Re: [igraph] Representation of Bipartite Networks

[Lorenzo Isella]

> Date: Sat, 6 Mar 2010 00:42:37 +0100 From: G?bor Cs?rdi 
> <address@hidden> Subject: Re: [igraph] Representation of 
> Bipartite Networks To: Help for igraph users <address@hidden> 
> Message-ID: 
> <address@hidden> 
> Content-Type: text/plain; charset=ISO-8859-1 Lorenzo, I don't know 
> what is recommended. We don't have any special layout algorithms for 
> bipartite networks, AFAIK. Maybe it is possible to work out something 
> meaningful using positive and negative edge weights in the 
> Fruchterman-Reingold algorithm. My idea would be to create the union 
> of the original and a full graph. In the full graph, vertices of 
> different type would repulse each other (negative weight on the edge) 
> and vertices of the same type would attract each other. Whether this 
> gives anything meaningful, I have no idea. Maybe it is enough to 
> simply use different vertex shapes for both vertex types. Best, Gabor 
> On Fri, Mar 5, 2010 at 3:57 PM, Lorenzo Isella 
> <address@hidden> wrote:
> > > Dear All,
> > > I am actually dealing with bipartite networks.
> > > What is the best way (if there is any) to plot a bipartite network?
> > > Is there any recommended layout in igraph?
> > > Cheers
> > >
> > > Lorenzo
> > >

Hi Gabor,
Well, let us say that I have a bipartite network, where I have nodes of 
two colors (blue and red): I would like to have e.g. all the red nodes 
on the left (possibly in a column) and all the blue nodes on the right.
See below an example with the fruchterman-reingold layout (which clearly 
is not what I want).
Any idea of how to get the job done? On top of that: where do I find a 
list of all the different vertex shapes available in igraph?
Many thanks

Lorenzo

library(igraph)
library(Cairo)



set.seed(1234)


nodes_1 <- round(runif(min=1000,max=1020, 200))
nodes_2 <- round(runif(min=2000,max=2020, 200))

tab <- cbind(as.data.frame(nodes_1), as.data.frame(nodes_2))

g <- graph.data.frame(as.data.frame(tab), dir=FALSE)

g <- simplify(g)

sel_1 <- which(as.numeric(V(g)$name)>=1000 & as.numeric(V(g)$name)<2000)

sel_2 <- which(as.numeric(V(g)$name)>=2000)

V(g)[sel_1-1]$color <- "red"
V(g)[sel_2-1]$color <- "blue"


l <- layout.fruchterman.reingold(g)
l <- layout.norm(l, -1,1, -1,1)


CairoPDF("bipartite_network.pdf")

plot(g, layout=l,
    vertex.label.dist=0.5,vertex.shape="rectangle",
    vertex.frame.color="#ff000033", 
edge.color="#55555533",vertex.label=NA, vertex.size=4
    )
dev.off()