Who am I?

What to expect during this workshop

Workshop resources

Course website: nrennie.rbind.io/r-pharma-2024-parameterized-reports

What do I mean by parameterized plots?

Why parameterize plots?

Essentially so that you Don’t Repeat Yourself:

• Once you’ve parameterized a plot, it’s quick to make different versions of the same plot.

• Scripts are shorter and easier to read because you’re not copying and pasting code.

• Ensure consistent layout between plots.

Start with variables…

Create variables that define the data you’re plotting:

month <- "Apr"
year <- 2024

…then build functions!

plot_function <- function(month, year) {
  # plot code goes here!
}

plot_function <- function(month, year, colour = "blue") {
  # plot code goes here!
}

Live Demo!

• Replacing hard-coded values with variables

• Transforming code into a function

• Using glue() for parameterized plot text

Exercise 1

1. Create a variable to define a year in the gapminder data that you want to plot.

2. Edit the code in the gapminder_year.R file to use this new year variable.

3. Turn your code into a function that takes the data and year variable as arguments (and check it works!)

4. Bonus: Add another argument to your function that allows a user to define the bar colour.

5. Bonus: How could you improve your function?

What is Quarto?

What makes a Quarto document?

YAML header

---
title: "A very cool title"
format: html
---

Output types

Quarto parameters

---
title: "A very cool title"
format: html
params:
  year: 2023
---

Note: you can do something similar with R Markdown.

Using parameters in code

Subset data based on parameters:

data_subset <- data |>
  dplyr::filter(
    year == params$year
  )

Parameters with Python code

```{python}
#| tags: [parameters]

alpha = 0.1
ratio = 0.1
```

Using in code:

```{python}
alpha
```

Live Demo!

• Adding a continent parameter to the document

• Accessing and using the parameter in plotting

• Using the parameter for tables

Exercise 2

1. Edit the report.qmd file to add a parameter for year.

2. Add your plotting function to the quarto document, and pass the parameter into the plotting function. Check it renders.

3. Change the value of the parameter, and re-render.

Inline text

Sometimes you want to use values from code in a sentence. We can add inline code:

The number of rows is `r nrow(gapminder)`.

Captions

You can also execute R code in figure captions:

```{r}
#| label: fig-gapminder
#| fig-cap: !expr 'paste("The number of rows is ", nrow(gapminder))'
#| output-location: column-fragment
#| eval: true
hist(gapminder$lifeExp)
```

Conditional content

Display content only for a specific format:

::: {.content-visible when-format="html"}

Will only appear in HTML.

:::

::: {.content-hidden unless-format="pdf"}

Will be hidden unless it's a PDF document.

:::

Conditional content

Display content only for a specific parameter value:

`r if (params$hide_answers) "::: {.content-hidden}"`

The answer is 4.

`r if (params$hide_answers) ":::"`

Quarto profiles

• Main project configuration: _quarto.yml

• solutions profile configuration: _quarto-solutions.yml

::: {.content-hidden when-profile="solutions"}

This content will be hidden in the advanced profile

:::

Live Demo!

• Writing inline R code

• Using parameter values in inline R code

• Using parameter values in figure captions

Exercise 3

1. Edit the sentence ...a plot of the median life expectancy in the year 1952 for each continent... to use inline code to add the document parameter for the year.

2. Use inline code for the title as well.

3. Edit the figure caption to use inline code to add the document parameter for the year.

Rendering multiple reports

Let’s say you want to create multiple different versions of your report.

• Manually change parameter values and hit render?

• Pass parameters in one at a time to the command line?

Using the {quarto} R package

Using the {purrr} package

Note: if you’re more a base R person, you can do something similar with the apply() family of functions.

Live Demo!

• Using the {quarto} R package.

• Using {purrr} to generate multiple reports.

Exercise 4

1. Render a version of your report for the year 1972 using the quarto_render() function.

2. Render a version of your report for all years in the data.

Workshop resources

Workshop website: nrennie.rbind.io/r-pharma-2024-parameterized-reports