testDHI: Perform a permutation test on DHI

In pietrofranceschi/HomogenMSI: Calculate drug homogeneity value for MSI data

Description

This function perform a permutation testing to assess the significance of the Drug Homogeneity Index.

Usage

testDHI(drugImg,permutations = 100, Nu=1, QuantLevel=NULL, maskImg=NULL)

Arguments

drugImg	Input matrix containing the pixel intensities
permutations	Number of permutations used to construct the null distribution. default = 100
Nu	Lowest size-zone value used for DHI calculation. default = 1
QuantLevel	Maximum possible gray-levels in drug ion image. default = 0, i.e., original image
maskImg	Matrix identifying the pixels belonging to the tissue (1 tissue, 0 background). It must have the same size as drugImg

Value

A list containing the DHI calculated for the original and the permuted images

Author(s)

Mridula Prasad [e-mail]

References

https://github.com/pietrofranceschi/HomogenMSI

Examples

## load package
library(HomogenMSI)
data("DHIimages")

## Input drug and mask images

drugImg = DHIimages[[20]]
maskImg = DHIimages[[1]]
maskImg[maskImg !=0] =1

# test the significance of the DHI
output <- testDHI(drugImg,maskImg=maskImg)

# Plot the results
hist(output$DHIPermutedvalues, 
     xlim = range(c(output$DHIoriginalvalue,output$DHIPermutedvalues)), 
     col = "steelblue", 
     xlab = "DHI",
     main = "DHI Permutation test")
abline(v = output$DHIoriginalvalue, lty = 2)

pietrofranceschi/HomogenMSI documentation built on May 21, 2019, 9:22 a.m.