CytoNCA hanging with calculations of betweenness and closeness centralities

35 views
Skip to first unread message

Ajit Ray

unread,
Sep 13, 2014, 8:27:05 AM9/13/14
to cytoscape...@googlegroups.com
Hi,

My network has 15 nodes and 27 edges (carrying two weights +1 and +1). CytoNCA is hanging when I am trying to calculate betweenness or closeness centralities for this network. It does work for some of the others such as eigenvector and degree centralities. Does this have anythings to do with the negative weights? Any other suggestions? Thanks in advance!

Regards,
Ajit

Alexander Pico

unread,
Sep 19, 2014, 10:52:43 PM9/19/14
to cytoscape...@googlegroups.com
Hi Ajit,

I've forwarded your question to the developers of CytoNCA. Hopefully, they will respond to you here soon.

 - Alex

Ajit Ray

unread,
Sep 20, 2014, 1:56:53 AM9/20/14
to cytoscape...@googlegroups.com
Dear Prof. Pico,

Thank you for your response. I have been able to figure out the reason for the problem. The matlab code that I had written to generate the edge list ended up feeding the information in to secondary sheets of an excel file based on the number of networks generated. Cytoscape only recognizes data from the first sheet. It did not strike me earlier as excel opens the non-empty data sheet by default. I apologize for any inconvenience caused. I would be grateful if you could let me know how to retract my query.

Regards,
Ajit
--
You received this message because you are subscribed to a topic in the Google Groups "cytoscape-helpdesk" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cytoscape-helpdesk/Ol15ZErOzfc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cytoscape-helpd...@googlegroups.com.
To post to this group, send email to cytoscape...@googlegroups.com.
Visit this group at http://groups.google.com/group/cytoscape-helpdesk.
For more options, visit https://groups.google.com/d/optout.

Alexander Pico

unread,
Sep 20, 2014, 9:05:09 PM9/20/14
to cytoscape...@googlegroups.com
Glad you found a solution. In case it's helpful for others searching this topic, I'll forward the reply I got from the app authors:

"In the current version CytoNCA, the edge weight represents the edge's reliability or the similarity between the two connecting nodes. The betweenness centrality and closeness centrality are calculated based on the shortest path algorithm SPFA, which can not produce shortest paths on the graph with negative edges in some cases.
 

We do not know the meaning of your edge weight. The new version of CytoNCA will work on the graph with negative edges soon. In another way, you can use 0.0001 instead of -1 if -1 in your work represents an unreliable edge or no edge there. 

Please do not hesitate to contact us if you have any questions when using CytoNCA.

Best regards,

Min

Homepage: http://netlab.csu.edu.cn/personal/limin/index_en.html

Phd, Associate Professor

Department of Computer Engineering

School of Information Science and Engineering

Central South University

P.R.China"


On Friday, September 19, 2014 10:56:53 PM UTC-7, Ajit Ray wrote:
Dear Prof. Pico,

Thank you for your response. I have been able to figure out the reason for the problem. The matlab code that I had written to generate the edge list ended up feeding the information in to secondary sheets of an excel file based on the number of networks generated. Cytoscape only recognizes data from the first sheet. It did not strike me earlier as excel opens the non-empty data sheet by default. I apologize for any inconvenience caused. I would be grateful if you could let me know how to retract my query.

Regards,
Ajit

 On Sep 20, 2014 8:22 AM, "Alexander Pico" <ap...@gladstone.ucsf.edu> wrote:
Hi Ajit,

I've forwarded your question to the developers of CytoNCA. Hopefully, they will respond to you here soon.

 - Alex

On Saturday, September 13, 2014 5:27:05 AM UTC-7, Ajit Ray wrote:
Hi,

My network has 15 nodes and 27 edges (carrying two weights +1 and +1). CytoNCA is hanging when I am trying to calculate betweenness or closeness centralities for this network. It does work for some of the others such as eigenvector and degree centralities. Does this have anythings to do with the negative weights? Any other suggestions? Thanks in advance!

Regards,
Ajit

--
You received this message because you are subscribed to a topic in the Google Groups "cytoscape-helpdesk" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cytoscape-helpdesk/Ol15ZErOzfc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cytoscape-helpdesk+unsub...@googlegroups.com.
To post to this group, send email to cytoscape-helpdesk@googlegroups.com.

Ajit Ray

unread,
Sep 21, 2014, 1:14:27 AM9/21/14
to cytoscape...@googlegroups.com
Dear Prof. Pico.

Thank you for posting the reply to the group. 

Regards,
Ajit

--
You received this message because you are subscribed to a topic in the Google Groups "cytoscape-helpdesk" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cytoscape-helpdesk/Ol15ZErOzfc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cytoscape-helpd...@googlegroups.com.
To post to this group, send email to cytoscape...@googlegroups.com.



--
Ajit Ray
Division of Cellular and Molecular Neuroscience,
National Brain Research Centre,
Nainwal Mode, NH-8,
Manesar, Gurgaon - 122 050
India
Reply all
Reply to author
Forward
0 new messages