Transmission Based Association Test for Multivariate Phenotype Using Quasi Likelihood

Article Type

Research Article

Publication Title

Sankhya B

Abstract

The transmission disequilibrium test (TDT) is a family-based alternative to the population-based case-control test for genetic association and is protected against inflated rates of false positives that may emerge due to potential population substructures. The majority of complex genetic traits are governed by quantitative precursors and it may be a more prudent strategy to analyze these quantitative traits instead of the binary end-point clinical trait. However, a single quantitative trait may not be a sufficiently good surrogate for the clinical end-point and tests based on a multivariate phenotype vector comprising the precursor variables may yield higher power to detect association. We propose a model-free family based test procedure for testing association in the presence of linkage for multivariate phenotypes. The parameters are estimated using resistance generalized estimating equations (RGEE). The test procedure is easily extended for the multivariate phenotypes consisting the mixture of continuous and categorical phenotypes. We carry out extensive simulations under a wide spectrum of genetic models to compare the performance of the proposed test procedure with family-based tests for multiple phenotypes FBAT-GEE and univariate test procedures. Our findings indicate that the statistical power of the proposed test is comparable to that of FBAT-GEE. The proposed test procedure is superior than the univariate test procedures for the phenotype that contain less genetic information. While, it is marginally less powerful than univariate method for the phenotypes that exhibit a strong association with genetic factors.

First Page

804

Last Page

832

DOI

10.1007/s13571-025-00365-z

Publication Date

11-1-2025

Link to Full Text

Share

COinS