dksSelectGenes

 An object of class <code>DKSGeneScores</code>, typically generated
		by <code>dksTrain</code>

data

 The number of genes, per class, to include in the classification 
		signature.


	The <code>DKSGeneScores</code> returned by <code><a rd-options="" href="/link/dksTrain?package=dualKS&version=1.32.0" data-mini-rdoc="dualKS::dksTrain">dksTrain</a></code> holds 
	the rank data for all the genes in the original <code>ExpressioSet</code>. 
	However, generally only the top n genes for each class are desired 
	for classification.  Rather than needing to re-run <code>dksTrain</code> 
	every time a signature of different size (n) is desired, you 
	simply extract that top n genes from this object using dksSelectGenes.


classif

This package implements a Kolmogorov Smirnov rank-sum based
algorithm for training (i.e. discriminant analysis--identification
of genes that discriminate between classes) and classification
of gene expression data sets.  One of the chief strengths of
this approach is that it is amenable to the "multiclass" problem.
That is, it can discriminate between more than 2 classes.

Eric J Kort

dualKS

Dual KS Discriminant Analysis and Classification

dksSelectGenes function

An object of class <code>DKSGeneScores</code>, typically generated
		by <code>dksTrain</code>

The number of genes, per class, to include in the classification 
		signature.

The <code>DKSGeneScores</code> returned by <code><a href="dksTrain" rd-options="">dksTrain</a></code> holds 
	the rank data for all the genes in the original <code>ExpressioSet</code>. 
	However, generally only the top n genes for each class are desired 
	for classification.  Rather than needing to re-run <code>dksTrain</code> 
	every time a signature of different size (n) is desired, you 
	simply extract that top n genes from this object using dksSelectGenes.

dksSelectGenes: Extract gene signatures from a DKSGeneScores object.

Description

Usage

Arguments

Value

See Also

Examples