paprpkgr

The goal of paprpkgr is to generate a skeleton to develop a research compendium containing the data, analysis, code, and images to reproduce one manuscript. I began working on this package around the same time that the more sophisticated rrtools package was being developed by Ben Marwick and colleagues. If you are looking to incorporate Docker or Travis in your workflows, that package will definitely be more useful to you.

Installation

You can install paprpkgr from GitHub with either of the following:

devtools::install_github("wmoldham/paprpkgr")
renv::install("wmoldham/paprpkgr")

Getting Started

1. Create a new package using usethis::create_package().
2. Start version control with Git by usethis::use_git().
3. Link version control to GitHub by usethis::use_github().
4. Restart the session when prompted.
5. Run paprpkgr::new_papr().

Compendium Development

1. Edit README.Rmd and DESCRIPTION files with basic information about your project.
2. Original data goes into the data-raw folder.
3. Analyze the raw data and generate cleaned data for analysis using Rmarkdown files in analysis/. Figures from these analyses should be saved in analysis/figures/.
4. The clean data should be saved by usethis::use_data() and documented appropriately.
5. Draft the manuscript and supplement using the Rmd templates installed in manususcript/.
6. Build processing pipeline using the targets package.

Research Compendium Resources

Many others have thought much more deeply about the hows and whys of using R packages in service of reproducible research efforts. I took my inspiration from their efforts:

• Ben Marwick and his rrtools package
• Francisco Rodriguez-Sanchez and his template package
• Carl Boettiger and his template package
• Jeff Hollister and his manuscriptPackage
• Robert Flight: http://rmflight.github.io/posts/2014/07/analyses_as_packages.html
• https://github.com/ropensci/rrrpkg
• https://github.com/Reproducible-Science-Curriculum/rr-init
• http://ropensci.github.io/reproducibility-guide/