@@ -3,8 +3,8 @@ Type: Package

 Title: A framework for cross-validated classification problems, with

        applications to differential variability and differential

        distribution testing

-Version: 3.2.0

-Date: 2022-10-25

+Version: 3.2.1

+Date: 2022-11-17

 Author: Dario Strbenac, Ellis Patrick, Sourish Iyengar, Harry Robertson, Andy Tran, John Ormerod, Graham Mann, Jean Yang

 Maintainer: Dario Strbenac <dario.strbenac@sydney.edu.au>

 VignetteBuilder: knitr

@@ -27,8 +27,6 @@ export(distribution)

 export(edgesToHubNetworks)

 export(featureSetSummary)

 export(finalModel)

-export(generateCrossValParams)

-export(generateModellingParams)

 export(interactorDifferences)

 export(models)

 export(performance)

@@ -87,7 +85,6 @@ exportMethods(selectionPlot)

 exportMethods(show)

 exportMethods(totalPredictions)

 exportMethods(tunedParameters)

-import(BiocParallel)

 import(MultiAssayExperiment)

 import(grid)

 import(methods)

@@ -86,7 +86,7 @@ setMethod("ROCplot", "ClassifyResult", function(results, ...) {

 #' @rdname ROCplot

 #' @export

-setMethod("ROCplot", "list", 

+setMethod("ROCplot", "list",

           function(results, mode = c("merge", "average"), interval = 95,

                    comparison = "auto", lineColours = "auto",

                    lineWidth = 1, fontSizes = c(24, 16, 12, 12, 12), labelPositions = seq(0.0, 1.0, 0.2),

@@ -101,10 +101,18 @@ setMethod("ROCplot", "list",

   if(comparison == "auto")

   {

     if(max(characteristicsCounts) == length(results))

-      comparison <- names(characteristicsCounts)[characteristicsCounts == max(characteristicsCounts)][1]

-    else

+    { # Choose a characteristic which varies the most across the results.

+      candidates <- names(characteristicsCounts)[characteristicsCounts == length(results)]

+      allCharacteristics <- do.call(rbind, lapply(results, function(result) result@characteristics))

+      distinctValues <- by(allCharacteristics[, "value"], allCharacteristics[, "characteristic"], function(values) length(unique(values)))

+      comparison <- names(distinctValues)[which.max(distinctValues)][1]

+    } else {

       stop("No characteristic is present for all results but must be.")

+    }

   }

+  resultsWithComparison <- sum(sapply(results, function(result) any(result@characteristics[, "characteristic"] == comparison)))

+  if(resultsWithComparison < length(results))

+    stop("Not all results have comparison characteristic ", comparison, ' but need to.')

   ggplot2::theme_set(ggplot2::theme_classic() + ggplot2::theme(panel.border = ggplot2::element_rect(fill = NA)))

   distinctClasses <- levels(actualOutcome(results[[1]]))

@@ -113,6 +113,15 @@ setMethod("calcExternalPerformance", c("Surv", "numeric"),

             .calcPerformance(actualOutcome, predictedOutcome, performanceType = performanceType)[["values"]]

           })

+#' @rdname calcPerformance

+#' @exportMethod calcExternalPerformance

+setMethod("calcExternalPerformance", c("factor", "tabular"), # table has class probabilities per sample.

+          function(actualOutcome, predictedOutcome, performanceType = "AUC")

+          {

+            performanceType <- match.arg(performanceType)

+            .calcPerformance(actualOutcome, predictedOutcome, performanceType = performanceType)[["values"]]

+          })

+

 #' @rdname calcPerformance

 #' @usage NULL

 #' @export

@@ -11,22 +11,26 @@

 #' same length as the number of samples in \code{measurements} or a character vector of length 1 containing the

 #' column name in \code{measurements} if it is a \code{\link{DataFrame}}. Or a \code{\link{Surv}} object or a character vector of

 #' length 2 or 3 specifying the time and event columns in \code{measurements} for survival outcome. If \code{measurements} is a

-#' \code{\link{MultiAssayExperiment}}, the column name(s) in \code{colData(measurements)} representing the outcome.

+#' \code{\link{MultiAssayExperiment}}, the column name(s) in \code{colData(measurements)} representing the outcome.  If column names

+#' of survival information, time must be in first column and event status in the second.

 #' @param outcomeTrain For the \code{train} function, either a factor vector of classes, a \code{\link{Surv}} object, or

 #' a character string, or vector of such strings, containing column name(s) of column(s)

-#' containing either classes or time and event information about survival.

