Last updated: 2018-12-20

workflowr checks: (Click a bullet for more information)
Expand here to see past versions:


This repository contains data and R source code accompanying our manuscript,

Zhengrong Xing and Matthew Stephens (2016). Smoothing via Adaptive Shrinkage (smash): denoising Poisson and heteroskedastic Gaussian signals. arXiv:1709.10066.

If you find any of the source code in this repository useful for your work, please cite our paper.

The new methods can be found in the smashr package.

Contents

The following analyses generate some of the empirical results presented in Xing & Stephens (2016). If you encounter a problem running any of the R code in these examples, please post an issue.

Illustrative applications

  1. Motorcycle acceleration data.

  2. ChIP-seq data.

What’s included in the git repository

See here for the source repository. This is the overall structure of the repository:

├── analysis
├── code
├── data
├── docs
├── dsc
├── output
└── shiny

License

Copyright (c) 2016-2018, Zhengrong Xing, Peter Carbonetto & Matthew Stephens.

Our numerical comparisons make use of some of the functions from WaveLab, so we have included the WaveLab source code in this repository. See the COPYING.m Wavelab850 subdirectory for more information about distributing WaveLab.

Our numerical comparisons also use of some functions from GaussianWaveDen, so we have included the GaussianWaveDen source in this repository. For information about distributing GaussianWaveDen, see the copyright.m in the WavDen subdirectory. Note that we made one small improvement to the code in blockJS.m to prevent an error that occurs when running the code in newer versions of MATLAB.

All other source code and software in this repository are made available under the terms of the MIT license. See the LICENSE file in the git repository for the full text of the license.

Credits

This project was developed by Zhengrong Xing at the University of Chicago, with support and contributions from Peter Carbonetto and Matthew Stephens.


This reproducible R Markdown analysis was created with workflowr 1.1.1