Genome-Wide Significance Levels and Weighted Hypothesis Testing
arXiv:1010.4637 · doi:10.1214/09-STS289
Abstract
Genetic investigations often involve the testing of vast numbers of related hypotheses simultaneously. To control the overall error rate, a substantial penalty is required, making it difficult to detect signals of moderate strength. To improve the power in this setting, a number of authors have considered using weighted -values, with the motivation often based upon the scientific plausibility of the hypotheses. We review this literature, derive optimal weights and show that the power is remarkably robust to misspecification of these weights. We consider two methods for choosing weights in practice. The first, external weighting, is based on prior information. The second, estimated weighting, uses the data to choose weights.
Published in at http://dx.doi.org/10.1214/09-STS289 the Statistical Science (http://www.imstat.org/sts/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (4)
Cited by in corpus (12)
- Higher Criticism for Large-Scale Inference, Especially for Rare and Weak Effects
- Covariate powered cross-weighted multiple testing
- Optimal weighting for false discovery rate control
- Causal Inference in Genetic Trio Studies
- A Tutorial on Statistically Sound Pattern Discovery
- Asymptotic optimality of the Westfall--Young permutation procedure for multiple testing under dependence
- Power-enhanced multiple decision functions controlling family-wise error and false discovery rates
- Single-index modulated multiple testing
- Optimal Multiple Testing Under a Gaussian Prior on the Effect Sizes
- Robust Tests in Genome-Wide Scans under Incomplete Linkage Disequilibrium
- Frasian Inference
- Classes of Multiple Decision Functions Strongly Controlling FWER and FDR