+#' containing either classes or time and event information about survival. If column names

+#' of survival information, time must be in first column and event status in the second.

 #' @param ... Parameters passed into \code{\link{prepareData}} which control subsetting and filtering of input data.

 #' @param nFeatures The number of features to be used for classification. If this is a single number, the same number of features will be used for all comparisons

 #' or assays. If a numeric vector these will be optimised over using \code{selectionOptimisation}. If a named vector with the same names of multiple assays, 

 #' a different number of features will be used for each assay. If a named list of vectors, the respective number of features will be optimised over. 

 #' Set to NULL or "all" if all features should be used.

-#' @param selectionMethod A character vector of feature selection methods to compare. If a named character vector with names corresponding to different assays, 

-#' and performing multiview classification, the respective classification methods will be used on each assay.

+#' @param selectionMethod Default: "auto". A character vector of feature selection methods to compare. If a named character vector with names corresponding to different assays, 

+#' and performing multiview classification, the respective classification methods will be used on each assay. If \code{"auto"} t-test (two categories) / F-test (three or more categories) ranking

+#' and top \code{nFeatures} optimisation is done. Otherwise, the ranking method is per-feature Cox proportional hazards p-value.

 #' @param selectionOptimisation A character of "Resubstitution", "Nested CV" or "none" specifying the approach used to optimise \code{nFeatures}.

-#' @param performanceType Default: \code{"auto"}. If \code{"auto"}, then balanced accuracy for classification or C-index for survival. Any one of the

+#' @param performanceType Default: \code{"auto"}. If \code{"auto"}, then balanced accuracy for classification or C-index for survival. Otherwise, any one of the

 #' options described in \code{\link{calcPerformance}} may otherwise be specified.

-#' @param classifier A character vector of classification methods to compare. If a named character vector with names corresponding to different assays, 

-#' and performing multiview classification, the respective classification methods will be used on each assay.

+#' @param classifier Default: \code{"auto"}. A character vector of classification methods to compare. If a named character vector with names corresponding to different assays, 

+#' and performing multiview classification, the respective classification methods will be used on each assay. If \code{"auto"}, then a random forest is used for a classification

+#' task or Cox proportional hazards model for a survival task.

 #' @param multiViewMethod A character vector specifying the multiview method or data integration approach to use.

 #' @param assayCombinations A character vector or list of character vectors proposing the assays or, in the case of a list, combination of assays to use

 #' with each element being a vector of assays to combine. Special value \code{"all"} means all possible subsets of assays.

@@ -108,12 +112,14 @@ setMethod("crossValidate", "DataFrame",

               if(!performanceType %in% c("auto", .ClassifyRenvir[["performanceTypes"]]))

                 stop(paste("performanceType must be one of", paste(c("auto", .ClassifyRenvir[["performanceTypes"]]), collapse = ", "), "but is", performanceType))

+              isCategorical <- is.character(outcome) && (length(outcome) == 1 || length(outcome) == nrow(measurements)) || is.factor(outcome)

               if(performanceType == "auto")

-              {

-                if(is.character(outcome) && (length(outcome) == 1 || length(outcome) == nrow(measurements)) || is.factor(outcome))

-                  performanceType <- "Balanced Accuracy"

-                else performanceType <- "C-index"

-              }

+                if(isCategorical) performanceType <- "Balanced Accuracy" else performanceType <- "C-index"

+              if(length(selectionMethod) == 1 && selectionMethod == "auto")

+                if(isCategorical) selectionMethod <- "t-test" else selectionMethod <- "CoxPH"

+              if(length(classifier) == 1 && classifier == "auto")

+                if(isCategorical) classifier <- "randomForest" else classifier <- "CoxPH"

+              

               # Which data-types or data-views are present?

               assayIDs <- unique(S4Vectors::mcols(measurements)$assay)

@@ -515,18 +521,6 @@ Using an ordinary GLM instead.")

     classifier

 }

-######################################

-######################################

-#' A function to generate a CrossValParams object

-#'

-#' @inheritParams crossValidate

-#'

-#' @return CrossValParams object

-#' @export

-#'

-#' @examples

-#' CVparams <- generateCrossValParams(nRepeats = 20, nFolds = 5, nCores = 8, selectionOptimisation = "none")

-#' @import BiocParallel

 generateCrossValParams <- function(nRepeats, nFolds, nCores, selectionOptimisation){

     seed <- .Random.seed[1]

@@ -549,32 +543,7 @@ generateCrossValParams <- function(nRepeats, nFolds, nCores, selectionOptimisati

     if(!any(tuneMode %in% c("Resubstitution", "Nested CV", "none"))) stop("selectionOptimisation must be Nested CV or Resubstitution or none")

     CrossValParams(permutations = nRepeats, folds = nFolds, parallelParams = BPparam, tuneMode = tuneMode)

 }

-######################################

-######################################

-#' A function to generate a ModellingParams object

-#'

-#' @inheritParams crossValidate

-#' @param assayIDs A vector of data set identifiers as long at the number of data sets.

