Git + GitHub As A Platform For Reproducible Research

Overview

This repository sets out the skeleton of an organizational structure used for scientific research. It loosely follows what I have used for several of my research projects and I hope it inspires you to conduct your research in an open, reproducible, and honest manner.

How to Use

To use this template for your research, fork this repository, change the name to something descriptive for your project, and adjust the licensing as you see fit.

To use this repository for your own research, simply click "Use this template" at the top of this page

⚠️ ⚠️ ⚠️ I wouldn't advise forking this repository. As you can only fork a given repository once, there is little utility in forking this repo if you hope to use it again in your future projects ⚠️ ⚠️ ⚠️

Alternatively, you can clone the repository and restructure as necessary. Via the command line,

> git clone git@github.com:gchure/reproducible_research

Layout

The repository is split into seven main directories, many of which have subdirectories. This structure has been designed to be easily navigable by humans and computers alike, allowing for rapid location of specific files and instructions. Within each directory is a README.md file which summarizes the purpose of that directory as well as some examples where necessary. This structure may not be perfect for your intended us and may need to be modified. Each section is briefly described below.

experiments

Where all of the experimental information lives, including any executed code. lives. This includes pipelines, scripts, and figure files.

• processing: Any code used to transform the data into another type should live here. This can include everything from parsing of text data, image segmentation/filtering, or simulations.
• analysis: Any code to to draw conclusions from an experiment or data set. This may include regression, dimensionality reduction, or calculation of various quantities.
• exploratory: A sandbox where you keep a record of your different approaches to transformation, interpretation, cleaning, or generation of data.
• figures: Any code used to generate figures for your finished work, presentations, or for any other use.

data

All raw data collected from your experiments as well as copies of the transformed data from your processing code.

miscellaneous

Files that may not be code, but are important for reproducibility of your findings.

• protocols: A well annotated and general description of your experiments. These protocols should be descriptive enough for someone to follow your experiments independently
• materials: Information regarding the materials used in your experiments or data generation. This could include manufacturer information, records of purity, and/or lot and catalog numbers.
• software details: Information about your computational environment that are necessary for others to execute your code. This includes details about your operating system, software version and required packages.

tests

All test suites for your code. Any custom code you've written should be thoroughly and adequately tested to make sure you know how it is working.

software_module

Custom code you've written that is not executed directly, but is called from files in the code directory. If you've written your code in Python, for example, this can be the root folder for your custom software module or simply house a file with all of your functions.

templates

Files that serve as blank templates that document the procedures taken for each experiment, simulation, or analysis routine.

Required Files

There are some files which I consider to be mandatory for any project.

1. LICENSE: A legal protection of your work. It is important to think deeply about the licensing of your work, and is not a decision to be made lightly. See this useful site for more information about licensing and choosing the correct license for your project.

2. README.md: A descriptive yet succinct description of your research project and information regarding the structure outlined below.

License Information

To the extent possible under law, Griffin Chure has waived all copyright and related or neighboring rights to A template for using git as a platform for reproducible scientific research. This work is published from: United States.