Polishing plots

class: center, middle, inverse, title-slide

# Polishing plots

---

background-image: url(https://raw.githubusercontent.com/allisonhorst/stats-illustrations/master/rstats-artwork/ggplot2_masterpiece.png)
background-size: 450px
background-position: 50% 90%
## graphics for communication

We have mostly been focused on **exploratory data analysis**

- i.e.; used plots as tools for exploration

After you understand your data, you need to **communicate** your understanding to others.

---
## `ggplot2` provides defaults ...

- but every aspect of the plot can be changed

- colors are controlled through **scales**

- **themes** control presentation of non-data elements

---
class: inverse
## Outline

### 1. labels & annotations

### 2. ggplot2 scales

### 3. scales & color choices

### 4. themes

---
## Adding labels

You add labels with the `labs()` function.

- Labels that can be modified include:  
  - `x`  
  - `y`  
  - `title`  
  - `subtitle`  
  - `caption`
  - `color`

Other methods of modifying labels:

- `ggtitle(main, subtitle)`: plot title & subtitle

- `xlab()`, `ylab()`: axes titles

---
## Adding labels

```r
ggplot(mpg, aes(displ, hwy)) + geom_point(aes(color = class)) +
  labs(title = "Fuel efficiency generally decreases with engine size",
    subtitle = "Two seaters (sports cars) are an exception because of their light weight",
    caption = "Data from fueleconomy.gov",
    x = "Engine displacement (L)", y = "Highway fuel economy (mpg)",
    colour = "Car type")
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-2-1.png)

---
## Adding text

Use `geom_text()` or `geom_label()` to label interesting observations.

```r
best_in_class <- mpg %>% group_by(class) %>%
 dplyr::filter(row_number(desc(hwy)) == 1)

ggplot(mpg, aes(displ, hwy)) +  geom_point(aes(colour = class)) +
  geom_text(aes(label = model), data = best_in_class)
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-4-1.png)

???
**Example**: pull out the most efficient car in each class with dplyr, and then label it on the plot:
---
## Adding text

```r
ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(colour = class)) +
  geom_point(size = 3, shape = 1, data = best_in_class) +
  ggrepel::geom_label_repel(aes(label = model), data = best_in_class)
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-5-1.png)

---
## Annotations

An **annotation** is a separate layer that doesn't connect to other elements in the plot and is used to add fixed elements to a data visualization

The `annotate()` function creates an annotation layer

- arguments include `geom`, and positions (`x`, `y`, `xmin`, `ymin`, etc.)
  
---
## Example

```r
p <- ggplot(mpg, aes(displ, hwy)) +
 geom_point(data = dplyr::filter(mpg, manufacturer == "subaru"), colour = "orange", size = 3) +
 geom_point() 
p
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-6-1.png)

---
## Adding annotations

```r
p + 
  annotate(geom = "curve", x = 4, y = 35, xend = 2.65, yend = 27, 
           curvature = .3, arrow = arrow(length = unit(2, "mm"))) +
  annotate(geom = "text", x = 4.1, y = 35, label = "subaru", hjust = "left")
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-7-1.png)

---
## Scales

**Recall**: Scales control the details of how data values are translated to visual properties.

- Every aes value has a corresponding family of scales functions

- `scale_{aes}_*()`, e.g. `scale_x_continuous()`  
  - Values of the `*` depend on the aes
  
- These scale functions have many arguments including:
  - `name`: label of the axis/legend
  - `breaks`: numeric positions of breaks on axes/legends
  - `labels`: labels of the breaks on axes/legends
  - `limits`: continuous axis limits
  - `expand`: padding around data
  - `na.value`: what to do with missings
  - `trans`: continuous transformations of data
  - `guide`: function to create guide/legend
  - `date_breaks`: breaks for date variables

---
## Scales for axes

.pull-left[
`scale_x_*()`, `scale_y_*()`
- continuous
- discrete
- binned
- log10
- sqrt 
- date
- datetime
- reverse
]

.pull-right[

]

---
## Scales for color

- Colors are controlled through scales
`scale_colour_discrete` (`scale_colour_hue`) and `scale_colour_continuous` (`scale_colour_gradient`) are the default choices for factor variables and numeric variables

- we can change parameters of the default scale, or we can change the scale function

.pull-left[
`scale_color_*()`, `scale_fill_*()`

- manual 
- continuous 
- brewer/distiller/fermenter
- gradient/gradient2/gradientn
- steps
- viridis 
]

---
## Continuous color scales

Default continuous colour scheme

