Introduction to the Tidyverse

# Introduction to the Tidyverse
## part 2: data visualization with ggplot2
### Martin Ndemo Sato
### 04/21/2020

---

# Working with data in R
## the tidyverse is a collection of friendly and consistent tools for data analysis and visualization.

---
class: inverse-ns, center

# Working with data in R
## the tidyverse is a collection of friendly and consistent tools for data analysis and visualization. 
## They live as R packages each of which does one thing well.

---

background-image: url(https://raw.githubusercontent.com/tidyverse/tidyverse/master/man/figures/logo.png)
background-position: 90% 10%
## `library(tidyverse)` will load  
## the core tidyverse packages:

#### [ggplot2](http://ggplot2.tidyverse.org), for data visualisation.
#### [dplyr](http://dplyr.tidyverse.org), for data manipulation.
#### [tidyr](http://tidyr.tidyverse.org), for data tidying.
#### [readr](http://readr.tidyverse.org), for data import.
#### [purrr](http://purrr.tidyverse.org), for functional programming.
#### [tibble](http://tibble.tidyverse.org), for tibbles, a modern re-imagining of data frames.
#### [stringr](https://github.com/tidyverse/stringr), for strings.
#### [forcats](https://github.com/hadley/forcats), for factors.

---
background-image: url(http://hexb.in/hexagons/ggplot2.png)
background-position: 90% 10%

# ggplot2: Elegant Data Visualizations in R

## a Layered Grammar of Graphics

---
background-image: url(http://hexb.in/hexagons/ggplot2.png)
background-position: 90% 10%

# ggplot2: Elegant Data Visualizations in R

## a Layered Grammar of Graphics
## Data is mapped to aesthetics; Statistics and plot are linked

---
background-image: url(http://hexb.in/hexagons/ggplot2.png)
background-position: 90% 10%

# ggplot2: Elegant Data Visualizations in R

## a Layered Grammar of Graphics
## Data is mapped to aesthetics; Statistics and plot are linked
## Sensible defaults; Infinitely extensible

---

background-image: url(http://hexb.in/hexagons/ggplot2.png)
background-position: 90% 10%

## Publication quality & beyond

### [https://nyti.ms/2jUp36n](https://nyti.ms/2jUp36n)

### [http://bit.ly/2KSGZLu](http://bit.ly/2KSGZLu)

---

```r
ggplot()
```

---

```r
ggplot(mtcars, aes(x = mpg, y = hp))
```

---

```r
ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point()
```

---

# `ggplot()`
#### `ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) + `
#### &nbsp;&nbsp; `<GEOM_FUNCTION>()`

## Add layers with <code>+</code>

## Put `+` at the end of a line

## map aesthetics with <code>aes()</code>

---

```r
diabetes <- read_csv("diabetes.csv")

ggplot(data = diabetes, mapping = aes(x = weight, y = hip)) +
  geom_point()
```

---

# Aesthetics: `aes()`

#### `ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) + `
#### &nbsp;&nbsp; `<GEOM_FUNCTION>()`
 
--

## Aesthetics map the data to the plot.
 
---

# Aesthetics: `aes()`

```r
ggplot(mtcars, aes(x = mpg, y = hp, color = cyl)) + geom_point()

ggplot(mtcars, aes(x = mpg, y = hp, size = cyl)) + geom_point()

ggplot(mtcars, aes(x = mpg, y = hp, alpha = cyl)) + geom_point()

ggplot(mtcars, aes(x = mpg, y = hp, shape = cyl)) + geom_point()
```

---

```r
ggplot(data = diabetes, 
       mapping = aes(x = weight, y = hip, color = gender)) +
  geom_point()
```

---

```r
ggplot(data = diabetes, 
       mapping = aes(x = weight, y = hip, size = gender)) +
  geom_point()
```

---

```r
ggplot(data = diabetes, 
       mapping = aes(x = weight, y = hip, alpha = gender)) +
  geom_point()
```

---

```r
ggplot(data = diabetes, 
       mapping = aes(x = weight, y = hip, shape = gender)) +
  geom_point()
```

---

```r
ggplot(data = diabetes) +
  geom_point(mapping = aes(x = weight, y = hip, 
                           color = gender, shape = gender))
```

---

# geoms

## What shape does the data take?

## `geom_point()`

## `geom_line()`

## `geom_violin()`

# Check the cheatsheet!

---

```r
ggplot(diabetes, aes(gender, chol)) + 
* geom_boxplot()
```

---

```r
ggplot(diabetes, aes(x = glyhb)) + 
* geom_histogram()
```

---

```r
ggplot(diabetes, aes(x = glyhb)) + 
* geom_density()
```

---

```r
diabetes %>% 
* ggplot(aes(x = frame)) +
*   geom_bar()
```

---

```r
diabetes %>% 
* drop_na() %>%
  ggplot(aes(x = frame)) + 
    geom_bar()
```

