paperTemplate

Template repository for scientific journal articles and associated materials

Input data

Use the data-raw folder to save code that extracts or downloads the input data used for your paper. This code can output data to the data directory.

Use the data directory to store input data. In this template, the data directory will not be tracked with git (see .gitignore file).

Analysis

Use the analysis directory to save analysis code. Analysis should be split into separate scripts appropriately. Operations that are common across analysis scripts can be functionalized and saved to the utils.R file to minimize repeated code.

Outputs of analysis can also be saved to the data directory as .rds files, or a new results directory can be created, as needed.

Writing

Rmarkdown files are useful for paper writing, because code can be embedded within, making number-plugging easier and more reproducible.

This template includes three Rmarkdown files:

1. manuscript
2. tables_figures
3. appendix

It may be appropriate to combine the tables and figures with the manuscript, depending on how many tables/figures are included and how long they take to knit.

Go here for Rmarkdown instructions.

PDF output from knit Rmarkdown files are not tracked with git in this template. Use the .gitignore file if you want to change this.

References

To save bibliographic information about sources used for your paper, use a .bib file. This template includes references.bib. For information about .bib file syntax, go here.

Reference manager softwares like Zotero and Mendeley can also export .bib files. These softwares have web plugins for easy extraction of reference metadata from web pages, and can be a nice way to store and track references across projects.

For information about citations in Rmarkdown, including instructions for in-line citations, go here.

Reference styles can be managed using .csl files. This template includes a .csl file for The Lancet, but other journal styles can be downloaded from the Zotero Style Repository.

The .bib and .csl files are listed in the YAML header of all Rmarkdown documents that use references:

---
title: "manuscript"
author: "author"
date: "date"
output: pdf_document
bibliography: references.bib
csl: the-lancet.csl
---

Figures

One recommended workflow for figures, represented in this template, is to save functions for generating figures in a figures directory, one per file, then load in data and call these functions within the Rmarkdown files.

This process makes it easier to view the code for a specific figure, and moves a lot of excess code out of the Rmarkdown file. Minimizing the amount of code in Rmarkdown files makes it easier to read and focus on the text and structure.

If figures take a long time to compile, it may be more appropriate to have tables and figures code export .csv and .jpg files, which are then included in the Rmarkdown scripts without needing to be recompiled from scratch.