This repository contains R scripts for risk score analysis and annotation clustering in An et al. (2018). The raw data are not included in this repository. For data generation, please refer to https://github.com/sanderslab/WGS-pipeline.
- Risk score analysis
- Compute the predictive R2 of rare annotations:
risk_score/predictive_r2.R
- Compute the p-value:
risk_score/significance_r2.R
- Compute the predictive R2 of rare annotations:
- Annotation clustering
- Spectral clustering of annotation categories:
clustering/clustering.R
- Spectral clustering of annotation categories:
- Network analysis
- Forming the graph and determining the significance of each cluster of nodes:
network/network.R
- Because many custom functions are needed for this step, a supplemental R package is located in
network/supernodeWGS
to encapsulate all the needed functions
- Forming the graph and determining the significance of each cluster of nodes: