Thank you for your reply
(and thanks to Tim for forwarding the mail)
----- Original Message -----
From: "Steve Horvath" <firstname.lastname@example.org>
To: ttriche at usc.edu, "Sudeep Sahadevan" <sudeep.sahadevan at="" scai-="" extern.fraunhofer.de="">
Cc: bioconductor at r-project.org, "Peter Langfelder"
<peter.langfelder at="" gmail.com="">
Sent: Wednesday, June 20, 2012 6:19:52 PM
Subject: RE: [BioC] WGCNA chooseTopHubInEachModule function
Dear Tim and Sudeep,
regarding your question one can invoke a general rule linking unsigned
and signed networks: if a power of beta is chosen for an unsigned
network then one should choose a power of 2*beta for corresponding
Therefore, I suggest to use a power of 4 for a signed network.
In any event, the good news is that weighted networks are fairly
robust with respect to (soft) threshold choices (i.e. the power) so
the result should be fairly robust irrespective of the choice of beta.
From: Tim Triche, Jr. [email@example.com]
Sent: Wednesday, June 20, 2012 6:38 AM
To: Sudeep Sahadevan
Cc: bioconductor at r-project.org; Horvath, Steve; Peter.Langfelder at
Subject: Re: [BioC] WGCNA chooseTopHubInEachModule function
WGCNA is not a BioC package, you should cc: the authors (Steve Horvath
and Peter Langfelder) on your email (IMHO)
On Wed, Jun 20, 2012 at 2:36 AM, Sudeep Sahadevan <sudeep.sahadevan at="" scai-extern.fraunhofer.de<mailto:sudeep.sahadevan="" at="" scai-="" extern.fraunhofer.de="">> wrote:
In WGCNA R package the default "power" argument for the function
"chooseTopHubInEachModule" is 2. My question is there anyway to test
what would be the optimum argument to use for a signed network ?
Thank you in advance.
Bioconductor mailing list
Bioconductor at r-project.org<mailto:bioconductor at="" r-project.org="">
Search the archives:
A model is a lie that helps you see the truth.
Skipper<http: cancerres.aacrjournals.org="" content="" 31="" 9="" 1173.full.pdf="">
IMPORTANT WARNING: This email (and any attachments) is