Re: [igraph] subsetting by vertex ids and keep all edges

[Gábor Csárdi]

Indeed. Btw. unlist(a) is much simpler than Reduce(c, a).

G.

On Wed, Feb 5, 2014 at 9:50 AM, Hermann Norpois <address@hidden> wrote:

I guess I found a satiesfieing solution:

a <- neighborhood (g,1,zu)
idx <- unique (Reduce (c, a))
ggg <- induced.subgraph (g, idx)

Hermann

2014-02-04 Hermann Norpois <address@hidden>:

Hello,

I have an igraph object g and I want to subset g by using vertex ids, for instance by

 zu
[1] "16052618"  "rs3013006".

I want to keep all edges of each defined vertix, also the edges my vertices share with other vertices. So induced.subgraph is not an option, isn't it?

I tried subgraph.edges but I guess my approach is very clumsy. So I ask for an amelioration.

I tried:

idx <- unique (as.vector (g[zu, edges=TRUE])) # get edges ids for vertices of interest
idx <- idx[-1] # remove 0
ggzu <- subgraph.edges (g, idx)

Do you have any suggestions?
Thanks
Hermann



dput (g)
structure(list(9, FALSE, c(6, 2, 7, 8, 7, 5), c(0, 1, 1, 1, 2,
3), c(1, 5, 0, 2, 4, 3), c(0, 1, 2, 3, 4, 5), c(0, 0, 0, 1, 1,
1, 2, 3, 5, 6), c(0, 1, 4, 5, 6, 6, 6, 6, 6, 6), list(c(1, 0,
1), structure(list(), .Names = character(0)), structure(list(
    name = c("16052361", "16052513", "16052618", "16052962",
    "rs11703994", "rs62224614", "rs78724352", "rs3013006", "rs55926024"
    ), x = c(16052361L, 16052513L, 16052618L, 16052962L, 16053791L,
    16053862L, 16053863L, 16054667L, 16054740L), y = c(2.156147,
    2.547104, 1.587586, 4.037532, 3.734964, 5.026869, 2.514876,
    1.543943, 0.579408)), .Names = c("name", "x", "y")), structure(list(
    weight = c(0.526203, 0.725426, 0.681783, 0.579408, 0.86216,
    0.718093), color = c("grey", "blue", "grey", "grey", "red",
    "blue")), .Names = c("weight", "color")))), class = "igraph")

    dput (zu)
c("16052618", "rs3013006")

_______________________________________________
igraph-help mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/igraph-help