Software used on the article

Stability-based comparison of class discovery methods for array-CGH profiles

Isabel Brito a,b,c,*, Philippe Hupéa,b,c,d, Pierre Neuviala,b,c and Emmanuel Barillota,b,c
a Institut Curie, 26 rue d'Ulm, Paris Cedex 05, F-75248 France,
b INSERM, U900, Paris, F-75248 France,
c Mines ParisTech, Fontainebleau, F-77300 France,
d CNRS UMR144, Paris, F-75248 France
*corresponding author: isabel.brito@curie.fr


Software

All software is implemented within the R programming language http://www.r-project.org. We used the following R Packages

package name available to download at
cluster http://www.r-project.org
hybridHclust http://www.r-project.org
WECCA http://www.few.vu.nl/~wvanwie/software/WECCA/WECCA.html
clusterv http://homes.dsi.unimi.it/valentini/software.html
mostclust http://homes.dsi.unimi.it/valentini/software.html

on which our script "StabClustCGH.R" depends.

Data sets

Data sets used on this article are available to download here blaveri , douglas , gysin , patil , veltman ,

How to use "StabClustCGH.R"

Once R packages(1), data sets (you can choose to run one or several data sets) and "StabClustCGH.R" script are dowloaded, use the following R commands,
> DATA<-list(data set(s))
> nb<-c(2:10) # number of clusters for each solution
> f<-0.8 # part of the data set which is randomly picked for each subsample in the resampling procedure
> nsub<-100 # half of the number of times the perturbation procedure is applied in the resampling procedure
> StabClustCGH(DATA,nb,f,nsub)


(1) cluster, hybridHclust, clusterv and mosclust should be put on your usual R local library and for WECCA authors' instructions should be followed.