An Adaptive Genetic Association Test Using Double Kernel Machines

Xiang Zhan, Michael P. Epstein, Debashis Ghosh

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Recently, gene set-based approaches have become very popular in gene expression profiling studies for assessing how genetic variants are related to disease outcomes. Since most genes are not differentially expressed, existing pathway tests considering all genes within a pathway suffer from considerable noise and power loss. Moreover, for a differentially expressed pathway, it is of interest to select important genes that drive the effect of the pathway. In this article, we propose an adaptive association test using double kernel machines (DKM), which can both select important genes within the pathway as well as test for the overall genetic pathway effect. This DKM procedure first uses the garrote kernel machines test for the purposes of subset selection and then the least squares kernel machine test for testing the effect of the subset of genes. An appealing feature of the kernel machine framework is that it can provide a flexible and unified method for multi-dimensional modeling of the genetic pathway effect allowing for both parametric and nonparametric components. This DKM approach is illustrated with application to simulated data as well as to data from a neuroimaging genetics study.

Original languageEnglish (US)
Pages (from-to)262-281
Number of pages20
JournalStatistics in Biosciences
Volume7
Issue number2
DOIs
StatePublished - Oct 1 2015

Fingerprint

Kernel Machines
Genetic Association
Pathway
Genes
Gene
Neuroimaging
Gene Expression Profiling
Subset Selection
Least-Squares Analysis
Gene expression
Noise
Profiling
Gene Expression
Least Squares
Testing
Subset

All Science Journal Classification (ASJC) codes

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

Cite this

Zhan, Xiang ; Epstein, Michael P. ; Ghosh, Debashis. / An Adaptive Genetic Association Test Using Double Kernel Machines. In: Statistics in Biosciences. 2015 ; Vol. 7, No. 2. pp. 262-281.
@article{0ae9b25e10b54821bd54c246374c385f,
title = "An Adaptive Genetic Association Test Using Double Kernel Machines",
abstract = "Recently, gene set-based approaches have become very popular in gene expression profiling studies for assessing how genetic variants are related to disease outcomes. Since most genes are not differentially expressed, existing pathway tests considering all genes within a pathway suffer from considerable noise and power loss. Moreover, for a differentially expressed pathway, it is of interest to select important genes that drive the effect of the pathway. In this article, we propose an adaptive association test using double kernel machines (DKM), which can both select important genes within the pathway as well as test for the overall genetic pathway effect. This DKM procedure first uses the garrote kernel machines test for the purposes of subset selection and then the least squares kernel machine test for testing the effect of the subset of genes. An appealing feature of the kernel machine framework is that it can provide a flexible and unified method for multi-dimensional modeling of the genetic pathway effect allowing for both parametric and nonparametric components. This DKM approach is illustrated with application to simulated data as well as to data from a neuroimaging genetics study.",
author = "Xiang Zhan and Epstein, {Michael P.} and Debashis Ghosh",
year = "2015",
month = "10",
day = "1",
doi = "10.1007/s12561-014-9116-2",
language = "English (US)",
volume = "7",
pages = "262--281",
journal = "Statistics in Biosciences",
issn = "1867-1764",
publisher = "Springer New York",
number = "2",

}

An Adaptive Genetic Association Test Using Double Kernel Machines. / Zhan, Xiang; Epstein, Michael P.; Ghosh, Debashis.

In: Statistics in Biosciences, Vol. 7, No. 2, 01.10.2015, p. 262-281.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An Adaptive Genetic Association Test Using Double Kernel Machines

AU - Zhan, Xiang

AU - Epstein, Michael P.

AU - Ghosh, Debashis

PY - 2015/10/1

Y1 - 2015/10/1

N2 - Recently, gene set-based approaches have become very popular in gene expression profiling studies for assessing how genetic variants are related to disease outcomes. Since most genes are not differentially expressed, existing pathway tests considering all genes within a pathway suffer from considerable noise and power loss. Moreover, for a differentially expressed pathway, it is of interest to select important genes that drive the effect of the pathway. In this article, we propose an adaptive association test using double kernel machines (DKM), which can both select important genes within the pathway as well as test for the overall genetic pathway effect. This DKM procedure first uses the garrote kernel machines test for the purposes of subset selection and then the least squares kernel machine test for testing the effect of the subset of genes. An appealing feature of the kernel machine framework is that it can provide a flexible and unified method for multi-dimensional modeling of the genetic pathway effect allowing for both parametric and nonparametric components. This DKM approach is illustrated with application to simulated data as well as to data from a neuroimaging genetics study.

AB - Recently, gene set-based approaches have become very popular in gene expression profiling studies for assessing how genetic variants are related to disease outcomes. Since most genes are not differentially expressed, existing pathway tests considering all genes within a pathway suffer from considerable noise and power loss. Moreover, for a differentially expressed pathway, it is of interest to select important genes that drive the effect of the pathway. In this article, we propose an adaptive association test using double kernel machines (DKM), which can both select important genes within the pathway as well as test for the overall genetic pathway effect. This DKM procedure first uses the garrote kernel machines test for the purposes of subset selection and then the least squares kernel machine test for testing the effect of the subset of genes. An appealing feature of the kernel machine framework is that it can provide a flexible and unified method for multi-dimensional modeling of the genetic pathway effect allowing for both parametric and nonparametric components. This DKM approach is illustrated with application to simulated data as well as to data from a neuroimaging genetics study.

UR - http://www.scopus.com/inward/record.url?scp=84943356309&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943356309&partnerID=8YFLogxK

U2 - 10.1007/s12561-014-9116-2

DO - 10.1007/s12561-014-9116-2

M3 - Article

C2 - 26640602

AN - SCOPUS:84943356309

VL - 7

SP - 262

EP - 281

JO - Statistics in Biosciences

JF - Statistics in Biosciences

SN - 1867-1764

IS - 2

ER -