not all edge attributes send from R to Cytoscape

[ramona]

Hi all,

I just started using the gaggle and tried to send an edge as well as
node annotated graph from R to Cytoscape. Unfortunately, not all of
the attributes are recieved by Cytoscape. Does anyone know why this
could happen?

Would be really happy about any advice!

Many thanks
Ramona

[Dan Tenenbaum]

Can you send the R code you used to generate the graph?

Thanks!

Dan

[Christopher Bare]

Hi Ramona,

To shorten your example a little:

graph=randomEGraph(LETTERS[1:8], edges=10)
nodeDataDefaults(graph, "attr1") <- NA
nodeData(graph, n=nodes(graph) , attr="attr1") <-
rep(list(1:numNodes(graph)),numNodes(graph))
broadcast(graph)

If I understand your question right, you're expecting each node in the
graph to have an attribute called attr1 with a value of
[1,2,3,4,5,6,7,8]? The problem is that although an R graph allows you
to assign a list as a node attribute, Gaggle networks don't allow
that.

I doubt Cytoscape could do anything with a multiple values for an
attribute either, but I don't know that for sure. The fact that you're
seeing attr1=8 on the Cytoscape side, where 8 is the last element of
list(1:numNodes(graph)), is an artifact of the funny way the R goose
transforms an R graph object into a Gaggle network object.

I hope this helps,

-chris

------------------------------------------------
J. Christopher Bare
Software Engineer, Baliga Lab
Institute for Systems Biology
------------------------------------------------

On Fri, Aug 28, 2009 at 11:15 AM, Dan
Tenenbaum<dtene...@systemsbiology.org> wrote:
>
>
> ---------- Forwarded message ----------
> From: ramona <ramona.s...@googlemail.com>
> Date: Fri, Aug 28, 2009 at 7:30 AM
> Subject: Re: not all edge attributes send from R to Cytoscape
> To: Dan Tenenbaum <dtene...@systemsbiology.org>
>
>
> Dear Dan,
>
> the program I wrote is quite big as well as is the resulting graph.I
> extracted a subgraph from my real graph which then has 50 nodes, 25
> edges and 4 node and 10 edge attributes assigned. Could it simply be
> too big and I do not get an error message by windows about the fact
> that not all data is send to/received by Cytoscape? Maybe it would
> help if I sent you my saved graph object?
> The generalized way i generate the graph is
> ##########
> graph=randomEGraph(LETTERS[1:8], edges=10)
> nodeDataDefaults(graph, "attr1") <- NA
> nodeData(graph,n=nodes(graph) , attr="attr1") <- rep(list(1:numNodes
> (graph)),numNodes(graph))
> nodeDataDefaults(graph, "attr2") <- NA
> nodeData(graph, n=nodes(graph), attr="attr2") <- rep(list(rep("ca",
> 5)),numNodes(graph))
>
> graph_edgeMatrix=edgeMatrix(graph)
> node_a=nodes(graph)[graph_edgeMatrix[1,]]
> node_b=nodes(graph)[graph_edgeMatrix[2,]]
>
> edgeDataDefaults(graph, "attr3") <- list(rep(NA, 5))
> edgeData(graph, from=node_a, to=node_b, attr="attr3") <- rep(list(rep
> ("ca", 5)),numEdges(graph))
>
> edgeDataDefaults(graph, "attr4") <- list(rep(NA, numEdges))
> edgeData(graph, from=node_a, to=node_b, attr="attr4") <- rep(list
> (1:numEdges(graph)),numEdges(graph))
> edgeDataDefaults(graph, "attr5") <- list(rep(NA, 5))
> edgeData(graph, from=node_a, to=node_b, attr="attr5") <- rep(list(rep
> ("ca", 5)),numEdges(graph))
>
> broadcast(graph)
> #######
>
> All of the attibute types are displayed in Cytoscape, what is not the
> case for my graph. What I did not realize before is in the case of
> vectors assigned, only the last element is displayed in Cytoscape.
>
> How do I have to assign vectors such that they are completely
> displayed by Cytoscape?
>
> Many thanks
> Ramona
>
>
> On 27 Aug., 17:59, Dan Tenenbaum <dtenenb...@systemsbiology.org>

> wrote:
>> Can you send the R code you used to generate the graph?Thanks!
>> Dan
>>
>> On Thu, Aug 27, 2009 at 3:46 AM, ramona

>> <ramona.schmid...@googlemail.com>wrote:

>>
>>
>>
>>
>>
>> > Hi all,
>>
>> > I just started using the gaggle and tried to send an edge as well as
>> > node annotated graph from R to Cytoscape. Unfortunately, not all of
>> > the attributes are recieved by Cytoscape. Does anyone know why this
>> > could happen?
>>
>> > Would be really happy about any advice!
>>
>> > Many thanks

>> > Ramona- Zitierten Text ausblenden -
>>
>> - Zitierten Text anzeigen -
>
>

[burak kutlu]

Hi Ramona
One way around the problem would be pass one concatanated character string. You may split them later in R.
-burak