Can I change size or select nodes and remove node labels selectively

[Adrian Johnson]

Hi:
I am using Cytoscape 3.0.

After I create a network with ~ 500 nodes, I want to select

While doing this I wanted to select all transcription factors I know,  say 10 of them in this network and increase node size to 200.

Currently I am doing this one of by one - right click - edit and choose edit node size  from 30 to 200.

Is is possible to select a subset of nodes from network and change node sizes one stroke.


Also - How can I remove node labels for selected nodes.



Thanks
Adrian.

[piet molenaar]

Yes, this is possible and in fact exactly what Cytoscape is meant for. If your network has annotation that distinguishes these known TF's you can map a visual property (size; having a node label) to this annotation as described here : http://wiki.cytoscape.org/Cytoscape_3/UserManual#Visual_Styles  Otherwise you'd have to annotate these specific nodes with an extra column of data by creating an extra column in the table panel (click the new data column icon; fifth from left) of e.g string type and set your selected nodes to a value of e.g. TF. Next you set the visual style again.

Hope this helps,

Piet  

--
You received this message because you are subscribed to the Google Groups "cytoscape-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cytoscape-disc...@googlegroups.com.
To post to this group, send email to cytoscap...@googlegroups.com.
Visit this group at http://groups.google.com/group/cytoscape-discuss.
For more options, visit https://groups.google.com/groups/opt_out.

--
Piet Molenaar
piet...@gmail.com
Department of Oncogenomics, M1-131
Academic Medical Center
University of Amsterdam
Meibergdreef 9
1105 AZ Amsterdam
the Netherlands

tel (+31) 20-5666592
fax (+31) 20-6918626

[Adrian Johnson]

Hi Piet,
Thanks for the tip. It worked.  I have another question may be you can suggest a way.

From a microarray experiment, I extracted the protein protein interaction data from HPRD for those genes that are differentially expressed.

From 750 differentially expressed genes, the interaction grew to 1700 binary relationships.


I want to represent a cluster diagram showing all NF-KB cluster, Interleukin and cytokine sub-network,  Interferon cluster with JAK-STAT signaling,  AP1 cluster, MYC and SMAD cluster.

Is there a module that can separate the data into these clusters. Basically do pathway analysis and put the clusters into respective sub-networks.


Thanks for your help again.

Adrian

[piet molenaar]

I don't know of a pathway overrepresentation cytoscape app after clustering; I'd cluster the network (using e.g. MCODE) and look for overrepresentation of GO categories in the clusters using e.g. BinGO. You can also use the lists of genes from the clusters in a tool like DAVID that does calculate pathway overrepresentation (http://david.abcc.ncifcrf.gov/)

Hope this helps,

Piet