receptLoss: Identify genes with expression loss

In receptLoss: Unsupervised Identification of Genes with Expression Loss in Subsets of Tumors

Description

This function allows you to identify genes with loss of expression

Usage

receptLoss(exprMatrNml, exprMatrTum, nSdBelow, minPropPerGroup)

Arguments

exprMatrNml	A matrix of expression values from normal tissue. Each row is a gene, and each column is a patient or sample. Genes should be in same order as exprMatrTum.
exprMatrTum	A matrix of expression values from tumor tissue. Each row is a gene, and each column is a patient or sample. Genes should be in same order as exprMatrNml.
nSdBelow	The number of SD below the mean of the adjacent normal tissue to set the boundary between tumor subgroups.
minPropPerGroup	A value between 0-1 that represents the minimum proportion of samples that should be present in each of the two subgroups (defined by the boundary set by nSdBelow) for a particular gene.

Value

a nx7 matrix, with n equaling the number of genes. The columns are as follows:

• geneNm - the gene name

• lowerBound - the lower bound, or the value 'nSdBelow' the mean of the normal tissue expression data.

• propTumLessThBound - the proportion of tumor samples with expression levels less than 'lowerBound'

• muAb - "mu above", the mean expression value of tumors greater than (ie above) the 'lowerBound'.

• 'muBl' - "mu below", the mean expression value of tumors less than (ie below) the 'lowerBound'.

• 'deltaMu' - the difference between 'muAb' and 'muBl'.

• meetsMinPropPerGrp - a logical indicating whether the proportion of samples in each group is greater than that set by 'minPropPerGroup'.

Examples

exprMatrNml <- matrix(abs(rnorm(100, mean=2)), nrow=10)
exprMatrTum <- matrix(abs(rnorm(100)), nrow=10)
geneNames <- paste0(letters[seq_len(nrow(exprMatrNml))],
seq_len(nrow(exprMatrNml)))
rownames(exprMatrNml) <- rownames(exprMatrTum) <- geneNames
nSdBelow <- 2
minPropPerGroup <- .2
rl <- receptLoss(exprMatrNml, exprMatrTum, nSdBelow, minPropPerGroup)
head(rl)

receptLoss documentation built on Nov. 8, 2020, 5:59 p.m.