported
source("https://bioconductor.org/biocLite.R")
biocLite("CNTools")
install.packages("TCGA2STAT_1.0.tar.gz", repos = NULL, type = "source")
The rest of this code repository is in python. We used the anaconda package manager which can be obtained at https://www.continuum.io/downloads, and we include a .yml file (ridge.yml) that we use for the analysis. You can simply create an "conda env" with the .yml file by
conda env create -f ridge.yml
#
# To activate this environment, use:
# $ source activate ridge
#
# To deactivate this environment, use:
# $ source deactivate
#
In order to run the code, first clone this repository.
git clone https://github.com/srmcc/determinstic-ridge-leverage-sampling.git
Then activate your conda environment:
source activate ridge
Change directory to the cloned repository:
cd /path/to/determinstic-ridge-leverage-sampling
And run the analysis file.
python ridge.py