wrProteo (version 1.5.0)

matrixNAinspect: Histogram of content of NAs in matrix

Description

matrixNAinspect makes histograms of the full data and shows sub-population of NA-neighbour values. The aim of this function is to investigate the nature of NA values in matrix (of experimental measures) where replicate measurements are available. If a given element was measured twice, and one of these measurements revealed a NA while the other one gave a (finite) numeric value, the non-NA-value is considered a NA-neighbour. The subpopulation of these NA-neighbour values will then be highlighted in the resulting histogram. In a number of experimental settiongs some actual measurements may not meet an arbitrary defined baseline (as 'zero') or may be too low to be distinguishable from noise that associated measures were initially recorded as NA. In several types of measurments in proteomics and transcriptomics this may happen. So this fucntion allows to collect all NA-neighbour values and compare them to the global distribution of the data to investigate if NA-neighbours are typically very low values. In case of data with multiple replicates NA-neighbour values may be distinguished for the case of 2 NA per group/replicate-set. The resulting plots are typically used to decide if and how NA values may get replaced by imputed random values or wether measues containing NA-values should rather me omitted. Of course, such decisions do have a strong impact on further steps of data-analysis and should be performed with care.

Usage

matrixNAinspect(
  dat,
  gr,
  retnNA = TRUE,
  xLab = NULL,
  tit = NULL,
  xLim = NULL,
  silent = FALSE,
  callFrom = NULL
)

Arguments

dat

(matrix or data.frame) main numeric data

gr

(charcter or factor) grouping of columns of dat indicating who is a replicate of whom (ie the length of 'gr' must be equivalent to the number of columns in 'dat')

retnNA

(logical) report number of NAs in graphic

xLab

(character) custom x-label

tit

(character) custom title

xLim

(numerical,length=2) custom x-axis limits

silent

(logical) suppress messages

callFrom

(character) allow easier tracking of messages produced

Value

graphic only

See Also

hist, na.fail, naOmit

Examples

# NOT RUN {
set.seed(2013)
datT6 <- matrix(round(rnorm(300)+3,1),ncol=6,dimnames=list(paste("li",1:50,sep=""),letters[19:24]))
datT6 <- datT6 +matrix(rep(1:nrow(datT6),ncol(datT6)),ncol=ncol(datT6))
datT6[6:7,c(1,3,6)] <- NA
datT6[which(datT6 < 11 & datT6 > 10.5)] <- NA
datT6[which(datT6 < 6 & datT6 > 5)] <- NA
datT6[which(datT6 < 4.6 & datT6 > 4)] <- NA
matrixNAinspect(datT6,gr=gl(2,3)) 
# }