We develop a general framework for performing large-scale significance testing in the presence of arbitrarily strong dependence. We derive a low-dimensional set of random vectors, called a dependence kernel, that fully captures the dependence structure in an observed high-dimensional dataset. This result shows a surprising reversal of the "curse of dimensionality" in the high-dimensional hypothesis testing setting. We show theoretically that conditioning on a dependence kernel is sufficient to render statistical tests independent regardless of the level of dependence in the observed data. This framework for multiple testing dependence has implications in a variety of common multiple testing problems, such as in gene expression studies, brain imaging, and spatial epidemiology.
Description
A general framework for multiple testing dependence. - PubMed - NCBI
%0 Journal Article
%1 Leek:2008:Proc-Natl-Acad-Sci-U-S-A:19033188
%A Leek, J T
%A Storey, J D
%D 2008
%J Proc Natl Acad Sci U S A
%K batch-effect bayes fulltext normalization rna-seq shouldread software
%N 48
%P 18718-18723
%R 10.1073/pnas.0808709105
%T A general framework for multiple testing dependence
%U https://www.ncbi.nlm.nih.gov/pubmed/19033188
%V 105
%X We develop a general framework for performing large-scale significance testing in the presence of arbitrarily strong dependence. We derive a low-dimensional set of random vectors, called a dependence kernel, that fully captures the dependence structure in an observed high-dimensional dataset. This result shows a surprising reversal of the "curse of dimensionality" in the high-dimensional hypothesis testing setting. We show theoretically that conditioning on a dependence kernel is sufficient to render statistical tests independent regardless of the level of dependence in the observed data. This framework for multiple testing dependence has implications in a variety of common multiple testing problems, such as in gene expression studies, brain imaging, and spatial epidemiology.
@article{Leek:2008:Proc-Natl-Acad-Sci-U-S-A:19033188,
abstract = {We develop a general framework for performing large-scale significance testing in the presence of arbitrarily strong dependence. We derive a low-dimensional set of random vectors, called a dependence kernel, that fully captures the dependence structure in an observed high-dimensional dataset. This result shows a surprising reversal of the "curse of dimensionality" in the high-dimensional hypothesis testing setting. We show theoretically that conditioning on a dependence kernel is sufficient to render statistical tests independent regardless of the level of dependence in the observed data. This framework for multiple testing dependence has implications in a variety of common multiple testing problems, such as in gene expression studies, brain imaging, and spatial epidemiology.},
added-at = {2019-09-03T12:08:53.000+0200},
author = {Leek, J T and Storey, J D},
biburl = {https://www.bibsonomy.org/bibtex/22a3fa7c7f8e246c123ccb146c98b80cf/marcsaric},
description = {A general framework for multiple testing dependence. - PubMed - NCBI},
doi = {10.1073/pnas.0808709105},
interhash = {ae98ae621a6fe5697b76598988a13d07},
intrahash = {2a3fa7c7f8e246c123ccb146c98b80cf},
journal = {Proc Natl Acad Sci U S A},
keywords = {batch-effect bayes fulltext normalization rna-seq shouldread software},
month = dec,
number = 48,
pages = {18718-18723},
pmid = {19033188},
timestamp = {2019-09-03T12:08:53.000+0200},
title = {A general framework for multiple testing dependence},
url = {https://www.ncbi.nlm.nih.gov/pubmed/19033188},
volume = 105,
year = 2008
}