Question: Additional high-scoring subnetworks in BioNet
0
gravatar for enricoferrero
3.4 years ago by
enricoferrero570
Switzerland
enricoferrero570 wrote:

I'd like to use BioNet to identify high-scoring subnetworks in my data.

The tutorial clearly explains how to identify the top scoring subnetwork (I don't have the CPLEX library so I'm using runFastHeinz()):

library(BioNet)
library(DLBCL)
data(dataLym)
data(interactome)

pvals <- cbind(t=dataLym$t.pval, s=dataLym$s.pval)
rownames(pvals) <- dataLym$label
pval <- aggrPvals(pvals, order=2, plot=FALSE)

subnet <- subNetwork(dataLym$label, interactome)
subnet <- rmSelfLoops(subnet)
subnet
A graphNEL graph with undirected edges
Number of Nodes = 2559
Number of Edges = 7788

fb <- fitBumModel(pval, plot=FALSE)
scores <- scoreNodes(subnet, fb, fdr=0.001)

module <- runFastHeinz(subnet, scores)
module
A graphNEL graph with undirected edges
Number of Nodes = 37
Number of Edges = 44

But what about other high-scoring subnetworks? It is likely that in a network with 2259 nodes there will be other interesting active subnetworks to explore other than the first one. How can I retrieve these?

Bonus question: are there any parameters that I can set to increase or decrease the size of the resulting subnetworks?

Thank you.

network bionet • 504 views
ADD COMMENTlink modified 3.4 years ago by branislav misovic120 • written 3.4 years ago by enricoferrero570
Answer: Additional high-scoring subnetworks in BioNet
1
gravatar for branislav misovic
3.4 years ago by
Netherlands/Amsterdam
branislav misovic120 wrote:

hi  Enrico 
 Last year ,during the Networks course,  I  asked same question Gunnar W. Klau  (one of  the authors who worked with Heinz &  link to Cytoscape Examine plugin )   ,   he said :  "you can remove your 1st Module nodes from you PPI DB  and rerun the Heinz to get 2nd Module ... in the past Heinz  gave more than 1  Module" .
  I do not know the reason why they stopped providing that output .

ADD COMMENTlink written 3.4 years ago by branislav misovic120

Thanks Branislav, that sounds like a reasonable (although not ideal) workaround.

Hopefully someone from the dev team can step in and comment.

ADD REPLYlink written 3.4 years ago by enricoferrero570
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 198 users visited in the last hour