Directed Acyclic Graphs

Authors

[Editor] Hu Zheng;

[Contributors]

Note

Hiplot website

This page is the tutorial for source code version of the Hiplot Directed Acyclic Graphs plugin. You can also use the Hiplot website to achieve no code ploting. For more information please see the following link:

https://hiplot.cn/basic/ggdag?lang=en

Visualizing directed acyclic graphs.

Setup

System Requirements: Cross-platform (Linux/MacOS/Windows)
Programming language: R
Dependent packages: ggdag

# Install packages
if (!requireNamespace("ggdag", quietly = TRUE)) {
  install.packages("ggdag")
}

# Load packages
library(ggdag)

Data Preparation

# Load data
tidy_ggdag <- dagify(
  y ~ x + z2 + w2 + w1,
  x ~ z1 + w1 + w2,
  z1 ~ w1 + v,
  z2 ~ w2 + v,
  w1 ~ ~w2, # bidirected path
  exposure = "x",
  outcome = "y") %>%
  tidy_dagitty()

# View data
head(tidy_ggdag)

$data
# A tibble: 13 × 8
   name       x        y direction to      xend   yend circular
   <chr>  <dbl>    <dbl> <fct>     <chr>  <dbl>  <dbl> <lgl>   
 1 v      1.50   0.00901 ->        z1     0.517 -1.00  FALSE   
 2 v      1.50   0.00901 ->        z2     0.504  1.01  FALSE   
 3 w1    -0.320 -0.464   ->        x     -0.932 -0.603 FALSE   
 4 w1    -0.320 -0.464   ->        y     -0.329  0.460 FALSE   
 5 w1    -0.320 -0.464   ->        z1     0.517 -1.00  FALSE   
 6 w1    -0.320 -0.464   <->       w2    -0.942  0.590 FALSE   
 7 w2    -0.942  0.590   ->        x     -0.932 -0.603 FALSE   
 8 w2    -0.942  0.590   ->        y     -0.329  0.460 FALSE   
 9 w2    -0.942  0.590   ->        z2     0.504  1.01  FALSE   
10 x     -0.932 -0.603   ->        y     -0.329  0.460 FALSE   
11 y     -0.329  0.460   <NA>      <NA>  NA     NA     FALSE   
12 z1     0.517 -1.00    ->        x     -0.932 -0.603 FALSE   
13 z2     0.504  1.01    ->        y     -0.329  0.460 FALSE   

$dag
dag {
v
w1
w2
x [exposure]
y [outcome]
z1
z2
v -> z1
v -> z2
w1 -> x
w1 -> y
w1 -> z1
w1 <-> w2
w2 -> x
w2 -> y
w2 -> z2
x -> y
z1 -> x
z2 -> y
}

Visualization

# Directed Acyclic Graphs
p <- ggdag(tidy_ggdag) +
  theme_dag() 

p