Brenna Kizer and WMRUG
Reproducible research encapsulates all aspects of the analytics process such that given the same data source, anyone anywhere can produce the same results. This eliminates mistakes by automating what would have otherwise required manual steps, reduces cognitive load on analysts, and helps create more modular code so that projects can expand without becoming more complex. Recording the development process in sufficient detail eliminates production lag resulting from complex, undocumented code. We will discuss our use of good coding practice in the R statistical programming language, version control through Git, and RMarkdown as a method of documentation. As students and interns in the realms of biostatistics, data science, and bioinformatics, we are committed to guaranteeing reproducibility in the code we generate for our projects. These practices are highly valuable to any organization wishing to amplify their agile approach to business problems.