東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Gene-based analyses in association s...

Liu, Qianying.

Linked to FindBook

Google Book

Amazon

博客來

Gene-based analyses in association studies: From testing for gene-environment interactions to population stratification adjustment.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Gene-based analyses in association studies: From testing for gene-environment interactions to population stratification adjustment./
Author:	Liu, Qianying.
Description:	101 p.
Notes:	Source: Dissertation Abstracts International, Volume: 76-02(E), Section: B.
Contained By:	Dissertation Abstracts International76-02B(E).
Subject:	Biostatistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3638630
ISBN:	9781321224375

Gene-based analyses in association studies: From testing for gene-environment interactions to population stratification adjustment.
Liu, Qianying.

Gene-based analyses in association studies: From testing for gene-environment interactions to population stratification adjustment. - 101 p.

Source: Dissertation Abstracts International, Volume: 76-02(E), Section: B.

Thesis (Ph.D.)--The University of Chicago, 2014.

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

Genome-wide association (GWA) studies have achieved numerous successes in identifying genetic risk factors for various complex diseases, whereas a large proportion of heritability remains unexplained. Gene-environment interactions (GxEs) and rare variants are believed to contribute to the missing heritability. However, genome-wide searches for GxE are not well-powered, possibly due to small effect sizes and sparsity of interactions in the genome. In addition, single-variant analysis are underpowered for rare variants. During the recent years, gene-based analysis has become a popular approach to improve the statistical power of association/interaction testing, especially in sequencing studies with rare variants. Motivated by the emergence of gene-based studies, this dissertation work aims at answering the following two questions:

ISBN: 9781321224375Subjects--Topical Terms:

1002712
Biostatistics.

Gene-based analyses in association studies: From testing for gene-environment interactions to population stratification adjustment.
LDR:06431nmm a2200385 4500 001 1988026
005 20150716112159.5
008 150803s2014 ||||||||||||||||| ||eng d
020 $a 9781321224375
035 $a (MiAaPQ)AAI3638630
035 $a AAI3638630
040 $a MiAaPQ $c MiAaPQ
100 1 $a Liu, Qianying. $3 2122877
245 1 0 $a Gene-based analyses in association studies: From testing for gene-environment interactions to population stratification adjustment.
300 $a 101 p.
500 $a Source: Dissertation Abstracts International, Volume: 76-02(E), Section: B.
500 $a Advisers: Lin Chen; Dan Nicolae.
502 $a Thesis (Ph.D.)--The University of Chicago, 2014.
506 $a This item must not be sold to any third party vendors.
506 $a This item must not be added to any third party search indexes.
520 $a Genome-wide association (GWA) studies have achieved numerous successes in identifying genetic risk factors for various complex diseases, whereas a large proportion of heritability remains unexplained. Gene-environment interactions (GxEs) and rare variants are believed to contribute to the missing heritability. However, genome-wide searches for GxE are not well-powered, possibly due to small effect sizes and sparsity of interactions in the genome. In addition, single-variant analysis are underpowered for rare variants. During the recent years, gene-based analysis has become a popular approach to improve the statistical power of association/interaction testing, especially in sequencing studies with rare variants. Motivated by the emergence of gene-based studies, this dissertation work aims at answering the following two questions:
520 $a 1. How to improve the power of detecting GxE at the gene-level, when the interaction effects are sparse in the human genome?
520 $a 2. What is the effect of population stratification on gene-based association studies with multiple rare and common variants?
520 $a Chapter 2 and 3 provide a solution to the first question. In Chapter 2, I introduce a general framework for jointly testing a set of parameters under sparse alternative hypotheses with two unified steps -- filtering and testing. I propose to conduct a filtering test to prioritize the parameters that are more likely to be non-null, and construct a set-based test statistic with only the parameters passing the filtering. In this way, the proportion of non-null parameters contributing to the set-based test statistic increases and power is improved. With derivation of the exact distribution of the overall set-based test statistic and approximation of its power function, the optimal filtering threshold can be chosen by maximizing the approximated power. The optimal filtering threshold depends on many factors and needs to be selected adaptively for different sets of parameters. The proposed method is widely applicable and can be extended to meta-analysis. The proposed set-based test statistic, which is essentially a weighted sum of individual-level testing statistics, also has a Bayesian interpretation.
520 $a The general set-based testing framework proposed in Chapter 2 is further developed in Chapter 3 to be tailored for unique data in genome-wide GxE analysis. I will introduce the filtering strategies for binary traits and quantitative traits, respectively. The optimal filtering threshold is chosen for each gene adaptively to maximize power, with considerations of potential linkage disequilibrium (LD). In case-control studies, a sequential sampling strategy is used to calculate p-values in the presence of LD. The proposed method outperforms existing two-stage methods in the literature. In particular, gene-based GxE test is more powerful with the proposed adaptive filtering threshold, compared to no filtering and with a fixed filtering threshold. In addition, the performance of the proposed method is robust to misspecification of parameters such as effect size and the number of single-nucleotide polymorphisms (SNPs) with GxE effects in a gene. As an example, I apply the proposed method to the PanScan GWA study to detect gene by gender interactions for pancreatic cancer, and show that the proposed method improves power by accumulating multiple GxE effects in a gene and filtering SNPs without GxE effects.
520 $a Chapter 4 examines the effects of population stratification on gene-based association studies. Population stratification is one of the main confounding factors that inflate the type I error rate in genetic association studies, and we have now efficient solutions for population stratification adjustment for single-SNP inference in GWA studies. However, it is unclear whether these solutions can be effectively applied to gene-based association tests with sequencing data. I examine, both theoretically and empirically, the performance of two commonly used approaches for population stratification adjustment -- genomic control and principal component analysis (PCA) -- when used on gene-based association tests.
520 $a This dissertation focuses on statistical issues related to gene-based or set-based tests with genetic association data, when there is additional filtering information for variant-selection, and when traditional SNP-based methods for controlling population stratification may not apply. The proposed unified set-based testing framework is motivated by GxE testing with GWA data but can be extended to various other settings, including meta-analysis of GxE effects, or GxE analysis with sequencing data involving both common and rare variants. The message for the effects of population stratification on set-based methods is also illuminating: score-type of set-based tests can accumulate unmeasured/unadjusted stratifications and result in various levels of inflated type I error rate for different genes. The work presented will contribute to the statistical literature for set-based association test and, in particular, GxE association test. Software will be distributed to facilitate analyses on cancer and other complex diseases. (Abstract shortened by UMI.).
590 $a School code: 0330.
650 4 $a Biostatistics. $3 1002712
650 4 $a Statistics. $3 517247
650 4 $a Genetics. $3 530508
690 $a 0308
690 $a 0463
690 $a 0369
710 2 $a The University of Chicago. $b Health Studies. $3 2049883
773 0 $t Dissertation Abstracts International $g 76-02B(E).
790 $a 0330
791 $a Ph.D.
792 $a 2014
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3638630