Class "biosign"

Description

The biosigner object class

Slots

methodVc

character vector: selected classifier(s) ('plsda', 'randomforest', or 'svm')

accuracyMN

numeric matrix: balanced accuracies for the full models, and the models restricted to the 'S' and 'AS' signatures

tierMC

character matrix: contains the tier ('S', 'A', 'B', 'C', 'D', or 'E') of each feature for each classifier

yFc

factor with two levels: response factor

modelLs

list: selected classifier(s) trained on the subset restricted to the 'S' features

signatureLs

list: 'S' signatures for each classifier

xSubMN

matrix: dataset restricted to the 'S' tier

AS

list: 'AS' signatures and corresponding trained classifiers, in addition to the dataset restricted to tiers 'S' and 'A' ('xMN') and the labels ('yFc')

eset

ExpressionSet: when 'biosign' has been applied to an ExpressionSet, the instance with additional columns in fData containing the selected features is stored here

Objects from the Class

Objects can be created by calls of the form new("biosign", ...) or by calling the biosign function

Author(s)

Philippe Rinaudo and Etienne Thevenot (CEA)

See Also

biosign

Examples

## loading the diaplasma dataset

data(diaplasma)
attach(diaplasma)

## restricting to a smaller dataset for this example

featureSelVl <- variableMetadata[, "mzmed"] >= 490 & variableMetadata[, "mzmed"] < 500
dataMatrix <- dataMatrix[, featureSelVl]
variableMetadata <- variableMetadata[featureSelVl, ]

## signature selection for all 3 classifiers
## a bootI = 5 number of bootstraps is used for this example
## we recommend to keep the default bootI = 50 value for your analyzes

set.seed(123)
diaSign <- biosign(dataMatrix, sampleMetadata[, "type"], bootI = 5)

detach(diaplasma)