A general framework for multiple testing dependence
Jeffrey T. Leek, John D. Storey · Proceedings of the National Academy of Sciences · 2008
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.
Cited by 374 works. doi:10.1073/pnas.0808709105
Authors
References
- Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing
- A Simple Sequentially Rejective Multiple Test Procedure
- In silico prediction of protein-protein interactions in human macrophages
- Significance analysis of microarrays applied to the ionizing radiation response
- Principal components analysis corrects for stratification in genome-wide association studies
- Statistical significance for genomewide studies
- The control of the false discovery rate in multiple testing under dependency
- Adaptive Control Processes
- Genomic Control for Association Studies
- A Direct Approach to False Discovery Rates