東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Bayes Factor Approaches for Region-b...

Xu, Jingxiong .

FindBook

Google Book

Amazon

博客來

Bayes Factor Approaches for Region-based Analysis of Rare Variants From Next Generation Sequencing Studies.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Bayes Factor Approaches for Region-based Analysis of Rare Variants From Next Generation Sequencing Studies./
作者:	Xu, Jingxiong .
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
面頁冊數:	132 p.
附註:	Source: Dissertations Abstracts International, Volume: 81-06, Section: B.
Contained By:	Dissertations Abstracts International81-06B.
標題:	Biostatistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27539473
ISBN:	9781392565865

Bayes Factor Approaches for Region-based Analysis of Rare Variants From Next Generation Sequencing Studies.
Xu, Jingxiong .

Bayes Factor Approaches for Region-based Analysis of Rare Variants From Next Generation Sequencing Studies. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 132 p.

Source: Dissertations Abstracts International, Volume: 81-06, Section: B.

Thesis (Ph.D.)--University of Toronto (Canada), 2019.

This item must not be sold to any third party vendors.

The emergence of new high-throughput genotyping technologies, such as Next Generation Sequencing (NGS), allows the study of the human genome at an unprecedented depth and scale. The discovery of germline rare variants (RVs) through NGS is a very challenging issue in the field of human genetics. Since RVs have extremely low frequencies, traditional strategies that analyze one variant at a time are underpowered for detecting associations with RVs. Gene-level or region-based statistics can provide a first step in the analysis of RVs that can lead to further experimental validation. Bayesian analysis is not well developed for RV analysis. Our goal in this thesis is to develop such approaches and show their interests for germline RV analyses in the context of case-control studies. Chapter 1 gives a general overview about NGS data analysis and methods for association tests with RV data. In Chapter 2, we propose a novel region-based statistical approach based on the Bayes Factor (BF) to assess evidence of association between a set of RVs located on the same genomic region and a disease outcome in the context of case-control design. We derive the theoretical null distribution of the BF under our prior setting. Informative priors are introduced using prior evidence of association from a Kolmogorov-Smirnov test statistic. In Chapter 3, we introduce a Bayesian procedure to control the False Discovery Rate (BFDR) in the context of genome-wide inference. We develop a simulation program, sim1000G, to generate RV data similar to the 1,000 genomes sequencing project and assess our BFDR procedure. Our simulation studies show that the new BF statistic outperforms standard methods (SKAT, SKAT-O, Burden test) in case-control studies with moderate sample sizes and is equivalent to them under large sample size scenarios. Chapter 4 concludes this thesis with an extension of the BF approach that integrates individual-level and variant-level covariates by using a Bayesian regression approach and inference based on the Integrated Nested Laplace Approximation (INLA). Finally, the interests of our methodological developments are illustrated throughout the thesis by real data applications to a lung cancer case-control study seeking RV association with known and novel cancer genes.

ISBN: 9781392565865Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Bayesian Analysis

Bayes Factor Approaches for Region-based Analysis of Rare Variants From Next Generation Sequencing Studies.
LDR:03567nmm a2200385 4500 001 2272841
005 20201105110238.5
008 220629s2019 ||||||||||||||||| ||eng d
020 $a 9781392565865
035 $a (MiAaPQ)AAI27539473
035 $a AAI27539473
040 $a MiAaPQ $c MiAaPQ
100 1 $a Xu, Jingxiong . $3 3550265
245 1 0 $a Bayes Factor Approaches for Region-based Analysis of Rare Variants From Next Generation Sequencing Studies.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 132 p.
500 $a Source: Dissertations Abstracts International, Volume: 81-06, Section: B.
500 $a Advisor: Briollais, Laurent;Xu, Wei.
502 $a Thesis (Ph.D.)--University of Toronto (Canada), 2019.
506 $a This item must not be sold to any third party vendors.
520 $a The emergence of new high-throughput genotyping technologies, such as Next Generation Sequencing (NGS), allows the study of the human genome at an unprecedented depth and scale. The discovery of germline rare variants (RVs) through NGS is a very challenging issue in the field of human genetics. Since RVs have extremely low frequencies, traditional strategies that analyze one variant at a time are underpowered for detecting associations with RVs. Gene-level or region-based statistics can provide a first step in the analysis of RVs that can lead to further experimental validation. Bayesian analysis is not well developed for RV analysis. Our goal in this thesis is to develop such approaches and show their interests for germline RV analyses in the context of case-control studies. Chapter 1 gives a general overview about NGS data analysis and methods for association tests with RV data. In Chapter 2, we propose a novel region-based statistical approach based on the Bayes Factor (BF) to assess evidence of association between a set of RVs located on the same genomic region and a disease outcome in the context of case-control design. We derive the theoretical null distribution of the BF under our prior setting. Informative priors are introduced using prior evidence of association from a Kolmogorov-Smirnov test statistic. In Chapter 3, we introduce a Bayesian procedure to control the False Discovery Rate (BFDR) in the context of genome-wide inference. We develop a simulation program, sim1000G, to generate RV data similar to the 1,000 genomes sequencing project and assess our BFDR procedure. Our simulation studies show that the new BF statistic outperforms standard methods (SKAT, SKAT-O, Burden test) in case-control studies with moderate sample sizes and is equivalent to them under large sample size scenarios. Chapter 4 concludes this thesis with an extension of the BF approach that integrates individual-level and variant-level covariates by using a Bayesian regression approach and inference based on the Integrated Nested Laplace Approximation (INLA). Finally, the interests of our methodological developments are illustrated throughout the thesis by real data applications to a lung cancer case-control study seeking RV association with known and novel cancer genes.
590 $a School code: 0779.
650 4 $a Biostatistics. $3 1002712
650 4 $a Genetics. $3 530508
650 4 $a Health sciences. $3 3168359
653 $a Bayesian Analysis
653 $a Biostatistical method
653 $a False discovery rate
653 $a Genome wide association study
653 $a Rare genetic variants
653 $a Sequencing data
690 $a 0308
690 $a 0369
690 $a 0566
710 2 $a University of Toronto (Canada). $b Dalla Lana School of Public Health. $3 3186959
773 0 $t Dissertations Abstracts International $g 81-06B.
790 $a 0779
791 $a Ph.D.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27539473