calculatePvalue: p-value that two PWM-positions are from the same distribution

In DiffLogo: DiffLogo: A comparative visualisation of biooligomer motifs

Description

Calculates the p-value for the null-hypothesis that two given probability vectors p1, p2 calculated from n1/n2 observations arise from the same distribution

Usage

calculatePvalue(p1, p2, n1, n2, stackHeight = shannonDivergence,
  numberOfPermutations = 100, plotGammaDistributionFit = FALSE)

Arguments

p1	first probability vector with one probability for each symbol of the alphabet
p2	second probability vector with one probability for each symbol of the alphabet
n1	number of observations for the calculation of p1
n2	number of observations for the calculation of p2
stackHeight	function for the calculation of a divergence measure for two probability vectors
numberOfPermutations	the number of permutations to perform for the calculation of stackHeights
plotGammaDistributionFit	if TRUE the fit of a gamma distribution to the sampled stackHeights is plotted

Value

a numeric p-value

Author(s)

Hendrik Treutler

Examples

p1 <- c(0.2, 0.3, 0.1, 0.4)
p2 <- c(0.2, 0.1, 0.3, 0.4)
n1 <- 100
n2 <- 200
numberOfPermutations = 100
plotGammaDistributionFit = TRUE

pValue <- calculatePvalue(p1 = p1, p2 = p2, n1 = n1, n2 = n2, stackHeight = shannonDivergence, numberOfPermutations = numberOfPermutations, plotGammaDistributionFit = plotGammaDistributionFit)

Example output

Loading required package: cba
Loading required package: grid
Loading required package: proxy

Attaching package: ‘proxy’

The following objects are masked from ‘package:stats’:

    as.dist, dist

The following object is masked from ‘package:base’:

    as.matrix

DiffLogo documentation built on Nov. 8, 2020, 6:09 p.m.