-#'

-#' @return ModellingParams object

-#' @export

-#'

-#' @examples

-#' data(asthma)

-#' # First make a toy example assay with multiple data types. We'll randomly assign different features to be clinical, gene or protein.

-#' set.seed(51773)

-#' measurements <- DataFrame(measurements, check.names = FALSE) 

-#' mcols(measurements)$assay <- c(rep("clinical",20),sample(c("gene", "protein"), ncol(measurements)-20, replace = TRUE))

-#' mcols(measurements)$feature <- colnames(measurements)

-#' modellingParams <- generateModellingParams(assayIDs = c("clinical", "gene", "protein"),

-#'                                           measurements = measurements, 

-#'                                           nFeatures = list(clinical = 10, gene = 10, protein = 10),

-#'                                           selectionMethod = list(clinical = "t-test", gene = "t-test", protein = "t-test"),

-#'                                           selectionOptimisation = "none",

-#'                                           classifier = "randomForest",

-#'                                           multiViewMethod = "merge")

-#' @import BiocParallel

 generateModellingParams <- function(assayIDs,

                                     measurements,

                                     nFeatures,

@@ -6,9 +6,11 @@ randomForestTrainInterface <- function(measurementsTrain, outcomeTrain, mTryProp

   if(verbose == 3)

     message("Fitting random forest classifier to training data.")

   mtry <- round(mTryProportion * ncol(measurementsTrain)) # Number of features to try.

-      

   # Convert to base data.frame as randomForest doesn't understand DataFrame.

-  ranger::ranger(x = as(measurementsTrain, "data.frame"), y = outcomeTrain, mtry = mtry, importance = "impurity_corrected", ...)

+  fittedModel <- ranger::ranger(x = as(measurementsTrain, "data.frame"), y = outcomeTrain, mtry = mtry, ...)

+  forImportance <- ranger::ranger(x = as(measurementsTrain, "data.frame"), y = outcomeTrain, mtry = mtry, importance = "impurity_corrected", ...)

+  attr(fittedModel, "forImportance") <- forImportance

+  fittedModel

 }

 attr(randomForestTrainInterface, "name") <- "randomForestTrainInterface"

@@ -37,7 +39,8 @@ randomForestPredictInterface <- function(forest, measurementsTest, ..., returnTy

 forestFeatures <- function(forest)

                   {

-                    rankedFeaturesIndices <- order(ranger::importance(forest), decreasing = TRUE)

-                    selectedFeaturesIndices <- which(ranger::importance(forest) > 0)

+                    forImportance <- attr(forest, "forImportance")

+                    rankedFeaturesIndices <- order(ranger::importance(forImportance), decreasing = TRUE)

+                    selectedFeaturesIndices <- which(ranger::importance(forImportance) > 0)

                     list(rankedFeaturesIndices, selectedFeaturesIndices)

                   }

\ No newline at end of file

@@ -5,8 +5,9 @@ rfsrcTrainInterface <- function(measurementsTrain, survivalTrain, mTryProportion

     stop("The package 'randomForestSRC' could not be found. Please install it.")

   if(verbose == 3)

     message("Fitting rfsrc classifier to training data and making predictions on test data.")

-    

-  bindedMeasurements <- cbind(measurementsTrain, event = survivalTrain[, 1], time = survivalTrain[, 2])

+

+  # Surv objects store survival information as a two-column table, time and event, in that order.    

+  bindedMeasurements <- cbind(measurementsTrain, time = survivalTrain[, 1], event = survivalTrain[, 2])

   mtry <- round(mTryProportion * ncol(measurementsTrain)) # Number of features to try.

   randomForestSRC::rfsrc(Surv(time, event) ~ ., data = as.data.frame(bindedMeasurements), mtry = mtry,

                           var.used = "all.trees", importance = TRUE, ...)

