Drawing network graphs (i.e. nodes and edges)
2
0
Entering edit mode
Peter ▴ 170
@peter-1556
Last seen 10.2 years ago
Does anyone know of any existing R (r-project) code/library/package for drawing graphs - the network variety consisting of nodes and edges (or points and lines/arrows)? e.g. Genes and interactions For example, I might have a list of three edges for four genes: Gene1 -> Gene2 Gene2 -> Gene3 Gene4 -> Gene3 Which would give the following simple graph: Gene1 -> Gene2 -> Gene3 <- Gene4 Or perhaps as a matrix, which for four genes would be 4 by 4: [ 0 1 0 0 ] [ 0 0 1 0 ] [ 0 0 0 0 ] [ 0 0 1 0 ] Here I have used the "rows" as the "from" and the columns as the "to" connections. I was hoping there was an existing library that would accept this sort of network definition and draw a picture of it. Maybe labelled circles for the nodes (genes) with arrows for the edges (interactions). The hard part of this is the spatial layout of the nodes to avoid the edges intersecting unrelated nodes. One simple answer is to draw the nodes equally spaces on a large circle. Apart from the special case of self interactions, all the edges would be inside the circle and thus never cross another node. e.g. This sort of image: http://www.cs.toronto.edu/~wayne/research/ramsey/r.3.9.35.gif Of course there is the extra gloss of being able to draw a connection weight using line thickness etc. Any ideas? I haven't found anything on Google... Thank you Peter
Network Network • 2.1k views
ADD COMMENT
0
Entering edit mode
@sean-davis-490
Last seen 3 months ago
United States
Look at Rgraphviz in bioconductor--it does just this. Sean On 2/2/06 10:57 AM, "Peter" <bioconductor-mailinglist at="" maubp.freeserve.co.uk=""> wrote: > Does anyone know of any existing R (r-project) code/library/package for > drawing graphs - the network variety consisting of nodes and edges (or > points and lines/arrows)? > > e.g. Genes and interactions > > For example, I might have a list of three edges for four genes: > > Gene1 -> Gene2 > Gene2 -> Gene3 > Gene4 -> Gene3 > > Which would give the following simple graph: > > Gene1 -> Gene2 -> Gene3 <- Gene4 > > Or perhaps as a matrix, which for four genes would be 4 by 4: > > [ 0 1 0 0 ] > [ 0 0 1 0 ] > [ 0 0 0 0 ] > [ 0 0 1 0 ] > > Here I have used the "rows" as the "from" and the columns as the "to" > connections. > > I was hoping there was an existing library that would accept this sort > of network definition and draw a picture of it. Maybe labelled circles > for the nodes (genes) with arrows for the edges (interactions). > > The hard part of this is the spatial layout of the nodes to avoid the > edges intersecting unrelated nodes. One simple answer is to draw the > nodes equally spaces on a large circle. Apart from the special case of > self interactions, all the edges would be inside the circle and thus > never cross another node. > > e.g. This sort of image: > > http://www.cs.toronto.edu/~wayne/research/ramsey/r.3.9.35.gif > > Of course there is the extra gloss of being able to draw a connection > weight using line thickness etc. > > Any ideas? I haven't found anything on Google... > > Thank you > > Peter > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor
ADD COMMENT
0
Entering edit mode
Peter wrote: >> Does anyone know of any existing R (r-project) code/library/package >> for drawing graphs - the network variety consisting of nodes and >> edges (or points and lines/arrows)? >> >> e.g. Genes and interactions Sean Davis wrote: > Look at Rgraphviz in bioconductor--it does just this. That looks like just the thing - thank you. It occurred to me that there are probably a lot gems like this tucked away in the PDF vignettes. I suspect that if they could be provided on the BioConductor webpage as normal HTML files, then Google and other search engines would be able to search them far more effectively - in particular for their images. As one of R's strengths is graphics, it would make sense to make samples of its output as easy to find as possible - and for me that means Google Image search. Anyway - Thanks again Sean, Peter
ADD REPLY
0
Entering edit mode
@saurin-d-jani-944
Last seen 10.2 years ago
Hi Peter, You can try http://graphexploration.cond.org/ which allows interface to R and its cool software with all lines, arrows, nodes etc, and you can do GeneOntology lay out and Query...as well......!! Howmany genes (neurotrnsmitter TAG Domain) --> interacts with --> endocytosis TAG Domain may be better than cytoscape..!! ( not sure ) Saurin -- |------------------------------------------------ | Saurin D. Jani,MS | Statistical and Research Analyst | | Department of Cell Biology and Anatomy | Medical University of South Carolina (MUSC) | 173 Ashley Ave | Charleston,SC - 29407 | | Email: jani at musc.edu | Phone: (843)792-1340 | Website: http://www.musc.edu/~jani |------------------------------------------------ Quoting Peter <bioconductor-mailinglist at="" maubp.freeserve.co.uk="">: > Does anyone know of any existing R (r-project) code/library/package for > drawing graphs - the network variety consisting of nodes and edges (or > points and lines/arrows)? > > e.g. Genes and interactions > > For example, I might have a list of three edges for four genes: > > Gene1 -> Gene2 > Gene2 -> Gene3 > Gene4 -> Gene3 > > Which would give the following simple graph: > > Gene1 -> Gene2 -> Gene3 <- Gene4 > > Or perhaps as a matrix, which for four genes would be 4 by 4: > > [ 0 1 0 0 ] > [ 0 0 1 0 ] > [ 0 0 0 0 ] > [ 0 0 1 0 ] > > Here I have used the "rows" as the "from" and the columns as the "to" > connections. > > I was hoping there was an existing library that would accept this sort > of network definition and draw a picture of it. Maybe labelled circles > for the nodes (genes) with arrows for the edges (interactions). > > The hard part of this is the spatial layout of the nodes to avoid the > edges intersecting unrelated nodes. One simple answer is to draw the > nodes equally spaces on a large circle. Apart from the special case of > self interactions, all the edges would be inside the circle and thus > never cross another node. > > e.g. This sort of image: > > http://www.cs.toronto.edu/~wayne/research/ramsey/r.3.9.35.gif > > Of course there is the extra gloss of being able to draw a connection > weight using line thickness etc. > > Any ideas? I haven't found anything on Google... > > Thank you > > Peter > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > >
ADD COMMENT

Login before adding your answer.

Traffic: 972 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6