screenExpr: Screens data from the original expression files

In syspremed/PanNETassigner: A cross-species analysis in pancreatic neuroendocrine tumors reveals molecular subtypes with distinctive clinical, metastatic, developmental, and metabolic characteristics.

Description

Screens for genes with sample variance above user-specified threshold. Consolidates data for duplicate genes by averaging across duplicates.

Usage

screenExpr(Yexpr, sdCutoff)

Arguments

Yexpr	Expression data.
sdCutoff	std dev threshold.

Details

Screens for genes with sample variance above user-specified threshold. Consolidates data for duplicate genes by averaging across duplicates.

Value

Returns expression data with unique gene names and variance greater than psecified threshold

Author(s)

Anguraj Sadanandam [e-mail]

References

Anguraj Sadanandam, et all (2015). A cross-species analysis in pancreatic neuroendocrine tumors reveals molecular subtypes with distinctive clinical, metastatic, developmental, and metabolic characteristics. Cancer Discovery.

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (Yexpr, sdCutoff) 
{
    data.all <- Yexpr
    print(paste("    Read expression data for ", dim(data.all)[1], 
        " genes.", sep = ""))
    data.mapped <- data.all[toupper(data.all[, 1]) != "UNMAPPED" & 
        !is.na(data.all[, 1]), ]
    print(paste("    Mapped data to ", dim(data.mapped)[1], " genes IDs.", 
        sep = ""))
    nLastCol <- dim(data.mapped)[2]
    print("Removing genes with low sample variance...")
    sampleVar <- apply(data.mapped[, 2:nLastCol], 1, var)
    idxHighVar <- sampleVar > sdCutoff^2
    data.highVar <- data.mapped[idxHighVar, ]
    sampleVar.highVar <- sampleVar[idxHighVar]
    print(paste("    Found ", dim(data.highVar)[1], " genes exceeding SD threshold of ", 
        sdCutoff, ".", sep = ""))
    nGenesHighVar <- dim(data.highVar)[1]
    genesUnique <- as.vector(unique(data.highVar[, 1]))
    nGenesUnique <- length(genesUnique)
    nSamples <- dim(data.highVar)[2] - 1
    data <- array(dim = c(nGenesUnique, nLastCol))
    data[, 1] <- genesUnique
    colnames(data) <- colnames(data.highVar)
    print("Removing duplicate genes (selecting for max standard deviation)...")
    for (gene in genesUnique) {
        idxGenes <- seq(along = 1:nGenesHighVar)[data.highVar[, 
            1] == gene]
        data.slice <- data.highVar[idxGenes, 2:nLastCol]
        if (length(idxGenes) > 1) {
            idxMaxVar <- which.max(sampleVar.highVar[idxGenes])
            data.slice <- data.slice[idxMaxVar, ]
        }
        data[data[, 1] == gene, 2:nLastCol] <- as.matrix(data.slice)
    }
    print(paste("    ", nGenesUnique, " unique genes IDs remain.", 
        sep = ""))
    print("Screened data is the output:... ")
    print("")
    rm(data.all, data.mapped, data.highVar)
    sdata <- data
    sdata
  }

syspremed/PanNETassigner documentation built on May 31, 2019, 12:48 a.m.