remove_background: Assess background expression

In calebclass/NanoTube: An Easy Pipeline for NanoString nCounter Data Analysis

Assess background expression

Description

Compare endogenous gene expression data against negative control genes and remove data for genes that fail the comparison. This step is conducted within processNanostringData, when normalization is set to "nCounter".

Usage

remove_background(
  dat,
  mode = c("threshold", "t.test"),
  numSD,
  proportionReq,
  pval,
  subtract = FALSE
)

Arguments

dat	Positive control-scaled NanoString data
mode	Either "threshold" (default) or "t.test". If "threshold", requires proportionReq of samples to have expression numSD standard deviations among the mean of negative control genes. If "t.test", each gene will be compared with all negative control genes in a one-sided two-sample t-test.
numSD	Number of standard deviations above mean of negative control genes to used as background threshold for each sample: mean(negative_controls) + numSD * sd(negative_controls). Required if mode == "threshold" or subtract == TRUE
proportionReq	Required proportion of sample expressions exceeding the sample background threshold to include gene in further analysis. Required if mode == "threshold" or subtract == TRUE
pval	p-value (from one-sided t-test) threshold to declare gene expression above background expression level. Genes with p-values above this level are removed from further analysis. Required if mode == "t.test"
subtract	Should calculated background levels be subtracted from reported expressions? If TRUE, will subtract mean+numSD*sd of the negative controls from the endogenous genes, and then set negative values to zero (default FALSE).

Value

NanoString data, with genes removed that fail the comparison test against negative control genes. Expression levels are updated for all genes if subtract == TRUE.

Examples

example_data <- system.file("extdata", "GSE117751_RAW", package = "NanoTube")

# Load data and positive control normalization
dat <- read_merge_rcc(list.files(example_data, full.names = TRUE))
dat <- normalize_pos_controls(dat)

# Remove endogenous genes that fail to reject the null hypothesis
# in a one-sided t test against negative control genes with p < 0.05.
dat <- remove_background(dat, mode = "t.test", pval = 0.05)

# Remove endogenous genes where fewer than 25% of samples have an expression
# 2 standard deviations above the average negative control gene. Also, 
# subtract this background level (mean + 2*sd) from endogenous genes.
dat <- remove_background(dat, mode = "threshold", 
                         numSD = 2, proportionReq = 0.25, subtract = TRUE)

calebclass/NanoTube documentation built on Nov. 21, 2023, 12:31 p.m.