@@ -4,7 +4,7 @@ coxphRanking <- function(measurementsTrain, survivalTrain, verbose = 3) # Clinic

   pValues <- rep(NA, ncol(measurementsTrain))

   names(pValues) <- colnames(measurementsTrain)

-  

+

   isCat <- sapply(measurementsTrain, class) %in% c("character", "factor")

   if(any(isCat))

   {

@@ -16,7 +16,8 @@

 #' \code{matrix} or \code{\link{DataFrame}}, the rows are samples, and the columns are features.

 #' @param outcomeTrain Either a factor vector of classes, a \code{\link{Surv}} object, or

 #' a character string, or vector of such strings, containing column name(s) of column(s)

-#' containing either classes or time and event information about survival.

+#' containing either classes or time and event information about survival. If column names

+#' of survival information, time must be in first column and event status in the second.

 #' @param measurementsTest Same data type as \code{measurementsTrain}, but only the test

 #' samples.

 #' @param outcomeTest Same data type as \code{outcomeTrain}, but for only the test

@@ -256,12 +257,19 @@ input data. Autmomatically reducing to smaller number.")

     {

       if(is.null(modellingParams@trainParams@getFeatures))

       selectedFeatures <- originalFeatures[selectedFeaturesIndices]

-      else selectedFeatures <- colnames(measurementsTrain)[rankedFeaturesIndices] 

+      else selectedFeatures <- colnames(measurementsTrain)[selectedFeaturesIndices] 

     } else {

       featureColumns <- na.omit(match(c("assay", "feature"), colnames(S4Vectors::mcols(measurementsTrain))))

-      if(is.null(modellingParams@trainParams@getFeatures))

-      selectedFeatures <- originalFeatures[selectedFeaturesIndices, ]

-      else selectedFeatures <- S4Vectors::mcols(measurementsTrain)[selectedFeaturesIndices, featureColumns]

+      if(length(featureColumns) == 1)

+      {

+         if(is.null(modellingParams@trainParams@getFeatures))

+            selectedFeatures <- originalFeatures[selectedFeaturesIndices]

+         else selectedFeatures <- S4Vectors::mcols(measurementsTrain)[selectedFeaturesIndices, featureColumns]            

+      } else {

+                if(is.null(modellingParams@trainParams@getFeatures))

+                  selectedFeatures <- originalFeatures[selectedFeaturesIndices, ]

+                else selectedFeatures <- S4Vectors::mcols(measurementsTrain)[selectedFeaturesIndices, featureColumns]            

+      }

     }

     importanceTable <- S4Vectors::DataFrame(selectedFeatures, performanceChanges)

     if(ncol(importanceTable) == 2) colnames(importanceTable)[1] <- "feature"

@@ -272,6 +280,7 @@ input data. Autmomatically reducing to smaller number.")

   if(!is.null(tuneDetailsSelect)) tuneDetails <- tuneDetailsSelect else tuneDetails <- tuneDetailsTrain

   # Convert back into original, potentially unsafe feature identifiers unless it is a nested cross-validation.

+  

   if(is.null(.iteration) || .iteration != "internal")

   {

     if(!is.null(rankedFeaturesIndices))

@@ -283,9 +292,16 @@ input data. Autmomatically reducing to smaller number.")

         else rankedFeatures <- colnames(measurementsTrain)[rankedFeaturesIndices]            

       } else {

         featureColumns <- na.omit(match(c("assay", "feature"), colnames(S4Vectors::mcols(measurementsTrain))))          

-        if(is.null(modellingParams@trainParams@getFeatures))

-          rankedFeatures <- originalFeatures[rankedFeaturesIndices, ]

-        else rankedFeatures <- S4Vectors::mcols(measurementsTrain)[rankedFeaturesIndices, featureColumns]

+        if(length(featureColumns) == 1)

+        {

+          if(is.null(modellingParams@trainParams@getFeatures))

+            rankedFeatures <- originalFeatures[rankedFeaturesIndices]

+          else rankedFeatures <- S4Vectors::mcols(measurementsTrain)[rankedFeaturesIndices, featureColumns]

+        } else {

+          if(is.null(modellingParams@trainParams@getFeatures))

+            rankedFeatures <- originalFeatures[rankedFeaturesIndices, ]

+          else rankedFeatures <- S4Vectors::mcols(measurementsTrain)[rankedFeaturesIndices, featureColumns] 

+        }

       }

     } else { rankedFeatures <- NULL}

     if(!is.null(selectedFeaturesIndices))

@@ -297,9 +313,16 @@ input data. Autmomatically reducing to smaller number.")

         else selectedFeatures <- colnames(measurementsTrain)[selectedFeaturesIndices]

       } else {

         featureColumns <- na.omit(match(c("assay", "feature"), colnames(S4Vectors::mcols(measurementsTrain))))  

-        if(is.null(modellingParams@trainParams@getFeatures))

-          selectedFeatures <- originalFeatures[selectedFeaturesIndices, ]

-        else selectedFeatures <- S4Vectors::mcols(measurementsTrain)[selectedFeaturesIndices, featureColumns]

+        if(length(featureColumns) == 1)

+        {

+          if(is.null(modellingParams@trainParams@getFeatures))

+            selectedFeatures <- originalFeatures[selectedFeaturesIndices]

+          else selectedFeatures <- S4Vectors::mcols(measurementsTrain)[selectedFeaturesIndices, featureColumns]            

+        } else {

+                if(is.null(modellingParams@trainParams@getFeatures))

+                  selectedFeatures <- originalFeatures[selectedFeaturesIndices, ]

+                else selectedFeatures <- S4Vectors::mcols(measurementsTrain)[selectedFeaturesIndices, featureColumns]            

+        }

       }

     } else { selectedFeatures <- NULL}

   } else { # Nested use in feature selection. No feature selection in inner execution, so ignore features. 

@@ -16,7 +16,8 @@

 #' containing either classes or time and event information about survival. If

 #' \code{measurements} is a \code{MultiAssayExperiment}, the names of the column (class) or

 #' columns (survival) in the table extracted by \code{colData(data)} that contain(s) the samples'

-#' outcome to use for prediction.

+#' outcome to use for prediction. If column names of survival information, time must be in first

+#' column and event status in the second.

 #' @param crossValParams An object of class \code{\link{CrossValParams}},

 #' specifying the kind of cross-validation to be done.

 #' @param modellingParams An object of class \code{\link{ModellingParams}},

@@ -13,7 +13,7 @@

 #' a matrix of pre-calculated metrics, for backwards compatibility.

 #' @param classes If \code{results} is a matrix, this is a factor vector of the

 #' same length as the number of columns that \code{results} has.

-#' @param comparison Default: "Classifier Name". The aspect of the experimental

+#' @param comparison Default: "auto". The aspect of the experimental

 #' design to compare. Can be any characteristic that all results share.

 #' @param metric Default: "Sample Error". The sample-wise metric to plot.

 #' @param featureValues If not NULL, can be a named factor or named numeric

@@ -44,6 +44,8 @@

 #' @param legendSize The size of the boxes in the legends.

 #' @param plot Logical. IF \code{TRUE}, a plot is produced on the current

 #' graphics device.

+#' @param ... Parameters not used by the \code{ClassifyResult} method that does

+#' list-packaging but used by the main \code{list} method.

 #' @return A plot is produced and a grob is returned that can be saved to a

 #' graphics device.

 #' @author Dario Strbenac

@@ -82,11 +84,17 @@

 setGeneric("samplesMetricMap", function(results, ...)

 standardGeneric("samplesMetricMap"))

+#' @rdname samplesMetricMap

+#' @export

+setMethod("samplesMetricMap", "ClassifyResult", function(results, ...) {

+    samplesMetricMap(list(assay = results), ...)

+})

+

 #' @rdname samplesMetricMap

 #' @export

 setMethod("samplesMetricMap", "list", 

           function(results,

-                   comparison = "Classifier Name",

+                   comparison = "auto",

                    metric = c("Sample Error", "Sample Accuracy", "Sample C-index"),

                    featureValues = NULL, featureName = NULL,

                    metricColours = list(c("#3F48CC", "#6F75D8", "#9FA3E5", "#CFD1F2", "#FFFFFF"),

@@ -103,6 +111,20 @@ setMethod("samplesMetricMap", "list",

     stop("The package 'gridExtra' could not be found. Please install it.")       

   if(!requireNamespace("gtable", quietly = TRUE))

     stop("The package 'gtable' could not be found. Please install it.")

+  

+  characteristicsCounts <- table(unlist(lapply(results, function(result) result@characteristics[["characteristic"]])))

+  if(comparison == "auto")

+  {

+    if(max(characteristicsCounts) == length(results))

+    { # Choose a characteristic which varies the most across the results.

+      candidates <- names(characteristicsCounts)[characteristicsCounts == length(results)]

+      allCharacteristics <- do.call(rbind, lapply(results, function(result) result@characteristics))

+      distinctValues <- by(allCharacteristics[, "value"], allCharacteristics[, "characteristic"], function(values) length(unique(values)))

+      comparison <- names(distinctValues)[which.max(distinctValues)][1]

+    } else {

+      stop("No characteristic is present for all results but must be.")

+    }

+  }

   resultsWithComparison <- sum(sapply(results, function(result) any(result@characteristics[, "characteristic"] == comparison)))

   if(resultsWithComparison < length(results))

     stop("Not all results have comparison characteristic ", comparison, ' but need to.')

@@ -7,6 +7,7 @@

 \alias{calcExternalPerformance,factor,factor-method}

 \alias{calcExternalPerformance,Surv,numeric-method}

 \alias{calcCVperformance,ClassifyResult-method}

+\alias{calcExternalPerformance,factor,tabular-method}

 \title{Add Performance Calculations to a ClassifyResult Object or Calculate for a

 Pair of Factor Vectors}

 \usage{

@@ -24,6 +25,12 @@ Pair of Factor Vectors}

   performanceType = "C-index"

 )

+\S4method{calcExternalPerformance}{factor,tabular}(

+  actualOutcome,

+  predictedOutcome,

+  performanceType = "AUC"

+)

+

 \S4method{calcCVperformance}{ClassifyResult}(

   result,

   performanceType = c("Balanced Accuracy", "Balanced Error", "Error", "Accuracy",

@@ -131,7 +131,8 @@ or a list of these objects containing the data.}

 same length as the number of samples in \code{measurements} or a character vector of length 1 containing the

 column name in \code{measurements} if it is a \code{\link{DataFrame}}. Or a \code{\link{Surv}} object or a character vector of

 length 2 or 3 specifying the time and event columns in \code{measurements} for survival outcome. If \code{measurements} is a

-\code{\link{MultiAssayExperiment}}, the column name(s) in \code{colData(measurements)} representing the outcome.}

+\code{\link{MultiAssayExperiment}}, the column name(s) in \code{colData(measurements)} representing the outcome.  If column names

+of survival information, time must be in first column and event status in the second.}

 \item{...}{Parameters passed into \code{\link{prepareData}} which control subsetting and filtering of input data.}

@@ -140,15 +141,17 @@ or assays. If a numeric vector these will be optimised over using \code{selectio

 a different number of features will be used for each assay. If a named list of vectors, the respective number of features will be optimised over. 

 Set to NULL or "all" if all features should be used.}

-\item{selectionMethod}{A character vector of feature selection methods to compare. If a named character vector with names corresponding to different assays, 

-and performing multiview classification, the respective classification methods will be used on each assay.}

+\item{selectionMethod}{Default: "auto". A character vector of feature selection methods to compare. If a named character vector with names corresponding to different assays, 

+and performing multiview classification, the respective classification methods will be used on each assay. If \code{"auto"} t-test (two categories) / F-test (three or more categories) ranking

+and top \code{nFeatures} optimisation is done. Otherwise, the ranking method is per-feature Cox proportional hazards p-value.}

 \item{selectionOptimisation}{A character of "Resubstitution", "Nested CV" or "none" specifying the approach used to optimise \code{nFeatures}.}

 \item{performanceType}{Performance metric to optimise if classifier has any tuning parameters.}

-\item{classifier}{A character vector of classification methods to compare. If a named character vector with names corresponding to different assays, 

-and performing multiview classification, the respective classification methods will be used on each assay.}

+\item{classifier}{Default: \code{"auto"}. A character vector of classification methods to compare. If a named character vector with names corresponding to different assays, 

+and performing multiview classification, the respective classification methods will be used on each assay. If \code{"auto"}, then a random forest is used for a classification

+task or Cox proportional hazards model for a survival task.}

 \item{multiViewMethod}{A character vector specifying the multiview method or data integration approach to use.}

@@ -167,7 +170,8 @@ with each element being a vector of assays to combine. Special value \code{"all"

 \item{outcomeTrain}{For the \code{train} function, either a factor vector of classes, a \code{\link{Surv}} object, or

 a character string, or vector of such strings, containing column name(s) of column(s)

-containing either classes or time and event information about survival.}

+containing either classes or time and event information about survival. If column names

+of survival information, time must be in first column and event status in the second.}

 \item{assayIDs}{A character vector for assays to train with. Special value \code{"all"}

 uses all assays in the input object.}

deleted file mode 100644

@@ -1,26 +0,0 @@

-% Generated by roxygen2: do not edit by hand

-% Please edit documentation in R/crossValidate.R

-\name{generateCrossValParams}

-\alias{generateCrossValParams}

-\title{A function to generate a CrossValParams object}

-\usage{

-generateCrossValParams(nRepeats, nFolds, nCores, selectionOptimisation)

-}

-\arguments{

-\item{nRepeats}{A numeric specifying the the number of repeats or permutations to use for cross-validation.}

-

-\item{nFolds}{A numeric specifying the number of folds to use for cross-validation.}

-

-\item{nCores}{A numeric specifying the number of cores used if the user wants to use parallelisation.}

-

-\item{selectionOptimisation}{A character of "Resubstitution", "Nested CV" or "none" specifying the approach used to optimise \code{nFeatures}.}

-}

-\value{

-CrossValParams object

-}

-\description{

-A function to generate a CrossValParams object

-}

-\examples{

-CVparams <- generateCrossValParams(nRepeats = 20, nFolds = 5, nCores = 8, selectionOptimisation = "none")

-}

deleted file mode 100644

@@ -1,61 +0,0 @@

-% Generated by roxygen2: do not edit by hand

-% Please edit documentation in R/crossValidate.R

-\name{generateModellingParams}

-\alias{generateModellingParams}

-\title{A function to generate a ModellingParams object}

-\usage{

-generateModellingParams(

-  assayIDs,

-  measurements,

-  nFeatures,

-  selectionMethod,

-  selectionOptimisation,

-  performanceType = "auto",

-  classifier,

-  multiViewMethod = "none"

-)

-}

-\arguments{

-\item{assayIDs}{A vector of data set identifiers as long at the number of data sets.}

-

-\item{measurements}{Either a \code{\link{DataFrame}}, \code{\link{data.frame}}, \code{\link{matrix}}, \code{\link{MultiAssayExperiment}} 

-or a list of these objects containing the data.}

-

-\item{nFeatures}{The number of features to be used for classification. If this is a single number, the same number of features will be used for all comparisons

-or assays. If a numeric vector these will be optimised over using \code{selectionOptimisation}. If a named vector with the same names of multiple assays, 

-a different number of features will be used for each assay. If a named list of vectors, the respective number of features will be optimised over. 

-Set to NULL or "all" if all features should be used.}

-

-\item{selectionMethod}{A character vector of feature selection methods to compare. If a named character vector with names corresponding to different assays, 

-and performing multiview classification, the respective classification methods will be used on each assay.}

-

-\item{selectionOptimisation}{A character of "Resubstitution", "Nested CV" or "none" specifying the approach used to optimise \code{nFeatures}.}

-

-\item{performanceType}{Performance metric to optimise if classifier has any tuning parameters.}

-

-\item{classifier}{A character vector of classification methods to compare. If a named character vector with names corresponding to different assays, 

-and performing multiview classification, the respective classification methods will be used on each assay.}

-

-\item{multiViewMethod}{A character vector specifying the multiview method or data integration approach to use.}

-}

-\value{

-ModellingParams object

-}

-\description{

-A function to generate a ModellingParams object

-}

-\examples{

-data(asthma)

-# First make a toy example assay with multiple data types. We'll randomly assign different features to be clinical, gene or protein.

-set.seed(51773)

-measurements <- DataFrame(measurements, check.names = FALSE) 

-mcols(measurements)$assay <- c(rep("clinical",20),sample(c("gene", "protein"), ncol(measurements)-20, replace = TRUE))

-mcols(measurements)$feature <- colnames(measurements)

-modellingParams <- generateModellingParams(assayIDs = c("clinical", "gene", "protein"),

-                                          measurements = measurements, 

-                                          nFeatures = list(clinical = 10, gene = 10, protein = 10),

-                                          selectionMethod = list(clinical = "t-test", gene = "t-test", protein = "t-test"),

-                                          selectionOptimisation = "none",

-                                          classifier = "randomForest",

-                                          multiViewMethod = "merge")

-}

@@ -34,7 +34,8 @@ are passed into and used by the \code{DataFrame} method or passed onwards to \co

 \item{outcomeTrain}{Either a factor vector of classes, a \code{\link{Surv}} object, or

 a character string, or vector of such strings, containing column name(s) of column(s)

-containing either classes or time and event information about survival.}

+containing either classes or time and event information about survival. If column names

+of survival information, time must be in first column and event status in the second.}

