The PPLD has advantages over conventional regression methods in application to moderately sized genome-wide association studies.

Veronica J Vieland,Sang-Cheol Seok,Ayse Ulgen

doi:10.1371/journal.pone.0257164

Veronica J Vieland, Sang-Cheol Seok + Show 1 more

Open Access

https://doi.org/10.1371/journal.pone.0257164

Copy DOI

Abstract

In earlier work, we have developed and evaluated an alternative approach to the analysis of GWAS data, based on a statistic called the PPLD. More recently, motivated by a GWAS for genetic modifiers of the X-linked Mendelian disorder Duchenne Muscular Dystrophy (DMD), we adapted the PPLD for application to time-to-event (TE) phenotypes. Because DMD itself is relatively rare, this is a setting in which the very large sample sizes generally assembled for GWAS are simply not attainable. For this reason, statistical methods specially adapted for use in small data sets are required. Here we explore the behavior of the TE-PPLD via simulations, comparing the TE-PPLD with Cox Proportional Hazards analysis in the context of small to moderate sample sizes. Our results will help to inform our approach to the DMD study going forward, and they illustrate several respects in which the TE-PPLD, and by extension the original PPLD, offer advantages over regression-based approaches to GWAS in this context.

Highlights

In previous work we have developed and evaluated a statistic called the Posterior Probability of Linkage Disequilibrium (PPLD) as a measure of evidence for or against trait-SNP association [1,2,3], and we have extended the PPLD to accommodate time-to-event (TE) data, yielding the TE-PPLD [4]
In what follows we evaluate the behavior of the TE-PPLD and Cox Proportional Hazards (CPH) in application to GWAS analysis in the context of our intended genetic application, using simulated data
In this paper we have evaluated the sampling behavior of the TE-PPLD in small to moderate samples sizes, and compared this behavior with the sampling behavior of CPH p-values, using simulations

Summary

Introduction

In previous work we have developed and evaluated a statistic called the Posterior Probability of Linkage Disequilibrium (PPLD) as a measure of evidence for or against trait-SNP association [1,2,3], and we have extended the PPLD to accommodate time-to-event (TE) data, yielding the TE-PPLD [4]. In this paper we compare and contrast the TE-PPLD with the more familiar regression-based approach to handling time-to-event phenotypes using the Cox Proportional Hazards (CPH) model, via simulations and with a focus on small to moderate sample sizes. Rather than real data, allows us to compare and contrast the behavior of the two statistics under conditions in which the true underlying genetic model is known, so that it is possible to definitively distinguish correct from incorrect results.

Methods

Results

Conclusion