. Author manuscript; available in PMC: 2021 Jan 1.

Published in final edited form as: Proc IEEE Inst Electr Electron Eng. 2019 Oct 29;108(1):125–162. doi: 10.1109/JPROC.2019.2947272

TABLE IV.

Example studies using multi-trait methods, where single-SNP-multi-QT or multi-SNP-multi-QT associations are examined.

Ref	Notes
[80]	Sum of powered score tests (SPU(γ)), adaptive SPU test for multi-trait-single-SNP associations (aSPU), selection of most powerful weighted test via adjusting weights to the studied data
[87]	Sum of powered score tests (SPU(γ₁, γ₂)), adaptive SPU test for multi-trait-multi-SNP associations (aSPUset), selection of most powerful weighted test via adjusting weights to the studied data
[88]	Adaptive SPU test for multi-trait-single-SNP associations (aSPU) under a proportional odds model (POM) instead of the generalized estimation equations (GEE) framework used in [80].
[44]	Brain-wide ROI QTs as a multivariate response, distance covariance between QT set and each SNP, local FDR modeling
[89]	Functional GWAS (FGWAS), multivariate varying coefficient model (MVCM), global sure independence screening (GSIS), GWAS of functional QTs including curves, surfaces and volumes