Repo_Template

A template for organizing and performing analysis (adapted from https://github.com/PhanstielLab/project-template).

Directory Structure

Prerequisites

• Install Docker Desktop for Windows and Mac.

How to use this repo as a template

1. Create a repository from this template according to these instructions: https://docs.github.com/en/repositories/creating-and-managing-repositories/creating-a-repository-from-a-template

2. Clone the repository to your home computer.

git clone https://github.com/bailey-lab/Repo_Template.git

3. Rename the directory to your project name

mv Repo_Template My_Project
cd My_Project

4. Build docker image (rename with the title of your project) in a terminal window of your project directory.

docker build -t my_project .

5. Run the docker container with the username and password below in a terminal window of your project directory.

docker run --rm -p 8787:8787 -e USER=rstudio -e PASSWORD=yourpassword --volume ${PWD}:/home/rstudio my_project

6. Open a web browser and navigate to localhost:8787. Log in with username rstudio and password yourpassword.

   • When you are logged in, you can run R code in the console or open R scripts in the src directory.
     • 
   • You will run the following commands in the terminal in RStudio.
     • 

7. As I already have the data/raw, config and src directories (with the analysis, processing,utils subdirectories) made, you can start adding your data and scripts to these directories.

8. Run make commands (in the terminal) to inialize the general dictionaries.

make dirs

8. Fill the config file with the necessary information for your project. The formatting of the config file is as follows:

# your relative file path
config_path: "config/YYYY_MM_DD_config.yaml"

# the date of the analysis
date: "YYYY_MM_DD"

# the name of the project
project_name: "my_project"

# the description of the project
description: "Cool analysis of something"

# the path to the raw data
sample_summary_csv: "data/raw/UMI_counts.csv"

# the path to the output report file
output_file: "YYYY_MM_DD_final.html"

9. Run the snakemake pipeline (in the terminal), that creates date specific subdirectories in the data/processed, plots and reports directories and then renders the quarto document with the information from your config file.

snakemake -s Snakefile -c 1 --configfile config/YYYY_MM_DD_config.yaml

10. To clear all output and rerun this pipeline, run the following commands:

snakemake -s Snakefile --delete-all-output -c 1 --configfile config/YYYY_MM_DD_config.yaml
make clean

What is usually in these directories?

• data:

  • data/raw: This is where you put your raw data files. These files are not modified in any way. They are the starting point of your analysis.
  • data/processed: This is where you put your processed data files. These files are created by your scripts and are used as input for your analysis.

• plots: This is where you put your plots. These plots are created by your scripts and are used in your report.

  • The subdirectories in plots are named after the date of the analysis.

• reports: This is where you put your reports. These reports are created by your scripts and are the final output of your analysis. They are named by date.

• src: This is where you put your source code.

  • Analysis scripts (performing PCA, etc) are stored in the analysis subdirectory.
  • Processing scripts (reformatting, etc) are stored in the processing subdirectory.
  • Source scripts (common functions you use and make) are stored in the utils subdirectory.

• config: This is where you put your configuration files. These files contain the information needed for your scripts to run.