1

what data format is necessary to create ggraph circlepack layout? It seems to require a hierarchy.

I tried normally vertices and nodes, which apparently doesn't work.

library(ggraph)
library(igraph)
edges=data.frame(from=c('a','b','c'), to= c('c','a','d'))
vertices=data.frame(nodes=c('a','b','c','d'), weight=c(1,2,3,4))
graph <- graph_from_data_frame(edges, vertices = vertices)
ggraph(graph, 'circlepack', weight = 'size') + 
geom_node_circle(size = 0.25, n = 50) + 
coord_fixed()

I then tried a dendrogram object which doesn't work either. If i want to show several groups with sub-items in packed circle, how shall i build the graph object?

the data frame is more like this

df <- data.frame(group=c("a","a","b","b","b"),    subitem=c("x","y","z,"u","v"), size=c(6,2,3,2,5))
rafa.pereira
  • 13,251
  • 6
  • 71
  • 109
santoku
  • 3,297
  • 7
  • 48
  • 76
  • 1
    What is the `graph` object you're passing to `ggraph`? – yeedle May 02 '17 at 14:02
  • that's what i'm trying to figure out. for the df in the last row, how can i make it into a graph object that accepts the circle layout? thanks. – santoku May 02 '17 at 15:14

1 Answers1

1

A circlepack layout models a hierarchical/tree-like structure with one root and no cycles. To model your df as a circlepack layout, you have to consider that a and b in the group column are both roots. If we add a root to the df, and have both a and b be children of that root, we can visualize it as a circlepack:


library(ggraph)
library(igraph)
library(dplyr)


df <- data.frame(group=c("root", "root", "a","a","b","b","b"),    
                 subitem=c("a", "b", "x","y","z","u","v"), 
                 size=c(0, 0, 6,2,3,2,5))

# create a dataframe with the vertices' attributes

vertices <- df %>% 
  distinct(subitem, size) %>% 
  add_row(subitem = "root", size = 0)

graph <- graph_from_data_frame(df, vertices = vertices)

ggraph(graph, layout = "circlepack", weight = 'size') + 
  geom_node_circle(aes(fill =depth)) +
# adding geom_text to see which circle is which node 
  geom_text(aes(x = x, y = y, label = paste(name, "size=", size))) +
  coord_fixed()

yeedle
  • 4,918
  • 1
  • 22
  • 22