Passes the output of generateSimulatedData() to differential expression tests, picking either t-tests or ANOVA for data with only two conditions or multiple conditions, respectively.

subDiffEx(tempData)

subDiffExttest(countMatrix, class.labels, test.type = "t.equalvar")

subDiffExANOVA(countMatrix, condition)

Arguments

tempData: Matrix. The output of generateSimulatedData(), where the first row contains condition labels.
countMatrix: Matrix. A simulated counts matrix, sans labels.
class.labels: Factor. The condition labels for the simulated cells. Will be coerced into 1's and 0's.
test.type: Type of test to perform. The default is t.equalvar.
condition: Factor. The condition labels for the simulated cells.

Value

subDiffEx(): A vector of fdr-adjusted p-values for all genes. Nonviable results (such as for genes with 0 counts in a simulated dataset) are coerced to 1. subDiffExttest(): A vector of fdr-adjusted p-values for all genes. Nonviable results (such as for genes with 0 counts in a simulated dataset) are coerced to 1. subDiffExANOVA(): A vector of fdr-adjusted p-values for all genes. Nonviable results (such as for genes with 0 counts in a simulated dataset) are coerced to 1.

Functions

subDiffEx: Get PCA components for a SCtkE object
subDiffExttest: Runs t-tests on all genes in a simulated dataset with 2 conditions, and adjusts for FDR.
subDiffExANOVA: Runs ANOVA on all genes in a simulated dataset with more than 2 conditions, and adjusts for FDR.

Examples

data("mouseBrainSubsetSCE")
res <- generateSimulatedData(
         totalReads = 1000, cells=10,
         originalData = assay(mouseBrainSubsetSCE, "counts"),
         realLabels = colData(mouseBrainSubsetSCE)[, "level1class"])
tempSigDiff <- subDiffEx(res)

data("mouseBrainSubsetSCE")
#sort first 100 expressed genes
ord <- rownames(mouseBrainSubsetSCE)[
  order(rowSums(assay(mouseBrainSubsetSCE, "counts")),
        decreasing = TRUE)][1:100]
#subset to those first 100 genes
subset <- mouseBrainSubsetSCE[ord, ]
res <- generateSimulatedData(totalReads = 1000, cells=10,
                             originalData = assay(subset, "counts"),
                             realLabels = colData(subset)[, "level1class"])
realLabels <- res[1, ]
output <- res[-1, ]
fdr <- subDiffExttest(output, realLabels)

data("mouseBrainSubsetSCE")
#sort first 100 expressed genes
ord <- rownames(mouseBrainSubsetSCE)[
  order(rowSums(assay(mouseBrainSubsetSCE, "counts")),
        decreasing = TRUE)][1:100]
# subset to those first 100 genes
subset <- mouseBrainSubsetSCE[ord, ]
res <- generateSimulatedData(totalReads = 1000, cells=10,
                             originalData = assay(subset, "counts"),
                             realLabels = colData(subset)[, "level2class"])
realLabels <- res[1, ]
output <- res[-1, ]
fdr <- subDiffExANOVA(output, realLabels)