netClass (version 1.2.1)

cv.stsvm: Cross validation for smoothed t-statistic to select significant top ranked differential expressed genes

Description

Cross validation for smoothed t-statistic to select significant top ranked differential expressed genes

Usage

cv.stsvm(x=x, x.mi=NULL,y=y, folds=5,Gsub=matrix(1,100,100),op.method=c("pt","spb"), 
		repeats=3, parallel=FALSE, cores=2,DEBUG=TRUE, pt.pvalue=0.05,op=0.85,
		aa=1000,a=1,p=2,allF=TRUE, seed=1234,Cs=10^c(-3:3))

Arguments

x

A p x n matrix of expression measurements with p samples and n genes.

x.mi

A p x m matrix of expression measurements with p samples and m miRNAs.

y

A factor of length p comprising the class labels.

folds

Folds number of folds to perform

Gsub

An adjacency matrix that represents the underlying biological network.

op.method

Method for selecet optimal feature subgoups: pt is permutation test, sp is span bound.

repeats

Number of how often to repeat the x-fold cross-validation

parallel

Use parallel computing or not

cores

Number of cores will used when parallel is TRUE

DEBUG

Show debugging information in screen more or less.

pt.pvalue

Cut off p-value of permutation test

op

Optimal on top op

aa

permutation test steps for permutation test (pt); low bounds top op

a

constant value of random walk kernel

p

random walk step(s) of random walk kernel

allF

Using all features (TRUE) or only these genes mapped to prior information (FALSE).

seed

seed for random sampling.

Cs

Soft-margin tuning parameter of the SVM. Defaults to 10^c(-3:3).

Value

• a LIST for Cross-Validation results
• auc
  The AUC values of each test fold
• fits
  The tranined models for traning folds
• feat
  The feature selected by each by the fits
• labels
  the original lables for training

References

Yupeng Cun, Holger Frohlich (2013) Network and Data Integration for Biomarker Signature Discovery via Network Smoothed T-Statistics. PLoS ONE 8(9): e73074.

Examples

library(netClass)
data(expr)
data(ad.matrix)
x <- expr$genes
y <- expr$y

r.stsvm <- cv.stsvm(x=x[,1:500],x.mi=NULL,y=y,folds=3,Gsub=ad.matrix,op.method="pt",
				repeats=1, parallel=FALSE, cores=2,DEBUG=TRUE,pt.pvalue=0.05,op=0.9,
				aa=5,a=1,p=2,allF=TRUE, seed=1234,Cs=10^(-3:3))