Small Sample Kernel Association Tests for Human Genetic and Microbiome Association Studies

Jun Chen, Wenan Chen, Ni Zhao, Michael C. Wu, Daniel J. Schaid

Research output: Contribution to journalArticle

Abstract

Kernel machine based association tests (KAT) have been increasingly used in testing the association between an outcome and a set of biological measurements due to its power to combine multiple weak signals of complex relationship with the outcome through the specification of a relevant kernel. Human genetic and microbiome association studies are two important applications of KAT. However, the classic KAT framework relies on large sample theory, and conservativeness has been observed for small sample studies, especially for microbiome association studies. The common approach for addressing the small sample problem relies on computationally intensive resampling methods. Here, we derive an exact test for KAT with continuous traits, which resolve the small sample conservatism of KAT without the need for resampling. The exact test has significantly improved power to detect association for microbiome studies. For binary traits, we propose a similar approximate test, and we show that the approximate test is very powerful for a wide range of kernels including common variant- and microbiome-based kernels, and the approximate test controls the type I error well for these kernels. In contrast, the sequence kernel association tests have slightly inflated genomic inflation factors after small sample adjustment. Extensive simulations and application to a real microbiome association study are used to demonstrate the utility of our method.

Original languageEnglish (US)
Pages (from-to)5-19
Number of pages15
JournalGenetic Epidemiology
Volume40
Issue number1
DOIs
StatePublished - Jan 1 2016
Externally publishedYes

Fingerprint

Microbiota
Medical Genetics
Genetic Association Studies
Economic Inflation
Politics

Keywords

  • Exact tests
  • Kernel machine based association tests
  • Overdispersion
  • Small sample problem

ASJC Scopus subject areas

  • Epidemiology
  • Genetics(clinical)

Cite this

Small Sample Kernel Association Tests for Human Genetic and Microbiome Association Studies. / Chen, Jun; Chen, Wenan; Zhao, Ni; Wu, Michael C.; Schaid, Daniel J.

In: Genetic Epidemiology, Vol. 40, No. 1, 01.01.2016, p. 5-19.

Research output: Contribution to journalArticle

Chen, Jun ; Chen, Wenan ; Zhao, Ni ; Wu, Michael C. ; Schaid, Daniel J. / Small Sample Kernel Association Tests for Human Genetic and Microbiome Association Studies. In: Genetic Epidemiology. 2016 ; Vol. 40, No. 1. pp. 5-19.
@article{63894a0b8f0b46669567680f2773b690,
title = "Small Sample Kernel Association Tests for Human Genetic and Microbiome Association Studies",
abstract = "Kernel machine based association tests (KAT) have been increasingly used in testing the association between an outcome and a set of biological measurements due to its power to combine multiple weak signals of complex relationship with the outcome through the specification of a relevant kernel. Human genetic and microbiome association studies are two important applications of KAT. However, the classic KAT framework relies on large sample theory, and conservativeness has been observed for small sample studies, especially for microbiome association studies. The common approach for addressing the small sample problem relies on computationally intensive resampling methods. Here, we derive an exact test for KAT with continuous traits, which resolve the small sample conservatism of KAT without the need for resampling. The exact test has significantly improved power to detect association for microbiome studies. For binary traits, we propose a similar approximate test, and we show that the approximate test is very powerful for a wide range of kernels including common variant- and microbiome-based kernels, and the approximate test controls the type I error well for these kernels. In contrast, the sequence kernel association tests have slightly inflated genomic inflation factors after small sample adjustment. Extensive simulations and application to a real microbiome association study are used to demonstrate the utility of our method.",
keywords = "Exact tests, Kernel machine based association tests, Overdispersion, Small sample problem",
author = "Jun Chen and Wenan Chen and Ni Zhao and Wu, {Michael C.} and Schaid, {Daniel J.}",
year = "2016",
month = "1",
day = "1",
doi = "10.1002/gepi.21934",
language = "English (US)",
volume = "40",
pages = "5--19",
journal = "Genetic Epidemiology",
issn = "0741-0395",
publisher = "Wiley-Liss Inc.",
number = "1",

}

TY - JOUR

T1 - Small Sample Kernel Association Tests for Human Genetic and Microbiome Association Studies

AU - Chen, Jun

AU - Chen, Wenan

AU - Zhao, Ni

AU - Wu, Michael C.

AU - Schaid, Daniel J.

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Kernel machine based association tests (KAT) have been increasingly used in testing the association between an outcome and a set of biological measurements due to its power to combine multiple weak signals of complex relationship with the outcome through the specification of a relevant kernel. Human genetic and microbiome association studies are two important applications of KAT. However, the classic KAT framework relies on large sample theory, and conservativeness has been observed for small sample studies, especially for microbiome association studies. The common approach for addressing the small sample problem relies on computationally intensive resampling methods. Here, we derive an exact test for KAT with continuous traits, which resolve the small sample conservatism of KAT without the need for resampling. The exact test has significantly improved power to detect association for microbiome studies. For binary traits, we propose a similar approximate test, and we show that the approximate test is very powerful for a wide range of kernels including common variant- and microbiome-based kernels, and the approximate test controls the type I error well for these kernels. In contrast, the sequence kernel association tests have slightly inflated genomic inflation factors after small sample adjustment. Extensive simulations and application to a real microbiome association study are used to demonstrate the utility of our method.

AB - Kernel machine based association tests (KAT) have been increasingly used in testing the association between an outcome and a set of biological measurements due to its power to combine multiple weak signals of complex relationship with the outcome through the specification of a relevant kernel. Human genetic and microbiome association studies are two important applications of KAT. However, the classic KAT framework relies on large sample theory, and conservativeness has been observed for small sample studies, especially for microbiome association studies. The common approach for addressing the small sample problem relies on computationally intensive resampling methods. Here, we derive an exact test for KAT with continuous traits, which resolve the small sample conservatism of KAT without the need for resampling. The exact test has significantly improved power to detect association for microbiome studies. For binary traits, we propose a similar approximate test, and we show that the approximate test is very powerful for a wide range of kernels including common variant- and microbiome-based kernels, and the approximate test controls the type I error well for these kernels. In contrast, the sequence kernel association tests have slightly inflated genomic inflation factors after small sample adjustment. Extensive simulations and application to a real microbiome association study are used to demonstrate the utility of our method.

KW - Exact tests

KW - Kernel machine based association tests

KW - Overdispersion

KW - Small sample problem

UR - http://www.scopus.com/inward/record.url?scp=84951771295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951771295&partnerID=8YFLogxK

U2 - 10.1002/gepi.21934

DO - 10.1002/gepi.21934

M3 - Article

C2 - 26643881

AN - SCOPUS:84951771295

VL - 40

SP - 5

EP - 19

JO - Genetic Epidemiology

JF - Genetic Epidemiology

SN - 0741-0395

IS - 1

ER -