iCluster (version 2.1.0)

tune.iCluster2: Model tuning function

Description

Model tuning process for choosing the number of clusters k and the lasso penalty parameters.

Usage

tune.iCluster2(datasets, k, n.lambda,nrep, mc.cores,max.iter)

Arguments

datasets

A list containing data matrices. For each data matrix, the rows represent samples, and the columns represent genomic features.

k

Number of classes for the samples.

nrep

Number of training and test data partition for computing the reproducibility index.

n.lambda

The number of sampled points for the uniform design. Use the default value by setting n.lambda=NULL

mc.cores

Number of cores to use for parallel computation.

max.iter

Number of EM iterations.

Value

A list with the following elements.

best.fit

Model fit under the optimal lambda values that give the highest reproducibility index.

RI

A vector of reproducibility index associated with each of the sampled lambda combination.

ud

Sampled lambda combinations under the uniform design

References

Ronglai Shen, Adam Olshen, Marc Ladanyi. (2009). Integrative clustering of multiple genomic data types using a joint latent variable model with application to breast and lung cancer subtype analysis. Bioinformatics 25, 2906-2912.

Ronglai Shen, Qianxing Mo, Nikolaus Schultz, Venkatraman E. Seshan, Adam B. Olshen, Jason Huse, Marc Ladanyi, Chris Sander. (2012). Integrative Subtype Discovery in Glioblastoma Using iCluster. PLoS ONE 7, e35236

See Also

iCluster2,plotiCluster, compute.pod, plotHeatmap

Examples

library(iCluster)
library(caTools, lib.loc="/apps/Rlib64/")
library(gdata, lib.loc="/apps/Rlib64/")
library(gtools, lib.loc="/apps/Rlib64/")
library(gplots, lib.loc="/apps/Rlib64/")
library(lattice, lib.loc="/apps/Rlib64/")
library(parallel, lib.loc="/apps/Rlib64/")

#data(simu.datasets)

#cv.fit=alist()
#for(k in 2:5){
#  cat(paste("K=",k,sep=""),'\n')
#  cv.fit[[k]]=tune.iCluster2(simu.datasets, k, mc.cores=6)
#}

##Reproducibility index (RI) plot
#plotRI(cv.fit)

##Based on the RI plot, k=3 is the best solution
#best.fit=cv.fit[[3]]$best.fit

##Try different color schemes
#plotHeatmap(fit=best.fit,datasets=simu.datasets,
#sparse=c(TRUE,TRUE),col.scheme=list(bluered(256), greenred(256)))