runClassifier-methods: Perform classification.

In geneClassifiers: Application of gene classifiers

Description

runClassifier performs classification by applying a classifier to gene expression data.

Usage

runClassifier(classifierParameters, fixedExpressionData, ...)


  ## S4 method for signature 'character,FixedExpressionData'
runClassifier(classifierParameters,
  fixedExpressionData, ...)


  ## S4 method for signature 'ClassifierParameters,FixedExpressionData'
runClassifier(classifierParameters,
  fixedExpressionData, ...)

Arguments

classifierParameters	Either a text value indicating a classifier name (see showClassifierList), or an object of class ClassifierParameters as returned by the getClassifier function.
fixedExpressionData	The data to be classified in the form of a FixedExpressionData object as returned by the setNormalizationMethod function.
...	see details

Details

A list of possible classifiers is obtained by showClassifierList. The data to be classified is first to be processed by the setNormalizationMethod function. By default the data is assumed to contain many (n>=25) samples with corresponding probe-sets needed for classification. If one of these conditions is not met, a classifier outcome might be seriously affected. By default an error is given. Although strongly discouraged, it is possible to circumvent the security checks. If not all required probe-sets are included in the input set, you can explicitly pass the parameter allow.reweighted = TRUE to the runClassifier function in order to determine the classifier outcome using less probe-sets (e.g. possible if the missing probe-sets are known to have minimal contribution).See vignette("MissingCovariates") for more information. If the input data has a small number of samples, the default batch correction becomes ineffective. If you are aware of the possible negative effects you can force to not use batch correction by passing the parameter do.batchcorrection=FALSE.

Value

The classification results as an object of class ClassifierResults.

See Also

Other workflow functions: getNormalizationMethod, setNormalizationMethod, showClassifierList

Examples

data(exampleMAS5)
myData<-setNormalizationMethod(exampleMAS5,"MAS5.0",targetValue=500)
runClassifier("EMC92",myData)

Example output

Note: Research use only
Classifier: EMC92 
	 > 0.827 : Risk-II 
classifications
 Risk-I Risk-II 
     19       6 

	Batch corrected   : yes 
	weighting type      : complete 
----------------

geneClassifiers documentation built on Nov. 8, 2020, 4:58 p.m.