Re: [igraph] Question about igraph and graph.data.frame in R

[Tamas Nepusz]

Hi Peter,

> My data frame (named forclus) has 796 rows (people) and 524 columns  
> (variables).
> I can create a distance matrix using the dist function:
>
> dist <- as.matrix(dist(forclus))
>
> I can create names for the rows using row.names(forclus).
>
> But in your documentation you list a rather small data frame, with  
> three nodes (foo,
> bar, and foobar), 2 edges, and the weights entered by hand.

> Do you have any suggestions as to how I can go from my large data  
> frame to a graph,
> and then to a minimum spanning tree?
I'm not particularly familiar with the R interface, but maybe  
graph.adjacency() helps:

http://cneurocvs.rmki.kfki.hu/igraph/doc/R/graph.adjacency.html

This function creates a graph based on a weighted adjacency matrix,  
which (according to my understanding) is exactly the distance matrix  
what you have. Vertex names can be assigned later to the graph as a  
vertex attribute.

Regards,
--
Tamas