---

```r
diabetes %>% 
  drop_na() %>% 
* ggplot(aes(x = frame, color = gender)) +
    geom_bar()
```

---

```r
diabetes %>% 
  drop_na() %>% 
* ggplot(aes(x = frame, fill = gender)) +
    geom_bar()
```

---

## Positions

### `geom_bar(position = "<POSITION>")`

## When we have aesthetics mapped, how are they positioned?

## bar: dodge, fill, stacked (default)

## point: jitter

---

```r
ggplot(mtcars, aes(x = factor(am), y = hp)) + 
  geom_point()
```

---

```r
ggplot(mtcars, aes(x = factor(am), y = hp)) + 
* geom_point(position = "jitter")
```

---

```r
ggplot(mtcars, aes(x = factor(am), y = hp)) + 
* geom_jitter(width = .1, height = 0)
```

---

```r
diabetes %>% 
  drop_na() %>% 
  ggplot(aes(x = frame, fill = gender)) + 
*   geom_bar(position = "stack")
```

---

```r
diabetes %>% 
  drop_na() %>% 
  ggplot(aes(x = frame, fill = gender)) + 
*   geom_bar(position = "dodge")
```

---

```r
diabetes %>% 
  drop_na() %>% 
  ggplot(aes(x = frame, fill = gender)) + 
*   geom_bar(position = "fill")
```

---

## Mapping vs setting

## Cool, but how do I just make everything blue?

## `geom_point(aes(x, y), color = "blue")`

## To set a color, put it outside `aes()`

---

```r
ggplot(mtcars, aes(x = mpg, y = hp, color = cyl)) + 
  geom_point(
*   color = "blue"
  )
```

---

```r
ggplot(mtcars, aes(x = mpg, y = hp, color = cyl)) + 
  geom_point(
*   aes(color = "blue")
  )
```

---

## Adding layers

```r
ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) +
* <GEOM_FUNCTION>() +
* <GEOM_FUNCTION>() +
* <SCALE_FUNCTION>() +
* <THEME_FUNCTION>()
```

---

## Live Code Part 1

#### 1. Predict what this code will do. Then run it. 
#### 2. Add a `linetype` aesthetic for `gender`. Run it again.
#### 3. Set the color of `geom_smooth()` to "black"
#### 4. Add `se = FALSE` to the `geom_smooth()`
#### 5. It's hard to see the lines well now. How about setting `alpha = .2` in `geom_point()`?
#### 6. Jitter the points. You can either change the geom or change the `position` argument.
#### 7. Add another layer, `theme_bw()`. Remember to use `+`.

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point() +
  geom_smooth()
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point() +
  geom_smooth()
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point() +
  geom_smooth(aes(linetype = gender))
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point() +
  geom_smooth(aes(linetype = gender), col = "black")
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point() +
  geom_smooth(aes(linetype = gender), col = "black", se = FALSE)
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point(alpha = .2) +
  geom_smooth(aes(linetype = gender), col = "black", se = FALSE)
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_jitter(alpha = .2) +
  geom_smooth(aes(linetype = gender), col = "black", se = FALSE)
```

---

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_jitter(alpha = .2) +
  geom_smooth(aes(linetype = gender), col = "black", se = FALSE) +
  theme_bw()
```

---

## Facets

## Easy peazy panels

## `facet_grid()`
## `facet_wrap()`

## `x ~ y`
## ` ~ y`
## `x ~ .`

---

```r
diamonds %>% 
  ggplot(aes(x = carat, price)) + 
  geom_point() + 
* facet_grid(cut ~ clarity)
```

---

### facet grid by `gender` and `location`

```r
ggplot(diabetes, aes(weight, hip)) + 
  geom_point() +
  geom_smooth() + 
* facet_grid(gender ~ location)
```

---

---

```r
diamonds %>% 
  ggplot(aes(x = carat, price)) + 
  geom_point() + 
* facet_wrap(~clarity)
```

---

## datasauRus

```r
library(datasauRus)
datasaurus_dozen 
```

```
*## # A tibble: 1,846 x 3
*## dataset x y
## <chr> <dbl> <dbl>
## 1 dino 55.4 97.2
## 2 dino 51.5 96.0
## 3 dino 46.2 94.5
## 4 dino 42.8 91.4
## 5 dino 40.8 88.3
## 6 dino 38.7 84.9
## 7 dino 35.6 79.9
## 8 dino 33.1 77.6
## 9 dino 29.0 74.5
## 10 dino 26.2 71.4
## # ... with 1,836 more rows
```

