/ggplot2_quick_lookup

Quick lookup to ggplot2 commands rather than StackOverflow'ing' every time

ggplot2 lookup guide

Packages

To check the list of all possible ggplot libraries look [here]{https://exts.ggplot2.tidyverse.org/gallery/}

  1. ggsci - To get Scientific Journal specific colour schemes
  2. ggfortify - To make PCA plots
  3. ggpubr, ggstatsplot - To attach stats to plots, and also to arrange multiple plots together
  4. patchwork - Quickly combine plots together
  5. corrplot, pheatmap - To make correlation plots, hierarchical clustering heatmaps
  6. GGally - To make ggpairs plots
  7. ggtranscript
  8. ggrepel - to repel annoying plot labels
  9. ggalluvial - To make sankey diagrams
  10. ggupset, ComplexUpset - To make upset plots

Template from ggplot2 cheatsheet

ggplot(data = <DATA>) +
     <GEOM_FUNCTION>(mapping = aes( <MAPPINGS>),
                     stat = <STAT>,
                     position = <POSITION>) +
     <COORDINATE_FUNCTION> +
     <FACET_FUNCTION> +
     <SCALE_FUNCTION> +
     <THEME_FUNCTION>

Plots

Scatter plot

geom_point(data = df1, aes(x = a, y = b))

Histogram

geom_hist(data = df1, aes(x = a))

Boxplot

Diverging bars

Heatmap

Dumbbell plot

Upset plot

Locus Zoom plot

plotGardener plot

Sankey diagram

Manhattan plot

Nightingale plot

ggpairs plot

Plot labels and addups

Title, xlabel, ylabel

labs(title = "This is the ggplot2 lookup guide!", subtitle = "GitHub", caption = "Source: ggplot2",
     tag = "A", x = "Command", y = "Function")

Fill area of data points with colour

geom_area(df1)

Draw vline to indicate important value

geom_vline(xintercept = 0.5, linetype = "dotted")

Connect all data points with a line

geom_line(df1)

Defining axis scales

  1. limits is the scale limit to be plotted/shown on graph
  2. breaks is the numbers to be displayed in the axis
  3. scales::comma displays 1000 as 1,000 and 10000 as 10,000, scales::percent displays 95 as 95%
  4. expand is the padding beyoond the limit, for e.g. right padding is 10000 + (10000 - 1)*1 + 2, left padding is 0 + (0-1)*1 -2
scale_x_continuous(limits = c(0,10000), breaks = c(0, 100, 1000, 10000), labels = scales::comma, expand = (1, 2)) +
scale_y_continuous(limits = c(0, 100), labels = scales::percent)

Theme

theme_set(theme_minimal(), legnend.position = "bottom", legend.title = "Tutorial")

Multi plots

Divide the data into multiple datasets based on the factor of any given column

facet_wrap(~ DxCondition, ncol = 1)

Combining multiple different plots

library(patchwork)
plot1 + plot2 + plot3 + plot4

Adding p-values to graphs

https://stackoverflow.com/questions/45476950/r-ggplot2-boxplots-ggpubr-stat-compare-means-not-working-properly