previousSelection: Automated Selection of Previously Selected Features

In ClassifyR: A framework for cross-validated classification problems, with applications to differential variability and differential distribution testing

Description

Uses the feature selection of the same cross-validation iteration of a previous classification for the current classification task.

Usage

  ## S4 method for signature 'matrix'
previousSelection(measurements, ...)
  ## S4 method for signature 'DataFrame'
previousSelection(measurements, classes, datasetName,
                  classifyResult, minimumOverlapPercent = 80,
                        selectionName = "Previous Selection", .iteration, verbose = 3)
  ## S4 method for signature 'MultiAssayExperiment'
previousSelection(measurements, ...)

Arguments

measurements	Either a matrix, DataFrame or MultiAssayExperiment containing the training data. For a matrix, the rows are features, and the columns are samples.
classes	Do not specify this variable. It is ignored and only used to create consistency of formal parameters with other feature selection methods.
...	Variables not used by the matrix nor the MultiAssayExperiment method which are passed into and used by the DataFrame method.
datasetName	A name for the data set used. Stored in the result.
classifyResult	An existing classification result from which to take the feature selections from.
minimumOverlapPercent	If at least this many selected features can't be identified in the current data set, then the selection stops with an error.
selectionName	A name to identify this selection method by. Stored in the result.
.iteration	Do not specify this variable. It is set by runTests if this function is being repeatedly called by runTests.
verbose	Default: 3. A number between 0 and 3 for the amount of progress messages to give. This function only prints progress messages if the value is 3.

Value

An object of class SelectResult.

Author(s)

Dario Strbenac

Examples

  #if(require(sparsediscrim))
  #{
    # Genes 76 to 100 have differential expression.
    genesMatrix <- sapply(1:25, function(sample) c(rnorm(100, 9, 2)))
    genesMatrix <- cbind(genesMatrix, sapply(1:25, function(sample)
                                 c(rnorm(75, 9, 2), rnorm(25, 14, 2))))
    colnames(genesMatrix) <- paste("Sample", 1:50)
    rownames(genesMatrix) <- paste("Gene", 1:100)                                 
    classes <- factor(rep(c("Poor", "Good"), each = 25))
    resubstitute <- ResubstituteParams(nFeatures = seq(10, 100, 10), 
                        performanceType = "error", better = "lower")
    result <- runTests(genesMatrix, classes, datasetName = "Example",
                       classificationName = "Differential Expression",
                       permutations = 2, fold = 2,
                       params = list(SelectParams(), TrainParams(), PredictParams()))
    features(result)
                       
    # Genes 50 to 74 have differential expression in new data set.
    newDataset <- sapply(1:25, function(sample) c(rnorm(100, 9, 2)))
    newDataset <- cbind(newDataset, rbind(sapply(1:25, function(sample) rnorm(49, 9, 2)),
                                          sapply(1:25, function(sample) rnorm(25, 14, 2)),
                                          sapply(1:25, function(sample) rnorm(26, 9, 2))))
                                          
    rownames(newDataset) <- rownames(genesMatrix)
    colnames(newDataset) <- colnames(genesMatrix)
                                           
    newerResult <- runTests(newDataset, classes, datasetName = "Latest Data Set",
                            classificationName = "Differential Expression",
                            permutations = 2, fold = 2,
                       params = list(SelectParams(previousSelection,
                                     intermediate = ".iteration",
                                     classifyResult = result),
                       TrainParams(), PredictParams()))
                                     
    # However, only genes 76 to 100 are chosen, because the feature selections are
    # carried over from the first cross-validated classification.
    features(newerResult)
  #}  

ClassifyR documentation built on Nov. 8, 2020, 6:53 p.m.