Assess the limit of detection of a given method

Share:

Description

This function assesses the limit of detection in one of two ways: (1) the distribution of expression estimates stratified by the proportion of poor quality values within replicates, (2) the average vs expected expression for the two most diluted sample types.

Usage

1
limitOfDetection(object, qcThreshold, plotType=c("boxplot","scatterplot","MAplot"))

Arguments

object

a list containing two elements: ct (the expression estiamtes) and qc (quality scores)

qcThreshold

a numeric threshold corresponding to object1$qc below which values are considered low quality.

plotType

the desired output type – boxplot is option (1); scatterplot is option (2), an MA-plot is option (3).

Value

This function assesses the limit of detection in several ways. If plotType is boxplot, then boxplots of expression estimates stratified by the proportion of poor quality values within replicates is displayed. The function also outputs a list with the values plotted in each box of the boxplot. If plotType is scatterplot, then the average within replicate expression vs expected expression (based on pure sample expression) is displayed for the 0.1/0.1 dilution and 0.01/0.01 dilution. If plotType is MAplot, then the difference in expression (average within replicate expression - expected expression) is displayed for the 0.1/0.1 dilution and 0.01/0.01 dilution. For both plotTypes, scatterplot and MAplot, the function outputs a matrix containing estimates of the limit of detection for four different tolerances. Specifically, the two columns correspond to the two dilutions (0.1/0.1 and 0.01/0.01) and rows correspond to the median difference between the observed and expected values. The values in the matrix are the expected expression values such that the median absolute difference of all larger expected expression values is approximately equal to the given tolerance.

Author(s)

Matthew N. McCall

Examples

1
2
3
4
  data(lifetech)
  tmp <- limitOfDetection(object=lifetech,qcThreshold=1.25)
  data(qpcRdefault)
  limitOfDetection(object=qpcRdefault,qcThreshold=0.99,plotType="scatter")