I want to draw and edit networks. Currently I will do something like this -- my actual networks are larger and more complex: 1. Plot network # use polygon with nodes as example network N <- 20 t. <- 2*pi*seq(N+1)/N plot(cos(t.), sin(t.), type="b", pch=19, cex=5, col="blue", axes=F, ann=F) 2. copy and paste the graphic as a metafile into Word. 3. edit the network there. The editing typically just involves moving nodes and the attached edges. Of course, Word does not know the edges are attached to the nodes so I have to move the nodes and then move the attached edges, as well. The networks are too complex to create the diagrams by hand which is why I want to create them in R and then edit them by hand. The above strategy is marginally acceptable but I was wondering if anyone had a better strategy which retains the simplicity of the present approach but gives me the capability to have the edges move along with attached nodes when editing them. (Note that the strategy must consist of generating the network under program control, preferably in R, and then editing it afterwards. I am not looking to create the entire network by hand since the networks are too complex for that. I had tried specifying the networks in dot from Bell Labs but 1. its still more work than generating the network in R 2. the automatic layout it chooses is so far from what I need that its a lot of work to fix it up by hand in dotty and 3. I could never get them to look just how I like. I have also tried Excel and Powerpoint, both of which support connectors, which are edges that move along with attached nodes, but to use them I have to create the entire network by hand and the networks have grown too complex to do that.)
Gabor Grothendieck wrote:> I want to draw and edit networks. Currently I will do something like this -- > my actual networks are larger and more complex: > > 1. Plot network > > # use polygon with nodes as example network > N <- 20 > t. <- 2*pi*seq(N+1)/N > plot(cos(t.), sin(t.), type="b", pch=19, cex=5, col="blue", axes=F, ann=F) > > 2. copy and paste the graphic as a metafile into Word. > > 3. edit the network > there. The editing typically just involves moving nodes and the attached > edges. Of course, Word does not know the edges are attached to the nodes so I > have to move the nodes and then move the attached edges, as well. > > The networks are too complex to create the diagrams by hand which is why > I want to create them in R and then edit them by hand. > > The above strategy is marginally acceptable but > I was wondering if anyone had a better strategy which retains the > simplicity of the present approach but gives me the capability to > have the edges move along with attached nodes when editing them. > > (Note that the strategy must consist of generating the network under program > control, preferably in R, and then editing it afterwards. I am not looking > to create the entire network by hand since the networks are too complex for > that. I had tried specifying the networks in dot from Bell Labs but 1. its > still more work than generating the network in R 2. the automatic > layout it chooses is so far from what I need that its a lot of work to > fix it up by hand in dotty and 3. I could never get them to look just how > I like. I have also tried Excel and Powerpoint, both > of which support connectors, which are edges that move along with attached > nodes, but to use them I have to create the entire network by hand and > the networks have grown too complex to do that.) > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.htmlHave you tried the CRAN package "dynamicGraph" by Jens Henrik Badsberg? Uwe Ligges
Or graph, RBGL and Rgraphviz from www.bioconductor.org.... On Sun, May 09, 2004 at 02:35:02PM +0200, Uwe Ligges wrote:> Gabor Grothendieck wrote: > > I want to draw and edit networks. Currently I will do something like this -- > > my actual networks are larger and more complex: > > > > 1. Plot network > > > > # use polygon with nodes as example network > > N <- 20 > > t. <- 2*pi*seq(N+1)/N > > plot(cos(t.), sin(t.), type="b", pch=19, cex=5, col="blue", axes=F, ann=F) > > > > 2. copy and paste the graphic as a metafile into Word. > > > > 3. edit the network > > there. The editing typically just involves moving nodes and the attached > > edges. Of course, Word does not know the edges are attached to the nodes so I > > have to move the nodes and then move the attached edges, as well. > > > > The networks are too complex to create the diagrams by hand which is why > > I want to create them in R and then edit them by hand. > > > > The above strategy is marginally acceptable but > > I was wondering if anyone had a better strategy which retains the > > simplicity of the present approach but gives me the capability to > > have the edges move along with attached nodes when editing them. > > > > (Note that the strategy must consist of generating the network under program > > control, preferably in R, and then editing it afterwards. I am not looking > > to create the entire network by hand since the networks are too complex for > > that. I had tried specifying the networks in dot from Bell Labs but 1. its > > still more work than generating the network in R 2. the automatic > > layout it chooses is so far from what I need that its a lot of work to > > fix it up by hand in dotty and 3. I could never get them to look just how > > I like. I have also tried Excel and Powerpoint, both > > of which support connectors, which are edges that move along with attached > > nodes, but to use them I have to create the entire network by hand and > > the networks have grown too complex to do that.) > > > > ______________________________________________ > > R-help at stat.math.ethz.ch mailing list > > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > > > Have you tried the CRAN package "dynamicGraph" by Jens Henrik Badsberg? > > Uwe Ligges > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html-- +---------------------------------------------------------------------------+ | Robert Gentleman phone : (617) 632-5250 | | Associate Professor fax: (617) 632-2444 | | Department of Biostatistics office: M1B20 | | Harvard School of Public Health email: rgentlem at jimmy.harvard.edu | +---------------------------------------------------------------------------+