ClassifyR (version 1.6.2)

previousSelection: Automated Selection of Previously Selected Features

Description

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

Usage

"previousSelection"(expression, classes, ...) "previousSelection"(expression, datasetName, classifyResult, minimumOverlapPercent = 80, selectionName = "Previous Selection", .iteration, verbose = 3)

Arguments

expression

Either a matrix or ExpressionSet containing the training data. For a matrix, the rows are features, and the columns are samples.

classes

A vector of class labels.

...

For the matrix method, variables passed to the ExpressionSet method.

datasetName

A name for the dataset 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 dataset, then the selection stops with an error.

selectionName

A name to identify this selection method by. Stored in the result.

.iteration

Not to be set by the user.

verbose

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.

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))))
    rownames(genesMatrix) <- paste("Gene", 1:100)                                 
    classes <- factor(rep(c("Poor", "Good"), each = 25))
    resubstitute <- ResubstituteParams(nFeatures = seq(10, 100, 10), 
                                       performanceType = "err", better = "lower")
    result <- runTests(genesMatrix, classes, "Ovarian Cancer", "Differential Expression",
                       resamples = 2, fold = 2,
                       params = list(SelectParams(limmaSelection, resubstituteParams = resubstitute),
                       TrainParams(dlda, TRUE, FALSE),
                       PredictParams(predict, TRUE, getClasses = function(result) result[["class"]])))
                       
    # Genes 74 to 98 have differential expression in new dataset.
    newDataset <- sapply(1:25, function(sample) c(rnorm(100, 9, 2)))
    newDataset <- cbind(newDataset, rbind(sapply(1:25, function(sample) rnorm(73, 9, 2)),
                                           sapply(1:25, function(sample) rnorm(25, 14, 2)),
                                           sapply(1:25, function(sample) rnorm(2, 14, 2))))
                                           
    newerResult <- runTests(newDataset, classes, "Ovarian Cancer Updated", "Differential Expression",
                       resamples = 2, fold = 2,
                       params = list(SelectParams(previousSelection, intermediate = ".iteration",
                                    classifyResult = result),
                       TrainParams(dlda, TRUE, FALSE),
                       PredictParams(predict, TRUE, getClasses = function(result) result[["class"]])))                                           
  }