dksWeights: Calculate gene weights based on average expression.

In dualKS: Dual KS Discriminant Analysis and Classification

Description

Prior to selecting genes it may be desirable to calculate weights for each genes so that some genes are more likely than others to be included in the gene signature all other things being equal. This function will calculate an N x M weight matrix for N genes in data and M unique classes in class. The weights are based on mean expression of each gene in each class such that genes that are highly expressed on average in a given class will be weighted more highly when scoring genes for that class.

The resulting weight matrix can be bassed to dksTrain as the weights argument.

Usage

dksWeights(eset, class)

Arguments

eset	An ExpressionSet or matrix containing the gene expression data to be used for bootstrapping.
class	A factor with two or more levels indicating which class each sample in the expression set belongs OR an integer indicating which column of pData(eset) contains this information.

Value

An N x M matrix containing the weights for each gene and each class.

Note

There are many metrics the user might want to use for weighting. This convenience function just implements one of the most obvious ones. The user can provide his/her own N x M weight matrix to dksTrain. The weight matrix calculated by this function will be calculated on the fly if the weights is set to TRUE when calling dksTrain. However, it multiple calls to dksTrain are being made (for example when performing some type of optimization or validation), it will save a lot of time if the weight matrix is pre-calculated by a call to this function and the resulting matrix supplied directly to dksTrain rather than having it re-calculate the weight matrix every time.

Author(s)

Eric J. Kort

See Also

dksTrain, dksSelectGenes, dksClassify, DKSGeneScores, DKSPredicted, DKSClassifier

Examples

	data("dks")
	wt <- dksWeights(eset, 1)
	str(wt)

dualKS documentation built on Nov. 8, 2020, 8:30 p.m.