betaUniformPvalueSumTest | R Documentation |
A variant of Fisher's method for P-value aggregation, but with a parametrically derived null-hypothesis that reflects the background signal in the data
betaUniformPvalueSumTest(testPvalues, betaUniformFit, na.rm = TRUE) ## S4 method for signature 'numeric,BetaUniformModel' betaUniformPvalueSumTest(testPvalues, betaUniformFit, na.rm = TRUE) ## S4 method for signature 'Pvalues,BetaUniformModel' betaUniformPvalueSumTest(testPvalues, betaUniformFit, na.rm = TRUE)
testPvalues |
A numerical vector of P-values to be tested. |
betaUniformFit |
A beta-uniform model of the P-value distribution. |
na.rm |
Remove NA - default TRUE |
Authors in the literature (e.g. Luo et al.) have used a chi-squared test for gene set enrichment with the null-hypothesis that all genes have P-values derived from a pure noise (uniform) distribution. This is typically demonstrably not true and will result in an inflated rate of false positives
combinedPval A single p-value combining the results of multiple hypothesis tests into one
testPvalues = numeric,betaUniformFit = BetaUniformModel
: Convert to P-values on the fly
testPvalues = Pvalues,betaUniformFit = BetaUniformModel
: Compute a P-value with the null-hypothesis being that they are drawn randomly from a mixture distribution
Luo, W., Friedman, M. S., Shedden, K., Hankenson, K. D., & Woolf, P. J. (2009). GAGE: generally applicable gene set enrichment for pathway analysis. BMC Bioinformatics
https://en.wikipedia.org/wiki/Phase-type_distribution
Sidje RB. Expokit: a software package for computing matrix exponentials. ACM Trans Math Softw. 1998;24: 130–156.
fishersCombindedProbabilityTest
fitBetaUniformParameters
BioNet::fitBumModel
Other P-value sum tests:
fishersPvalueSumTest()
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.