CODEWITHSUNDEEP
rgraphigraph
saamar rajput
saamar rajput

Reputation: 75

Edge conditions in igraph in R

I am working on igraph trying to make my own visualization of the graph taken from KEGG xml. I have retreived an adjacency matrix for the graph and an edge list.Now I have a few conditions for my edges for example I have inhibition, activation and binding association (not weighted). Now I want to color the edges differently and I also want the shape of the edges differently for each condition. For example an edge with arrow and green in color for activation. An edge with a vertical lie after edge and red in color for inhibition. And maybe a dotted line for binding association.

My edge List named reactions looks like this

    > entry1    entry2  name 

    > 59         62     activation
    > 62         57     Inhibition
    > 61         60     binding association
    > 53         42     activation

My nodes are in the form of an directed adjacency matrix.

  plot(G,vertex.shape= "rectangle", edge.arrow.size=.3, edge.color=ifelse(reactions$name =="activation", "green", "red"),vertex.color="gold", vertex.size2=1,vertex.frame.color="gray", vertex.label.color="black", vertex.label.cex=1, vertex.label.dist=0.5, edge.curved=0.2)

I was just tryin to check if the code works for activation first and then i deal with other conditions but all my edges are green not just the activation ones.

Could some body help me with this. I tried using edge.color with ifelse but don't really know how to use it.

Upvotes: 2

Views: 1471

Answers (1)

Michael Davidson
Michael Davidson

Reputation: 1411

Instead of referencing the variable with your conditions in the data frame, try referencing the data in the graph itself. Here I can recreate your graph:

library(igraph)

d <- data.frame(ego=c('59', '62', '61', '53'), alter=c('62', '57', '60', '42'), status=c("activation","Inhibition","binding association","activation"))

G <- graph_from_data_frame(d)

Lets see what the graph includes:

str(G)

The (e/c) is telling us that status is a characteristic of the edges, just what we want to condition the color and or shape of the edges.

plot(G)

plot(G, vertex.shape= "rectangle", edge.arrow.size=.3, edge.color=ifelse(E(G)$status =="activation", "green", "red"), vertex.color="gold", vertex.size2=1, vertex.frame.color="gray", vertex.label.color="black", vertex.label.cex=1, vertex.label.dist=0.5, edge.curved=0.2)

Upvotes: 3

Related Questions