80’s idea from D. Knuth:

write code for humans, not for computer

One document to rule them all with both text & code
=> export to different formats

source: Rmarkdown cheatsheet

Science reproducibility crisis

• Is most scientific research false ? Ioanidis et al.

• have you had to redo a whole analysis with just slightly different data
  (e.g. 1 patient added or removed ?)

Fench book on reproducible research:

Vers une recherche reproductible : faire évoluer ses pratique

• markdown: lightweight structured text

• Rmarkdown = R + markdown

• Rmarkdown + pandoc
  => output in many formats (.docx, .pdf, .html, etc…)

source: Rmarkdown cheatsheet

These materials have been prepared in Rmarkdown

1. YAML Header

---
title: 'BADAS: Part II'
subtitle: "Dynamic & Reproducible reporting with `Rmarkdown`"
author: "Boris Hejblum & Loïc Ferrer"
date: "June 17, 2019"
output: 
  ioslides_presentation:
    number_sections: no
---

• Text
  Narration

• Code Chunks

```{r}
```

• results (default is 'markup')

  • 'asis' - passthrough results
  • 'hide' - do not display results
  • 'hold' - put all results below all code

• eval - Run code in chunk (default is TRUE)

• echo - Display code in output document (default is TRUE)

• message - display code messages in document (default is TRUE)

• error - Display error messages in output document (TRUE) or stop render when errors occur (FALSE)
  (default is FALSE)

• tidy - tidy code for display (default is FALSE)
• warning - display code warnings in document (default = TRUE)
• collapse - collapse all output into single block (default = FALSE)
• highlight - highlight source code (default = TRUE)
• include - include chunk in doc after running (default = TRUE)
• comment - prefix for each line of results (default = ‘##’)

• fig.align - figure postion: 'left', 'right', or 'center'
• fig.cap - figure caption as character string (default is NULL)
• fig.height, fig.width - Dimensions of plots in inches
• out.width - only in pdf: size of the figure

Long running code chunks

• cache - cache results for future knits (default is FALSE)
• cache.path - directory to save cached results in (default is "cache/")
• dependson - chunk dependencies for caching (default is NULL)

Multiple files

• child - file(s) to knit and then include (default is NULL)

not just R (python, SAS…)

• engine - code language used in chunk (default = ‘R’)

• *italic* — italic
• **bold** — bold
• ~~no~~ — strikethrough
• `code` — verbatim code
• ^2^ — superscript²
• ~2~ — subscript₂
• $\alpha = \beta^2$ — LaTeX equations: \(\alpha = \beta^2\)
• End a line with two spaces to start a new paragraph

Headings

# Header1 {#anchor}
## Header 2

Jump to [Header 1](#anchor)

Footnotes

A footnote [^1]
[^1]: Here is the footnote.

Include images

![Caption](pathtoimage.png)

 - Bullet 1
 - Bullet 2

• Bullet 1
• Bullet 2

1. item 1
2. item 2

1. item 1
2. item 2

| Right | Left | Default | Center |
|------:|:-----|---------|:------:|
| 12 | 12 | 12 | 12|
| 123 | 123 | 123 | 123 |
| 1 | 1 | 1 | 1 |

---
bibliography: refs.bib 
csl: style.csl
---

Smith cited [@smith04].
Smith cited without author [-@smith04].
@smith04 cited in line.

• Rstudio cheatsheet

• Rmarkdown book by Yihui Xie, JJ Allaire & G Grolemund
  (very) comprehensive…

1. create a new .Rmd file
2. play along with the various options described here
3. try various output formats
4. turn your answers to the first practical into a .Rmd