```r
p1 <- ggplot(mpg, aes(x = displ, y = hwy, colour = cyl)) +
 geom_point()
```

```r
p1 + scale_colour_continuous()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-10-1.png)

---
## Discrete Color Scales

Default discrete colour scheme

```r
p2 <- ggplot(mpg, aes(x = displ, y = hwy, colour = factor(cyl))) +
 geom_point()
```

```r
p2 + scale_colour_discrete()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-12-1.png)

---
## Color and Fill

- Area geoms (barcharts, histograms, polygons) use `fill` to map values to the fill color

- only discrete color scales can be used:  `scale_fill_hue`, `scale_fill_brewer`, `scale_fill_grey`, ...

- most general: `scale_fill_manual(..., values)`

- `values` is a vector of color values.  
  
  - At least as many colors as levels in the variable have to be listed

**Color Values**:

- can be defined as a hex value or a name of a color
  
  - [R colors pdf](http://www.stat.columbia.edu/~tzheng/files/Rcolor.pdf)

---
## Colour gradients

**Sequential schemes**:

`scale_colour_gradient(..., low = "#132B43", high = "#56B1F7", space = "Lab", na.value = "grey50", guide = “colourbar")`

- colors can be specified by hex code, name or through rgb()

- gradient goes from low to high - that should match the interpretation of the mapped variable

**Divergent schemes**:

`scale_colour_gradient2(..., low = muted("red"), mid = "white",  
high = muted("blue"), midpoint = 0, space = "Lab", na.value = "grey50",  guide = "colourbar")`

- midpoint is value of the ‘neutral’ color gradient2 is a divergent color scheme

- best matches a variable that goes from large negative to zero to large positive (or below mean, above mean)

---
## Discrete color schemes

**Manually created**:

`scale_color_hue (..., h = c(0, 360) + 15, c = 100, l = 65, h.start = 0, direction = 1, na.value = "grey50")`

- uses hue, chroma and luminance (=value)

- each level of a variable is assigned a different level of h

**ColorBrewer**:

`scale_colour_brewer(..., type = "seq", palette = 1, direction = 1)`

- brewer schemes are defined in RColorBrewer (Neuwirth, 2014) palettes can be specified by name or index

- see also http://colorbrewer2.org/ (Brewer et al 2002)

---
## All brewer schemes

```r
library(RColorBrewer)
display.brewer.all()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-13-1.png)

---
background-image: url(https://www.datanovia.com/en/wp-content/uploads/dn-tutorials/ggplot2/figures/030-ggplot-colors-wes-anderson-color-palettes-r-1.png)
background-size: 325px
background-position: 85% 75%
## Predefined ggplot color palettes

.pull-left[
The most commonly used color scales, include:

- Viridis color scales [[viridis package](https://cran.r-project.org/web/packages/viridis/vignettes/intro-to-viridis.html)]

- Colorbrewer palettes [[RColorBrewer package](http://colorbrewer2.org/)]

- Scientific journal color palettes [[ggsci package](https://cran.r-project.org/web/packages/ggsci/vignettes/ggsci.html)]

- Wes Anderson color palettes [[wesanderson package](https://github.com/karthik/wesanderson/blob/master/README.md)]

For the most extensive list I've found, look [here](https://github.com/EmilHvitfeldt/r-color-palettes)
]
---
## Legends

The **guide** or **legend** connects non-axis aesthetics in the data visualization like color and size to the data

The `guides()` function controls all legends by connecting to the aes.

