This package provides tools to check whether a vector of p-values respects the assumptions of classical FDR (false discovery rate) control procedures.
It is built to be easily used by non-statisticians in the context of quantitative proteomics (yet, it can be applied in other contexts).
Concretely, it allows estimating the proportion of true null hypotheses (i.e. proportion of non-differentially abundant proteins or peptides in a relative quantification experiment), as well as checking whether the p-values are adequately distributed for further FDR control.
In addition, the package allows performing an adequately chosen adaptive FDR control procedure to get adjusted p-values.
A tutorial giving a practical introduction to this package is available in the supplementary material of Giai Gianetto et al. (2016).
|Depends:||multtest, qvalue, limma, MESS, graphics, stats|
This package is composed of three functions that take as input a vector of p-values resulting from multiple two-sided hypothesis testing (such as multiple t-tests for equal means for instance).
First, the function
estim.pi0 allows determining the proportion of true null hypotheses among the set of tests using eight different estimation methods proposed in literature.
Second, the function
calibration.plot proposes an intuitive plot of the p-values, so as to visually assess their behavior and well-calibration.
Third, the function
adjust.p allows obtaining adjusted p-values in view to perform an adaptive FDR control from a chosen level.
Two proteomic datasets named
LFQRatio25 allow to use these functions in a concrete framework where the proportion of non-differentially abundant proteins is known.
Quentin Giai Gianetto, Florence Combes, Claire Ramus, Christophe Bruley, Yohann Cout<c3><a9>, Thomas Burger
Maintainer: Quentin Giai Gianetto <[email protected]>
Giai Gianetto, Q., Combes, F., Ramus, C., Bruley, C., Cout<c3><a9>, Y., Burger, T. (2016). Calibration plot for proteomics: A graphical tool to visually check the assumptions underlying FDR control in quantitative experiments. Proteomics, 16(1), 29-32.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.