SPreFuGED (version 1.0)

generateGED: A function to simulate two groups gene expression data (GED)

Description

This function draws two sets of vectors (train and test samples' labels) from a binomial distribution and generates two gene expression datasets (train and test data) from a multivariate normal distribution with a mean vector U[6,10] and a given within-class covariance matrix, at each iteration.

Usage

generateGED(covAll, nTrain, nTest, log2FC = 1, niter = 3, prob = 0.5)

Arguments

covAll

an object returned by covMax or a list containing a covariance matrix cov and the proportion of DE genes pie.

nTrain

the number of samples in the training set

nTest

the number of samples in the test set

log2FC

the absolute Log2 fold changes (effect sizes) for DE genes (Default is 1)

niter

the number of iterations (train/test datasets to be generated). Default is 3

prob

the probability of success for the binomial sampling. Default is 0.5

Value

A list of length niter. Each element of which is a list containing: A list of length niter. Each element of which is a list containing:

References

Jong VL, Novianti PW, Roes KCB & Eijkemans MJC. Selecting a classification function for class prediction with gene expression data. Bioinformatics (2016) 32(12): 1814-1822

See Also

covMat, directClass and plotDirectClass

Examples

myCov<-covMat(pAll=100, lambda=2, corrDE=0.75, sigma=0.25);
myData<-generateGED(covAll=myCov, nTrain=30, nTest=10);
myFirstTrainData<-myData[[1]]$trainData; myFirstTrainLabels<-myData[[1]]$trainLabels;
myFirstTestData<-myData[[1]]$testData; myFirstTestLabels<-myData[[1]]$testLabels;