Implements the conditionally symmetric multidimensional Gaussian mixture model (csmGmm) for large-scale testing of composite null hypotheses in genetic association applications such as mediation analysis, pleiotropy analysis, and replication analysis. In such analyses, we typically have J sets of K test statistics where K is a small number (e.g. 2 or 3) and J is large (e.g. 1 million). For each one of the J sets, we want to know if we can reject all K individual nulls. Please see the vignette for a quickstart guide. The paper describing these methods is "Testing a Large Number of Composite Null Hypotheses Using Conditionally Symmetric Multidimensional Gaussian Mixtures in Genome-Wide Studies" by Sun R, McCaw Z, & Lin X (Journal of the American Statistical Association 2025,
This R package implements the conditionally symmetric multidimensional Gaussian mixture model (csmGmm) for large-scale testing of composite null hypotheses in genetic association applications such as mediation analysis, pleiotropy analysis, and replication analysis. In such analyses, we typically have K sets of test statistics corresponding to the same J SNPs, where K is a small number (e.g. 2 or 3) and J is large (e.g. 1 million). For each SNP, we want to know if we can reject all K individual nulls. The preprint “Testing a Large Number of Composite Null Hypotheses Using Conditionally Symmetric Multidimensional Gaussian Mixtures in Genome-Wide Studies” by R Sun, Z McCaw, and X Lin is available upon request from the package maintainer. Please see the vignette for a quickstart guide.