Data Carpentry for Biologists

Quarto

Notes

Literate programming

  • Combine text, code, figures, tables, etc.
  • Quickly generate websites
  • Automatically generate reports
  • Easy to share results and code with collaborators

Getting started

install.packages(c("quarto", "dplyr", "ggplot2", "gt", "readr"))
  • File -> New File -> Quarto Document
  • Title: Portal Population Dynamics
  • Default Output Format: HTML
  • Uncheck Use visual markdown editor
  • Create
  • Generates a basic stub of a .qmd document
  • The material between the ---’s is called “front matter”
  • Metadata about the document
  • Title, author, and what we want to create from this file by default
  • Delete everything below the second ---

Markdown

  • Basic approach to formatting text
  • Let’s you do
    • # Headers
    • *italics*
    • **bold**
    • [links](http://google.com)
    • Lists
      • *
      • 1.

## Concept

Exploration of population dynamic patterns at **The Portal Project**.
How do counts of rodents like *Dipodomys* species change through time?

In this document I will:

1. Load data from the [Portal Project](https://portal.weecology.org)
2. Process it into population time series
3. And make initial visualizations
  • Easy to read for humans

  • Easy to convert into other things for computers

  • Press Render to create HTML from the document
  • Can also create pdf & Word versions of our files

  • Change output to pdf or docx

  • Switch back to html
  • Click the gear icon and select Preview in Viewer Pane
  • Render

  • This shows the rendered page right in RStudio

  • Markdown is common on lots of websites
  • Used to create all of the exercises and lectures in this course

Visual editor

  • We can also view and interact with the markdown in a style similar to a word processor
  • Click Visual
  • Markdown is now rendered in the editor
  • Can use the editor bar to make markdown changes
  • UNBOLD: The Portal Project
  • Click Source
  • See that it is no longer bolded using **

R chunks

  • Quarto allows you to include code to run in the document
  • Click on +C button, which stands for “add code”
  • Or in the visual editor type / and select R Code Chunk

## Required Packages

```{r}
library(dplyr)
library(ggplot2)
library(gt)
library(readr)
```
  • Can run this code by clicking the green arrow
  • Load the data and look at it

## Data

```{r}
data <- read_csv("https://ndownloader.figshare.com/files/2292172")
data
```
  • If we run the code with the green arrow we’ll get a nicely formatted table

  • If we click the Render button the output will be included in the html web page

  • You can run code inside your text, too:
The data includes `r length(unique(data$species_id))` species.

Analysis Example


## Analysis

Get the time-series of counts for all species.
          
```{r}
time_series <-
  data |>
  group_by(species_id, year) |>
  summarize(count = n()) |>
  filter(sum(count) > 200)
time_series
```

Including Graphs

  • Graphs will be included as figures in the document

## Plot the time-series.

```{r}
ggplot(time_series, aes(x = year, y = count)) +
  geom_line() +
  facet_wrap(vars(species_id), scales = "free_y")
```

Chunk options

  • Each block of code is called a “chunk”
  • Chunks have lots of useful options
  • Can add them at the start of the chunk using #| , the option name, :, the option value
  • #| echo: false let’s you show the results of the code chunk without showing the code.
  • ADD: #| echo: false to plot chunk
  • #| message: false will remove messages returned by R
  • ADD: #| message: false to package, data, and analysis chunks

  • RENDER

  • There are also chunk options to include figure legends and alt text
  • ADD TO FIGURE CHUNK:
#| fig-cap: "Figure 1. Time-series of rodent population dynamics at Portal."
#| fig-alt: "Time-seris of 12 rodent species showing a range of dynamics from increases to decreases to variable, but not directionally changing, response"

Tables

  • We can also make nicely formated tables using the gt() function
  • Let’s say we need to know who many total individuals of each species were included in the analysis

```{r}
time_series |>
  group_by(species_id) |>
  summarize(total_count = sum(count)) |>
  gt() |>
  cols_label(
      species_id = "Species",
      total_count = "Total Population"
      )
```
  • You can add a label and then cite the table/figure in the text
#| label: tbl-total-count
#| tbl-cap: "Total population counts for each species in the study"
  • You can then reference the table in the text using @tbl-total-count
The population dynamics at the site varies by species (@tbl-total-count).

Citations

  • Insert -> Citation -> From DOI -> 10.1101/332783 -> + -> Insert
  • Creates a bibliography.bib file with the citations
  • This is called a bibtex file
  • Can now cite papers using [@ernest2018]
  • It will appear as an in-text citation and in a References section at the bottom of the document

Other Output Formats

  • Slides
  • Websites