---

```r
datasaurus_dozen %>% 
* group_by(dataset) %>%
  summarize(corr = cor(x, y)) %>% 
  mutate(corr = round(corr, 3))
```

```
## # A tibble: 13 x 2
## dataset corr
## <chr> <dbl>
## 1 away -0.064
## 2 bullseye -0.069
## 3 circle -0.068
## 4 dino -0.064
## 5 dots -0.06 
## 6 h_lines -0.062
## 7 high_lines -0.069
## 8 slant_down -0.069
## 9 slant_up -0.069
## 10 star -0.063
## 11 v_lines -0.069
## 12 wide_lines -0.067
## 13 x_shape -0.066
```

---

```r
datasaurus_dozen %>% 
  ggplot(aes(x, y)) + 
    geom_point() + 
*   facet_wrap(~dataset)
```

---

---

## Scales

## position scales

### `scale_x_continuous()`
### `scale_y_date()`
### `scale_x_log10()`

---

## Scales

## aesthetic scales

### `scale_color_hue()`
### `scale_fill_brewer()`
### `scale_shape_manual()`

---

```r
mtcars %>% 
ggplot(aes(hp, mpg, col = factor(cyl))) + 
  geom_point() + 
  scale_x_log10() + 
* scale_colour_brewer(palette = "Set2")
```

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = weight)) + 
  geom_point() 
```

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = weight)) + 
  geom_point() + 
* scale_color_viridis_c()
```

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = gender)) + 
  geom_point() + 
* scale_color_brewer()
```

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = gender)) + 
  geom_point() + 
* scale_color_manual(values = c("#E69F00", "#56B4E9"))
```

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = gender)) + 
  geom_point() + 
* scale_color_manual(name = "Sex", values = c("#E69F00", "#56B4E9"))
```

---

## Color Palettes

## [https://github.com/EmilHvitfeldt/r-color-palettes](https://github.com/EmilHvitfeldt/r-color-palettes)

---

## Themes

## Non-data ink (text, background, etc)

## Prespecified themes: `theme_gray()` (default), `theme_minimal()`, `theme_light()`, etc.

## `theme()`

---

```r
mtcars %>% 
ggplot(aes(hp, mpg, col = factor(cyl))) + 
  geom_point(size = 3) + 
  scale_x_log() + 
  scale_colour_brewer(name = "Cylinders", palette = "Set2") +
* theme_minimal() +
  theme(axis.text = element_text(size = 16),  
        legend.text = element_text(size = 8, face = "bold"),  
        legend.direction = "horizontal")  
```

---

```r
mtcars %>% 
ggplot(aes(hp, mpg, col = factor(cyl))) + 
  geom_point(size = 3) + 
  scale_x_log() + 
  scale_colour_brewer(name = "Cylinders", palette = "Set2") +
  theme_minimal() +  
* theme(axis.text = element_text(size = 16),
*       legend.text = element_text(size = 8, face = "bold"),
*       legend.direction = "horizontal")
```

---

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = weight)) + 
  geom_point() +
  scale_color_viridis_c()
```

---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = weight)) + 
  geom_point() +
  scale_color_viridis_c() +
* theme_minimal() +
  theme(legend.position = "bottom", 
        axis.ticks = element_blank(), 
        axis.title = element_text(size = 16)) 
```
---

```r
diabetes %>% 
  ggplot(aes(waist, hip, col = weight)) + 
  geom_point() +
  scale_color_viridis_c() +
  theme_minimal() + 
* theme(legend.position = "bottom",
*       axis.ticks = element_blank(),
*       axis.title = element_text(size = 16))
```

---

---

## High-density plots

```r
diamonds
```

```
*## # A tibble: 53,940 x 10
## carat cut color clarity depth table price x y z
## <dbl> <ord> <ord> <ord> <dbl> <dbl> <int> <dbl> <dbl> <dbl>
## 1 0.23 Ideal E SI2 61.5 55 326 3.95 3.98 2.43
## 2 0.21 Premium E SI1 59.8 61 326 3.89 3.84 2.31
## 3 0.23 Good E VS1 56.9 65 327 4.05 4.07 2.31
## 4 0.290 Premium I VS2 62.4 58 334 4.2 4.23 2.63
## 5 0.31 Good J SI2 63.3 58 335 4.34 4.35 2.75
## 6 0.24 Very Good J VVS2 62.8 57 336 3.94 3.96 2.48
## 7 0.24 Very Good I VVS1 62.3 57 336 3.95 3.98 2.47
## 8 0.26 Very Good H SI1 61.9 55 337 4.07 4.11 2.53
## 9 0.22 Fair E VS2 65.1 61 337 3.87 3.78 2.49
## 10 0.23 Very Good H VS1 59.4 61 338 4 4.05 2.39
## # ... with 53,930 more rows
```

---

