Directed acyclic diagrams (DAG) with R

In preparation for my comprehensive exam I have been looking for easy and fast ways to draw causal diagrams. In LaTeX this can be achieved with some effort using the TikZ package, some examples are this and this.

I am planning to write the report for the data analysis portion of my exam using Rmarkdown, so I want to draw the causal diagram directly from R. Enter the DiagrammeR package.

“With the DiagrammeR package, you can create graph diagrams and flowcharts using R. Markdown-like text is used to describe a diagram and, by doing this in R, we can also add some R code into the mix and integrate these diagrams in the R console, through R Markdown, and in shiny web apps.”

A short Example using DiagrammeR

The DiagrammeR package let us access the drawing capacities of mermaid and graphViz.

For this example I will be reproducing Example 1.4 “A causal diagram of factors affecting fertility in cows”, from the textbook by Dohoo, I., W. Martin, and H. Stryhn. “Veterinary epidemiologic research.” (2009).

Let’s start by using mermaid:

	library(DiagrammeR)

	mermaid("
	        graph LR
	        A(Age)-->F(Fertility)
	        A-->O(Cistic ovarian <br> disease)
	        A-->R(Retained <br> placenta)
	        R-->O
	        R-->M(Metritis)
	        M-->O
	        O-->F
	        M-->F
	        ")

That looks good but since I am trying to reproduce what is in the book, I want to get rid of the boxes, for this we define a style for each node (white fill and white border).

	        mermaid("
	        graph LR
	        A(Age)-->F(Fertility)
	        A-->O(Cistic ovarian <br> disease)
	        A-->R(Retained <br> placenta)
	        R-->O
	        R-->M(Metritis)
	        M-->O
	        O-->F
	        M-->F
	
	        style A fill:#FFFFFF, stroke-width:0px
	        style R fill:#FFFFFF, stroke-width:0px
	        style M fill:#FFFFFF, stroke-width:0px
	        style O fill:#FFFFFF, stroke-width:0px
	        style F fill:#FFFFFF, stroke-width:0px
	        ")

The same can be achieved with the flexible and nicer but more complicated graphViz:

	grViz("
	digraph causal {
	
	  # Nodes
	  node [shape = plaintext]
	  A [label = 'Age']
	  R [label = 'Retained\n Placenta']
	  M [label = 'Metritis']
	  O [label = 'Cistic ovarian\n disease']
	  F [label = 'Fertility']
	  
	  # Edges
	  edge [color = black,
	        arrowhead = vee]
	  rankdir = LR
	  A->F
	  A->O
	  A->R
	  R->O
	  R->M
	  M->O
	  O->F
	  M->F
	  
	  # Graph
	  graph [overlap = true, fontsize = 10]
	}")

I got a very similar, yet a reflection of the original figure in the book.

Conclusion

While graphViz allows for a lot of customization I think I will stick with the simple syntax of mermaid for the report. For publication however, I think graphViz is a serious contender for TikZ. Just visit the DiagrammeR docs page and learn how to draw very fancy diagrams without leaving the R environment.

Thanks for reading!

Share this: