calcPhenotype: Generate predicted drug sensitivity scores

In oncoPredict: Drug Response Modeling and Biomarker Discovery

Generate predicted drug sensitivity scores

Description

This function predicts a phenotype (drug sensitivity score) when provided with microarray or bulk RNAseq gene expression data of different platforms. The imputations are performed using ridge regression, training on a gene expression matrix where phenotype is already known. This function integrates training and testing datasets via a user-defined procedure, and power transforming the known phenotype.

Usage

calcPhenotype(
  trainingExprData,
  trainingPtype,
  testExprData,
  batchCorrect,
  powerTransformPhenotype = TRUE,
  removeLowVaryingGenes = 0.2,
  minNumSamples,
  selection = 1,
  printOutput,
  pcr = FALSE,
  removeLowVaringGenesFrom,
  report_pc = FALSE,
  cc = FALSE,
  percent = 80,
  rsq = FALSE,
  folder = FALSE,
  parallel = FALSE,
  cores = 1
)

Arguments

trainingExprData	The training data. A matrix of expression levels. rownames() are genes, colnames() are samples (cell line names or cosmic ides, etc.). rownames() must be specified and must contain the same type of gene ids as "testExprData"
trainingPtype	The known phenotype for "trainingExprData". This data must be a matrix with training samples as rows and drugs or phenotypes as columns. This matrix can contain NA values, that is ok (they are removed in the calcPhenotype() function).
testExprData	The test data where the phenotype will be estimated. It is a matrix of expression levels, rows contain genes and columns contain samples, "rownames()" must be specified and must contain the same type of gene ids as "trainingExprData".
batchCorrect	How should training and test data matrices be homogenized. Choices are "eb" (default) for ComBat, "qn" for quantile normalization, "standardize" for within-dataset z-score standardization, "rank", "rank_then_eb", or "none" for no homogenization.
powerTransformPhenotype	Should the phenotype be power transformed before we fit the regression model? Default to TRUE, set to FALSE if the phenotype is already known to be highly normal.
removeLowVaryingGenes	What proportion of low varying genes should be removed? 20 percent be default
minNumSamples	How many training and test samples are required. Print an error if below this threshold
selection	How should duplicate gene ids be handled. Default is -1 which asks the user. 1 to summarize by their or 2 to disguard all duplicates.
printOutput	Set to FALSE to supress output.
pcr	Indicates whether or not you'd like to use pcr for feature (gene) reduction. Options are 'TRUE' and 'FALSE'. If you indicate 'report_pc=TRUE' you need to also indicate 'pcr=TRUE'
removeLowVaringGenesFrom	Determine method to remove low varying genes. Options are 'homogenizeData' and 'rawData'.
report_pc	Indicates whether you want to output the training principal components. Options are 'TRUE' and 'FALSE'.
cc	Indicate if you want correlation coefficients for biomarker discovery.
percent	Indicate percent variability (of the training data) you'd like principal components to reflect if pcr=TRUE. Default is 80 for 80% These are the correlations between a given gene of interest across all samples vs. a given drug response across samples. These correlations can be ranked to obtain a ranked correlation to determine highly correlated drug-gene associations.
rsq	Indicate whether or not you want to output the R^2 values for the data you train on from true and predicted values. These values represent the percentage in which the optimal model accounts for the variance in the training data. Options are 'TRUE' and 'FALSE'.
folder	If TRUE, write calcPhenotype outputs to calcPhenotype_Output in the current working directory. The default is FALSE.
parallel	If TRUE, fit drug models in parallel after the shared homogenization and gene-filtering steps are complete. The default is FALSE.
cores	The number of cores to use when parallel is TRUE. Parallel execution uses forked processes via parallel::mclapply, which is not available for multicore execution on Windows PCs; on Windows, calcPhenotype will warn and run serially.

Value

A matrix of predicted drug response values. If rsq, cc, or report_pc is TRUE, returns a list containing the predictions and requested optional outputs. If folder is TRUE, the same object is returned invisibly after files are written.

Examples

set.seed(1)
genes <- paste0("gene", 1:30)
trainingExprData <- matrix(rnorm(30 * 8), nrow=30,
                           dimnames=list(genes, paste0("train", 1:8)))
testExprData <- matrix(rnorm(30 * 3), nrow=30,
                       dimnames=list(genes, paste0("test", 1:3)))
trainingPtype <- matrix(rnorm(8), ncol=1,
                        dimnames=list(colnames(trainingExprData), "drug1"))
predictions <- calcPhenotype(trainingExprData, trainingPtype, testExprData,
                             batchCorrect="none",
                             powerTransformPhenotype=FALSE,
                             removeLowVaryingGenes=0,
                             minNumSamples=0,
                             selection=1,
                             printOutput=FALSE,
                             pcr=FALSE,
                             removeLowVaringGenesFrom="rawData")
head(predictions)

oncoPredict documentation built on June 29, 2026, 5:07 p.m.