```r
diamonds %>% 
  ggplot(aes(x = carat, price)) + 
  geom_point()
```

---

## High-density plots

## Transparency

## Binning

---

```r
diamonds %>% 
  ggplot(aes(x = carat, price)) + 
* geom_point(alpha = .05)
```

---

```r
diamonds %>% 
  ggplot(aes(x = carat, price)) + 
* geom_bin2d()
```

---

```r
diamonds %>% 
  ggplot(aes(x = carat, price)) + 
* geom_hex()
```

---

## Labels, titles, and legends

## Add a title: 
#### `ggtitle()`, `labs(title = "My Awesome Plot")`

## Change a label: 
#### `xlab()`, `ylab()`, `labs(x = "X!", y = "Y!!")`

## Change a legend: 
#### scale function, `labs(color = "Wow, labs does everything", fill = "Yup")`

---

## Saving plots

## `ggsave(p, path = "figure_name.png", dpi = 320)`

---

## Live Code Part 2

#### 1. Add a title. 
#### 2. Change the x and y axis labels to include the unites (inches for `hip` and pounds for `weight`). You can use either `labs()` or `xlab()` and `ylab()`
#### 3. Add `scale_linetype()` and set the `name` argument to "Sex".
#### 4. Save the plot.

```r
ggplot(diabetes, aes(weight, hip, linetype = gender)) +
  geom_jitter(alpha = .2, size = 2.5) +
  geom_smooth(color = "black", se = FALSE) + 
  theme_bw(base_size = 12)
```

---

```r
ggplot(diabetes, aes(weight, hip, linetype = gender)) +
  geom_jitter(alpha = .2, size = 2.5) +
  geom_smooth(color = "black", se = FALSE) + 
  theme_bw(base_size = 12) +
  labs(x = "Weight (lbs)", y = "Hip (inches)") +
  ggtitle("Hip and Weight by Sex") + 
  scale_linetype(name = "Sex")
```

---

---

```r
ggsave(path = "diabetes_figure.png", dpi = 320)
```

---

```r
*diabetes_plot <-
 ggplot(diabetes, aes(weight, hip, linetype = gender)) + 
 geom_jitter(alpha = .2, size = 2.5) +
 geom_smooth(color = "black", se = FALSE) + 
 theme_bw(base_size = 12) +
 labs(x = "Weight (lbs)", y = "Hip (inches)") +
 ggtitle("Hip and Weight by Sex") + 
 scale_linetype(name = "Sex")

ggsave(diabetes_plot, path = "diabetes_figure.png", dpi = 320) 
```

---

```r
diabetes_plot <-
 ggplot(diabetes, aes(weight, hip, linetype = gender)) + 
 geom_jitter(alpha = .2, size = 2.5) +
 geom_smooth(color = "black", se = FALSE) + 
 theme_bw(base_size = 12) +
 labs(x = "Weight (lbs)", y = "Hip (inches)") +
 ggtitle("Hip and Weight by Sex") + 
 scale_linetype(name = "Sex")

*ggsave(diabetes_plot, path = "diabetes_figure.png", dpi = 320)
```

---

# You can use this code template to make thousands of graphs with ggplot2.

```r
ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) +
 <GEOM_FUNCTION>() +
 <SCALE_FUNCTION>() + 
 <THEME_FUNCTION>()
```

---

![](https://media.giphy.com/media/OWefeMtyJex4A/giphy.gif)

---

# Resources
## [R for Data Science](http://r4ds.had.co.nz/): A comprehensive but friendly introduction to the tidyverse. Free online.
## [DataCamp](https://www.datacamp.com/): ggplot2 courses and tidyverse courses
## [ggplot2: Elegant Graphics for Data Analysis](https://smile.amazon.com/ggplot2-Elegant-Graphics-Data-Analysis/dp/331924275X/ref=sr_1_2?ie=UTF8&qid=1524362742&sr=8-2&keywords=ggplot2): The official ggplot2 book

---

![](https://media.giphy.com/media/KJ1f5iTl4Oo7u/giphy.gif)

### [ndemosato](https://github.com/ndemosato/)
### [@SatoMartin](https://twitter.com/SatoMartin)

Slides created via the R package [xaringan](https://github.com/yihui/xaringan).

---

## Bonus!

---

## Arranging plots

## Spatial plots

## Interactive plots

## Animated plots

---