Question

I need to do a simple data conversion in R for usage with igraph. My dataframe is in this format, grouped by GROUP:

    A   GROUP
1   1       a
2   2       a
3   3       a
4   4       a
5   1       b
6   3       b
7   5       b

1. How do I expand the groups to get an undirected edgelist el in this format?

    A   B
1   1   2
2   1   3
3   1   4
4   2   3
5   2   4
6   3   4
7   1   3
8   1   5
9   3   5

Note: no self-references 1-1, 2-2, 3-3, …

2. How do I count A-B occurrences and create a weighted edgelist from el?

    A   B   weight
1   1   2        1
2   1   3        2
3   1   4        1
4   2   3        1
5   2   4        1
6   3   4        1
7   1   5        1
8   3   5        1

Answer 1

Here is a way to get the edgelist with plyr:

foo <- data.frame(
  A = c(1,2,3,4,1,3,5),   
  GROUP = c("a","a","a","a","b","b","b"))

library("plyr")

E1 <- do.call(rbind,dlply(foo,.(GROUP),function(x)t(combn(x$A,2))))

E1

Returns:

      [,1] [,2]
 [1,]    1    2
 [2,]    1    3
 [3,]    1    4
 [4,]    2    3
 [5,]    2    4
 [6,]    3    4
 [7,]    1    3
 [8,]    1    5
 [9,]    3    5

Then to get the weights (here I use that combn puts the lowest number first):

W <- apply(E1,1,function(x)sum(E1[,1]==x[1]&E1[,2]==x[2]))
E2 <- cbind(E1,weight=W)
E2 <- E2[!duplicated(E2),]

E2

Which returns:

         weight
[1,] 1 2      1
[2,] 1 3      2
[3,] 1 4      1
[4,] 2 3      1
[5,] 2 4      1
[6,] 3 4      1
[7,] 1 5      1
[8,] 3 5      1