leha: Legume consumption and hepatobiliary disease

• Protocol:
• Preprint: doi.org/10.1101/2024.12.09.24315856
• Analysis code:

This project investigates the association between substituting legumes for meats, fish, or poultry and the risk of developing NAFLD and gallbladder diseases.

Installing and setting up the project

If dependencies have been managed by using usethis::use_package("packagename") through the DESCRIPTION file, installing dependencies is as easy as opening the .Rproj file and running this command in the console:

pak::pak()

Steps to select and download the data

The data-raw/ folder contains the scripts to select, process, and prepare the data on the RAP to eventually be downloaded.

The steps to take to select the variables you want, create the CSV file on the RAP, convert it to Parquet format (for faster loading), and download to your project on RAP. The order is:

1. Select the variables you want in data-raw/project-variables.csv.
2. Follow the instructions in the data-raw/create-data.R script and run it to create the CSV file on the RAP server.
3. Open the _targets.R and change the download_project_data target in tar_target() line from "parquet" to "csv". Then run targets::tar_make() to download the CSV file to data/. RESTART SESSION
4. Open and run the data-raw/convert-to-parquet.R script to convert the CSV file to the Parquet format.
5. Go back into _targets.R and change the "csv" to "parquet" (the opposite of what was done in item 3). Run targets::tar_make() in the Console to download the Parquet file and store in the data/ folder.

Processing and saving progress

It is very timely to rerun all code every time you have made changes in the data, so you can save and reload your work along the way by following the layout in "data-raw/processing.r" script. This is also helpful to do, when you have completed all data management tasks and want to save your changes to have a "ready to go" data frame to run your analysis on.

After having completed the data management and running the processing.R script, you can start doing descriptive and comparative analyses of your data.

Brief description of folder and file contents

The following folders contain:

• data/: Will contain the UK Biobank data (not saved to Git) as well as the intermediate results output files.

• data-raw/: Contains the R script to download the data, as well as the CSV files that contain the project variables and the variable list as named in the RAP.

• doc/: This file contains the R Markdown, Word, or other types of documents with written content, like the manuscript and protocol.

• R/: Contains the R scripts and functions to create the figures, tables, and results for the project.