CODEWITHSUNDEEP
rigraph
Aline
Aline

Reputation: 347

How to convert an igraph ego() list into a meaningful dataframe?

I have the following network:

edges <- data.frame(from=c('a','1','2','b'), to=c('1','2','3','4'))
edges
   from to
1    a  1
2    1  2
3    2  3
4    b  4

And want to identify the neighborhood of "to" nodes in the following way:

1 a NA NA
2 1 a  NA
3 2 1  a
4 b NA NA

Ideally I would get only

3 2 1  a
4 b NA NA

because I am only interested in getting the full path from 'a' to '3' and '4' to 'b'.

Using the ego() function from the igraph package I get a list with this information but I have not managed to convert it into a dataframe in the form above:

test <- ego(graph,4,edges[,2], "in")
test
[[1]]
+ 1/6 vertex, named:
[1] a

[[2]]
+ 2/6 vertices, named:
[1] 1 a

[[3]]
+ 3/6 vertices, named:
[1] 2 1 a

[[4]]
+ 1/6 vertex, named:
[1] b

Here are my unsuccessful trials:

require(plyr)
> data.frame(ldply(test, rbind))
   a X1 X2  b
1  1 NA NA NA
2  1  2 NA NA
3  1  2  3 NA
4 NA NA NA  4

data.frame(t(unlist(test)))
a X1 a.1 X2 X1.1 a.2 b
1 1  2   1  3    2   1 4

Upvotes: 1

Views: 452

Answers (2)

Sandipan Dey
Sandipan Dey

Reputation: 23099

We could use rbind.fill too:

library(plyr)
rbind.fill(lapply(test, function(x)as.data.frame(t(names(x)))))
#  V1   V2   V3
#1  a <NA> <NA>
#2  1    a <NA>
#3  2    1    a
#4  b <NA> <NA>

Upvotes: 2

Aline
Aline

Reputation: 347

Ok, I've found a way of doing it, but I am not sure if it is the most efficient.

neighbors <- lapply(1:length(test), function(x) names(test[[x]]))
require(plyr)
data.frame(edges$to,ldply(neighbors, rbind))
  edges.to X1   X2   X3
1        1  a <NA> <NA>
2        2  1    a <NA>
3        3  2    1    a
4        4  b <NA> <NA>

Upvotes: 0

Related Questions