A Powerful Test for SNP Effects on Multivariate Binary Outcomes Using Kernel Machine Regression

Clemontina A. Davenport, Arnab Maity, Patrick F. Sullivan, Jung Ying Tzeng

Research output: Research - peer-reviewArticle

Abstract

Evaluating multiple binary outcomes is common in genetic studies of complex diseases. These outcomes are often correlated because they are collected from the same individual and they may share common marker effects. In this paper, we propose a procedure to test for effect of a single nucleotide polymorphism-set on multiple, possibly correlated, binary responses. We develop a score-based test using a non-parametric modeling framework that jointly models the global effect of the marker set. We account for the non-linear effects and potentially complicated interaction between markers using reproducing kernels. Our testing procedure only requires estimation under the null hypothesis and we use multivariate generalized estimating equations to estimate the model components to account for the correlation among the outcomes. We evaluate finite sample performance of our test via simulation study and demonstrate our methods using the Clinical Antipsychotic Trials of Intervention Effectiveness antibody study data and the CoLaus study data.

LanguageEnglish (US)
Pages1-22
Number of pages22
JournalStatistics in Biosciences
DOIs
StateAccepted/In press - Mar 24 2017

Fingerprint

Kernel Machines
Binary Outcomes
Regression
Single Nucleotide Polymorphism
Polymorphism
Antipsychotic Agents
Nucleotides
Antibodies
Testing
Multiple Outcomes
Binary Response
Generalized Estimating Equations
Single nucleotide Polymorphism
Reproducing Kernel
Nonlinear Effects
Component Model
Antibody
Null hypothesis
Clinical Trials
Simulation Study

Keywords

  • Correlated binary responses
  • Generalized estimating equations
  • IBS kernel
  • Kernel machine
  • Non-parametric regression

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology (miscellaneous)

Cite this

A Powerful Test for SNP Effects on Multivariate Binary Outcomes Using Kernel Machine Regression. / Davenport, Clemontina A.; Maity, Arnab; Sullivan, Patrick F.; Tzeng, Jung Ying.

In: Statistics in Biosciences, 24.03.2017, p. 1-22.

Research output: Research - peer-reviewArticle

@article{b0fc98973424485ba7a3e7ddc3f5c2e0,
title = "A Powerful Test for SNP Effects on Multivariate Binary Outcomes Using Kernel Machine Regression",
abstract = "Evaluating multiple binary outcomes is common in genetic studies of complex diseases. These outcomes are often correlated because they are collected from the same individual and they may share common marker effects. In this paper, we propose a procedure to test for effect of a single nucleotide polymorphism-set on multiple, possibly correlated, binary responses. We develop a score-based test using a non-parametric modeling framework that jointly models the global effect of the marker set. We account for the non-linear effects and potentially complicated interaction between markers using reproducing kernels. Our testing procedure only requires estimation under the null hypothesis and we use multivariate generalized estimating equations to estimate the model components to account for the correlation among the outcomes. We evaluate finite sample performance of our test via simulation study and demonstrate our methods using the Clinical Antipsychotic Trials of Intervention Effectiveness antibody study data and the CoLaus study data.",
keywords = "Correlated binary responses, Generalized estimating equations, IBS kernel, Kernel machine, Non-parametric regression",
author = "Davenport, {Clemontina A.} and Arnab Maity and Sullivan, {Patrick F.} and Tzeng, {Jung Ying}",
year = "2017",
month = "3",
doi = "10.1007/s12561-017-9189-9",
pages = "1--22",
journal = "Statistics in Biosciences",
issn = "1867-1764",
publisher = "Springer New York",

}

TY - JOUR

T1 - A Powerful Test for SNP Effects on Multivariate Binary Outcomes Using Kernel Machine Regression

AU - Davenport,Clemontina A.

AU - Maity,Arnab

AU - Sullivan,Patrick F.

AU - Tzeng,Jung Ying

PY - 2017/3/24

Y1 - 2017/3/24

N2 - Evaluating multiple binary outcomes is common in genetic studies of complex diseases. These outcomes are often correlated because they are collected from the same individual and they may share common marker effects. In this paper, we propose a procedure to test for effect of a single nucleotide polymorphism-set on multiple, possibly correlated, binary responses. We develop a score-based test using a non-parametric modeling framework that jointly models the global effect of the marker set. We account for the non-linear effects and potentially complicated interaction between markers using reproducing kernels. Our testing procedure only requires estimation under the null hypothesis and we use multivariate generalized estimating equations to estimate the model components to account for the correlation among the outcomes. We evaluate finite sample performance of our test via simulation study and demonstrate our methods using the Clinical Antipsychotic Trials of Intervention Effectiveness antibody study data and the CoLaus study data.

AB - Evaluating multiple binary outcomes is common in genetic studies of complex diseases. These outcomes are often correlated because they are collected from the same individual and they may share common marker effects. In this paper, we propose a procedure to test for effect of a single nucleotide polymorphism-set on multiple, possibly correlated, binary responses. We develop a score-based test using a non-parametric modeling framework that jointly models the global effect of the marker set. We account for the non-linear effects and potentially complicated interaction between markers using reproducing kernels. Our testing procedure only requires estimation under the null hypothesis and we use multivariate generalized estimating equations to estimate the model components to account for the correlation among the outcomes. We evaluate finite sample performance of our test via simulation study and demonstrate our methods using the Clinical Antipsychotic Trials of Intervention Effectiveness antibody study data and the CoLaus study data.

KW - Correlated binary responses

KW - Generalized estimating equations

KW - IBS kernel

KW - Kernel machine

KW - Non-parametric regression

UR - http://www.scopus.com/inward/record.url?scp=85015985735&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015985735&partnerID=8YFLogxK

U2 - 10.1007/s12561-017-9189-9

DO - 10.1007/s12561-017-9189-9

M3 - Article

SP - 1

EP - 22

JO - Statistics in Biosciences

T2 - Statistics in Biosciences

JF - Statistics in Biosciences

SN - 1867-1764

ER -