- `guide_colorbar()`: continuous colors

- `guide_legend()`: discrete values (shapes, colors)

- `guide_axis()`: control axis text/spacing, add a secondary axis

- `guide_bins()`: creates "bins" of values in the legend

- `guide_colorsteps()`: makes colorbar discrete

---
## Themes

The **theme** describes the appearance of the plot, such as the background color, font size, positions of labels, etc.

Specific themes:

- `theme_grey()`: default

- `theme_bw()`: white background, gray gridlines

- `theme_classic()`: looks more like base R plots

- `theme_void()`: removes all background elements, all axes elements, keeps legends

---
## theme Examples

```r
p <- ggplot(mpg, aes(x = displ, y = cty, colour= factor(class))) + geom_point()
```

.pull-left[

```r
p + theme_grey()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-15-1.png)
]
.pull-right[

```r
p + theme_minimal()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-16-1.png)
]

---
## theme Examples

```r
p <- ggplot(mpg, aes(x = displ, y = cty, colour= factor(class))) + geom_point()
```

.pull-left[

```r
p + theme_light()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-18-1.png)
]
.pull-right[

```r
p + theme_dark()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-19-1.png)
]

---
## More themes

```r
library(ggthemes)
```

.pull-left[

```r
p + theme_excel() 
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-21-1.png)
]
.pull-right[

```r
p + theme_fivethirtyeight()
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-22-1.png)
]

---
## Theme customization

The `theme()` function can modify any non-data element of the plot.

- adjust the appearance every "non-data element" of the viz

- fonts, background, text positioning, legend appearance, facet appearance, etc.

**Rule of thumb**: when changing an element that shows data, use aes() and scales. Otherwise, use themes.

---
## Elements of themes

- **Line elements**: axis lines, minor and major grid lines, plot panel border, axis ticks background color, etc.

- **Text elements**: plot title, axis titles, legend title and text, axis tick mark labels, etc.

- **Rectangle elements**: plot background, panel background, legend background, etc.

There is a specific function to modify each of these three elements :

- `element_line()` to modify the line elements of the theme

- `element_text()` to modify the text elements

- `element_rect()` to change the appearance of the rectangle elements

- `element_blank()` to draw nothing and assign no space
  
---
## Elements of themes

- Axis: `axis.line`, `axis.text.x`, `axis.text.y`, `axis.ticks`, `axis.title.x`, `axis.title.y`

- Legend: `legend.background`, `legend.key`, `legend.text`, `legend.title`

- Panel: `panel.background`, `panel.border`, `panel.grid.major`, `panel.grid.minor`

- Strip (facetting): `strip.background`, `strip.text.x`, `strip.text.y`

For a complete overview see `?theme`

---
## Changing elements manually

- to change an element add the theme function and specify inside:

```r
ggplot(mpg, aes(x = manufacturer)) + 
  geom_bar() +
  theme(axis.text.x = element_text(angle=90, vjust=0.5, hjust=1))
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-23-1.png)

---

## Saving your Work

We can save the results of a plot to a file (as an image) using the `ggsave()` function:

```r
iris_plot <- ggplot(data = iris, aes(x = Sepal.Width, y = Sepal.Length, color = Species)) + geom_point() +
 scale_color_manual(values = c("turquoise3", "green3", "darkorchid2")) + 
 theme(legend.position = "bottom", legend.background = element_blank())
iris_plot 
```

![](01_polishing-plots_files/figure-html/unnamed-chunk-24-1.png)

```r
ggsave("iris-scatter.png", plot = iris_plot)
```

---
## Resources

- reference/document: http://ggplot2.tidyverse.org/reference/

- RStudio cheat sheet for [ggplot2](https://www.rstudio.com/wp-content/uploads/2015/03/ggplot2-cheatsheet.pdf)

- Artwork by [@allison_horst](https://twitter.com/allison_horst?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor)

- Some material borrowed from [Sam Tyner](https://sctyner.github.io/user20-proposal.html)'s ggplot2 workshop