Novel Data Mining Approaches for Detecting Quantitative Trait Loci of Bone Mineral Density in Genome-Wide Linkage Analysis

Qiao Li; Alexander J. Macgregor; Wenjia Wang

doi:10.1007/978-3-642-23878-9_59

Novel Data Mining Approaches for Detecting Quantitative Trait Loci of Bone Mineral Density in Genome-Wide Linkage Analysis

Qiao Li, Alexander J. Macgregor, Wenjia Wang

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Abstract

Haseman-Elston (H-E) regression is a commonly used conventional approach for detecting quantitative trait loci (QTLs), which regulate the quantitative phenotype based on the Identical-By-Descent (IBD) information between twins in Genome-wide scan. However, this approach only considers genetic effect at individual loci, but not any interaction between genes. A Pair-Wise H-E regression (PWH-E) and a Feature Screening Approach (FSA) are proposed in this paper to take gene-gene interaction into account when detecting QTLs. After testing these approaches with several series of simulation studies, they are applied to a real-world bone mineral density (BMD) dataset, and find three site specific sets of potential QTLs. Further comparison analyses show that our results not only corroborate the 14 findings from previous published studies, but also suggest 22 new QTLs of BMD.

Original language	English
Title of host publication	Intelligent Data Engineering and Automated Learning - IDEAL 2011
Editors	Hujun Yin, Wenjia Wang, Victor Rayward-Smith
Publisher	Springer
Pages	498-510
Number of pages	13
DOIs	https://doi.org/10.1007/978-3-642-23878-9_59
Publication status	Published - 2011

Access to Document

10.1007/978-3-642-23878-9_59

Cite this

@inbook{c17e80b9dcea4b4f9172f3b0a3d98e1b,

title = "Novel Data Mining Approaches for Detecting Quantitative Trait Loci of Bone Mineral Density in Genome-Wide Linkage Analysis",

abstract = "Haseman-Elston (H-E) regression is a commonly used conventional approach for detecting quantitative trait loci (QTLs), which regulate the quantitative phenotype based on the Identical-By-Descent (IBD) information between twins in Genome-wide scan. However, this approach only considers genetic effect at individual loci, but not any interaction between genes. A Pair-Wise H-E regression (PWH-E) and a Feature Screening Approach (FSA) are proposed in this paper to take gene-gene interaction into account when detecting QTLs. After testing these approaches with several series of simulation studies, they are applied to a real-world bone mineral density (BMD) dataset, and find three site specific sets of potential QTLs. Further comparison analyses show that our results not only corroborate the 14 findings from previous published studies, but also suggest 22 new QTLs of BMD.",

author = "Qiao Li and Macgregor, {Alexander J.} and Wenjia Wang",

year = "2011",

doi = "10.1007/978-3-642-23878-9_59",

language = "English",

pages = "498--510",

editor = "Hujun Yin and Wenjia Wang and Victor Rayward-Smith",

booktitle = "Intelligent Data Engineering and Automated Learning - IDEAL 2011",

publisher = "Springer",

address = "Germany",

}

Novel Data Mining Approaches for Detecting Quantitative Trait Loci of Bone Mineral Density in Genome-Wide Linkage Analysis. / Li, Qiao; Macgregor, Alexander J.; Wang, Wenjia.

Intelligent Data Engineering and Automated Learning - IDEAL 2011. ed. / Hujun Yin; Wenjia Wang; Victor Rayward-Smith. Springer, 2011. p. 498-510.

Research output: Chapter in Book/Report/Conference proceeding › Chapter

TY - CHAP

T1 - Novel Data Mining Approaches for Detecting Quantitative Trait Loci of Bone Mineral Density in Genome-Wide Linkage Analysis

AU - Li, Qiao

AU - Macgregor, Alexander J.

AU - Wang, Wenjia

PY - 2011

Y1 - 2011

N2 - Haseman-Elston (H-E) regression is a commonly used conventional approach for detecting quantitative trait loci (QTLs), which regulate the quantitative phenotype based on the Identical-By-Descent (IBD) information between twins in Genome-wide scan. However, this approach only considers genetic effect at individual loci, but not any interaction between genes. A Pair-Wise H-E regression (PWH-E) and a Feature Screening Approach (FSA) are proposed in this paper to take gene-gene interaction into account when detecting QTLs. After testing these approaches with several series of simulation studies, they are applied to a real-world bone mineral density (BMD) dataset, and find three site specific sets of potential QTLs. Further comparison analyses show that our results not only corroborate the 14 findings from previous published studies, but also suggest 22 new QTLs of BMD.

AB - Haseman-Elston (H-E) regression is a commonly used conventional approach for detecting quantitative trait loci (QTLs), which regulate the quantitative phenotype based on the Identical-By-Descent (IBD) information between twins in Genome-wide scan. However, this approach only considers genetic effect at individual loci, but not any interaction between genes. A Pair-Wise H-E regression (PWH-E) and a Feature Screening Approach (FSA) are proposed in this paper to take gene-gene interaction into account when detecting QTLs. After testing these approaches with several series of simulation studies, they are applied to a real-world bone mineral density (BMD) dataset, and find three site specific sets of potential QTLs. Further comparison analyses show that our results not only corroborate the 14 findings from previous published studies, but also suggest 22 new QTLs of BMD.

U2 - 10.1007/978-3-642-23878-9_59

DO - 10.1007/978-3-642-23878-9_59

M3 - Chapter

SP - 498

EP - 510

BT - Intelligent Data Engineering and Automated Learning - IDEAL 2011

A2 - Yin, Hujun

A2 - Wang, Wenjia

A2 - Rayward-Smith, Victor

PB - Springer

ER -