epoc.validation {epoc}R Documentation

epoc.validation

Description

Model validation using random split or cross-validation

Usage

epoc.validation(type=c('pred','concordance'),repl,Y,U,lambdas=NULL,
   method='G',thr=1e-10,trace=0,...)

Arguments

type

'pred' for 10-fold CV of prediction error. 'concordance' for random split network concordance using Kendall W.

repl

The number of replicates

Y

mRNA, samples x genes

U

CNA, samples x genes

lambdas

series of relative \lambdas or default=NULL which means let EPoC choose

method

'G' means EPoC G and 'A' means EPoC A.

thr

Threshold for convergence to the LASSO solver

trace

Debug information

...

Extra parameters passed through to the EPoC solver

Details

In the case of 'pred' assess CV prediction error using 10-fold cross-validation with repl replicates. In the case of 'concordance' assess network concordance using random split and Kendall W with repl replicates.

Value

A list of class plot.EPoC.validation.pred or plot.EPoC.validation.W respectively.

References

Rebecka Jörnsten, Tobias Abenius, Teresia Kling, Linnéa Schmidt, Erik Johansson, Torbjörn Nordling, Bodil Nordlander, Chris Sander, Peter Gennemark, Keiko Funa, Björn Nilsson, Linda Lindahl, Sven Nelander. (2011) Network modeling of the transcriptional effects of copy number aberrations in glioblastoma. Molecular Systems Biology 7 (to appear)

See Also

epoc, plot.EPoC.validation


[Package epoc version 0.2.6-1.1 Index]