A note on p-value multiple-testing adjustment for two-step genome-wide gene-environment interactions scans
Lewinger, J. P.; Kawaguchi, E. S.; Gauderman, W. J.
Show abstract
Two-step testing has emerged as the leading approach for genome-wide gene-environment interaction scans, providing greater power than standard single-step methods in nearly all plausible scenarios. This gain in power has enabled several recent discoveries in gene-environment interactions. However, despite controlling the genome-wide type I error rate, two-step tests typically lack valid p-values, hindering comparisons with single-step results. We demonstrate how multiple-testing adjusted p-values can be defined for two-step tests using standard multiple-testing theory, and how these can be further scaled to allow valid comparisons with single-step tests.
Matching journals
The top 1 journal accounts for 50% of the predicted probability mass.
Similar papers in this journal
- Improved two-step testing of genome-wide gene-environment interactions 98%
- A robust association test leveraging unknown genetic interactions: Application to cystic brosis lung disease 96%
- A Two-Sample Robust Bayesian Mendelian Randomization Method Accounting for Linkage Disequilibrium and Idiosyncratic Pleiotropy with Applications to the COVID-19 Outcome 95%
Similar papers in this journal
Similar papers in this journal
- A mixed-model approach for powerful testing of genetic associations with cancer risk incorporating tumor characteristics 95%
- Overcoming the impacts of two-step batch effect correction on gene expression estimation and inference 94%
- Estimating the overall fraction of phenotypic variance attributed to high-dimensional predictors measured with error 94%
Similar papers in this journal
- Two-Stage Multivariate Mendelian Randomization on Multiple Outcomes with Mixed Distributions 96%
- Causal Mediation Analysis with Multiple Causally Ordered and Non-ordered Mediators based on Summarized Genetic Data 94%
- A bivariate zero-inflated negative binomial model and its applications to biomedical settings 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.