 \item{measurementsTest}{Same data type as \code{measurementsTrain}, but only the test

 samples.}

@@ -35,7 +35,8 @@ a character string, or vector of such strings, containing column name(s) of colu

 containing either classes or time and event information about survival. If

 \code{measurements} is a \code{MultiAssayExperiment}, the names of the column (class) or

 columns (survival) in the table extracted by \code{colData(data)} that contain(s) the samples'

-outcome to use for prediction.}

+outcome to use for prediction. If column names of survival information, time must be in first

+column and event status in the second.}

 \item{crossValParams}{An object of class \code{\link{CrossValParams}},

 specifying the kind of cross-validation to be done.}

@@ -4,11 +4,14 @@

 \alias{samplesMetricMap}

 \alias{samplesMetricMap,list-method}

 \alias{samplesMetricMap,matrix-method}

+\alias{samplesMetricMap,ClassifyResult-method}

 \title{Plot a Grid of Sample Error Rates or Accuracies}

 \usage{

+\S4method{samplesMetricMap}{ClassifyResult}(results, ...)

+

 \S4method{samplesMetricMap}{list}(

   results,

-  comparison = "Classifier Name",

+  comparison = "auto",

   metric = c("Sample Error", "Sample Accuracy", "Sample C-index"),

   featureValues = NULL,

   featureName = NULL,

@@ -55,7 +58,10 @@

 \item{results}{A list of \code{\link{ClassifyResult}} objects. Could also be

 a matrix of pre-calculated metrics, for backwards compatibility.}

-\item{comparison}{Default: "Classifier Name". The aspect of the experimental

+\item{...}{Parameters not used by the \code{ClassifyResult} method that does

+list-packaging but used by the main \code{list} method.}

+

+\item{comparison}{Default: "auto". The aspect of the experimental

 design to compare. Can be any characteristic that all results share.}

 \item{metric}{Default: "Sample Error". The sample-wise metric to plot.}

@@ -1,35 +1,16 @@

 ---

-title: "An Introduction to **ClassifyR**"

+title: "Getting Started with ClassifyR"

 author: Dario Strbenac, Ellis Patrick, Graham Mann, Jean Yang, John Ormerod <br>

         The University of Sydney, Australia.

 output: 

   BiocStyle::html_document:

+    css: style.css

     toc: true

 vignette: >

   %\VignetteEngine{knitr::rmarkdown}

   %\VignetteIndexEntry{An Introduction to the ClassifyR Package}

 ---

-<style>

-    body .main-container {

-        max-width: 1600px;

-    }

-    p {

-      padding: 20px;

-    }

-    .table {

-      border: 2px solid #e64626;

-    }

-    .table>thead>tr>th

-    {

-     border-bottom: 2px solid #e64626;

-     border-right: 1px solid black;

-    }

-    .table>tbody>tr>td {

-     border-right: 1px solid black;

-    }    

-</style>

-

 ```{r, echo = FALSE, results = "asis"}

 options(width = 130)

 ```

@@ -50,7 +31,7 @@ Driver functions can use parallel processing capabilities in R to speed up cross

 ```{r, echo = FALSE}

-htmltools::img(src = knitr::image_uri("ClassifyRprocedure.png"), 

+htmltools::img(src = knitr::image_uri("images/ClassifyRprocedure.png"), 

                style = 'margin-left: auto;margin-right: auto')

 ```

@@ -90,7 +71,7 @@ Data Type | $n \times p$ | $p \times n$

 <span style="font-family: 'Courier New', monospace;">MultiAssayExperiment</span> |  | ✔

 <span style="font-family: 'Courier New', monospace;">list</span> of <span style="font-family: 'Courier New', monospace;">data.frame</span>s | ✔ | 

-*crossValidate* must also be supplied with *outcomes*, which represents the prediction to be made in a variety of possible ways.

+*crossValidate* must also be supplied with *outcome*, which represents the prediction to be made in a variety of possible ways.

 * A *factor* that contains the class label for each observation. *classes* must be of length $n$.

 * A *character* of length 1 that matches a column name in a data frame which holds the classes. The classes will automatically be removed before training is done.

new file mode 100755

Binary files /dev/null and b/vignettes/images/ClassifyRprocedure.png differ

new file mode 100644

@@ -0,0 +1,9 @@

+body .main-container {max-width: 1600px;}

+p {padding: 20px;}

+.table {border: 2px solid #e64626;}

+.table>thead>tr>th

+{

+  border-bottom: 2px solid #e64626;

+  border-right: 1px solid black;

+}

+.table>tbody>tr>td {border-right: 1px solid black;}    

\ No newline at end of file