Blindly Using Wald's Test Can Miss Rare Disease-Causal Variants in Case-Control Association Studies

Guan Xing, Chang Yun Lin, Stephen P. Wooding, Chao Xing

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

There are four tests - the likelihood ratio (LR) test, Wald's test, the score test and the exact test - commonly employed in genetic association studies. On comparison of the four tests, we found that Wald's test, popular in genome-wide screens due to its low computational demands, exhibited a paradoxical behaviour in that the test statistic decreased as the effect size of the variant increased, resulting in a loss of power. The LR test always achieved the most significant P-values, followed by the exact test. We further examined the results in a real data set composed of high- and low-cholesterol subjects from the Dallas Heart Study (DHS). We also compared the single-variant LR test with two multi-variant analysis approaches - the burden test and the C-alpha test - in analysing the sequencing data by simulation. Our results call for caution in using Wald's test in genome-wide case-control association studies and suggest that the LR test is a better alternative in spite of its computational demands.

Original languageEnglish (US)
Pages (from-to)168-177
Number of pages10
JournalAnnals of Human Genetics
Volume76
Issue number2
DOIs
StatePublished - Mar 2012

Fingerprint

Rare Diseases
Case-Control Studies
Cholesterol
Genome
Genetic Association Studies
Datasets

Keywords

  • Case-control study
  • Low-frequency variants
  • Wald's test

ASJC Scopus subject areas

  • Genetics(clinical)
  • Genetics

Cite this

Blindly Using Wald's Test Can Miss Rare Disease-Causal Variants in Case-Control Association Studies. / Xing, Guan; Lin, Chang Yun; Wooding, Stephen P.; Xing, Chao.

In: Annals of Human Genetics, Vol. 76, No. 2, 03.2012, p. 168-177.

Research output: Contribution to journalArticle

Xing, Guan ; Lin, Chang Yun ; Wooding, Stephen P. ; Xing, Chao. / Blindly Using Wald's Test Can Miss Rare Disease-Causal Variants in Case-Control Association Studies. In: Annals of Human Genetics. 2012 ; Vol. 76, No. 2. pp. 168-177.
@article{575c106b3d2e4ae8a589e2ebc9627524,
title = "Blindly Using Wald's Test Can Miss Rare Disease-Causal Variants in Case-Control Association Studies",
abstract = "There are four tests - the likelihood ratio (LR) test, Wald's test, the score test and the exact test - commonly employed in genetic association studies. On comparison of the four tests, we found that Wald's test, popular in genome-wide screens due to its low computational demands, exhibited a paradoxical behaviour in that the test statistic decreased as the effect size of the variant increased, resulting in a loss of power. The LR test always achieved the most significant P-values, followed by the exact test. We further examined the results in a real data set composed of high- and low-cholesterol subjects from the Dallas Heart Study (DHS). We also compared the single-variant LR test with two multi-variant analysis approaches - the burden test and the C-alpha test - in analysing the sequencing data by simulation. Our results call for caution in using Wald's test in genome-wide case-control association studies and suggest that the LR test is a better alternative in spite of its computational demands.",
keywords = "Case-control study, Low-frequency variants, Wald's test",
author = "Guan Xing and Lin, {Chang Yun} and Wooding, {Stephen P.} and Chao Xing",
year = "2012",
month = "3",
doi = "10.1111/j.1469-1809.2011.00700.x",
language = "English (US)",
volume = "76",
pages = "168--177",
journal = "Annals of Human Genetics",
issn = "0003-4800",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Blindly Using Wald's Test Can Miss Rare Disease-Causal Variants in Case-Control Association Studies

AU - Xing, Guan

AU - Lin, Chang Yun

AU - Wooding, Stephen P.

AU - Xing, Chao

PY - 2012/3

Y1 - 2012/3

N2 - There are four tests - the likelihood ratio (LR) test, Wald's test, the score test and the exact test - commonly employed in genetic association studies. On comparison of the four tests, we found that Wald's test, popular in genome-wide screens due to its low computational demands, exhibited a paradoxical behaviour in that the test statistic decreased as the effect size of the variant increased, resulting in a loss of power. The LR test always achieved the most significant P-values, followed by the exact test. We further examined the results in a real data set composed of high- and low-cholesterol subjects from the Dallas Heart Study (DHS). We also compared the single-variant LR test with two multi-variant analysis approaches - the burden test and the C-alpha test - in analysing the sequencing data by simulation. Our results call for caution in using Wald's test in genome-wide case-control association studies and suggest that the LR test is a better alternative in spite of its computational demands.

AB - There are four tests - the likelihood ratio (LR) test, Wald's test, the score test and the exact test - commonly employed in genetic association studies. On comparison of the four tests, we found that Wald's test, popular in genome-wide screens due to its low computational demands, exhibited a paradoxical behaviour in that the test statistic decreased as the effect size of the variant increased, resulting in a loss of power. The LR test always achieved the most significant P-values, followed by the exact test. We further examined the results in a real data set composed of high- and low-cholesterol subjects from the Dallas Heart Study (DHS). We also compared the single-variant LR test with two multi-variant analysis approaches - the burden test and the C-alpha test - in analysing the sequencing data by simulation. Our results call for caution in using Wald's test in genome-wide case-control association studies and suggest that the LR test is a better alternative in spite of its computational demands.

KW - Case-control study

KW - Low-frequency variants

KW - Wald's test

UR - http://www.scopus.com/inward/record.url?scp=84862823427&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862823427&partnerID=8YFLogxK

U2 - 10.1111/j.1469-1809.2011.00700.x

DO - 10.1111/j.1469-1809.2011.00700.x

M3 - Article

C2 - 22256951

AN - SCOPUS:84862823427

VL - 76

SP - 168

EP - 177

JO - Annals of Human Genetics

JF - Annals of Human Genetics

SN - 0003-4800

IS - 2

ER -