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

Mixed-model methods for genome-wide ...

Zhong, Sheng.

FindBook

Google Book

Amazon

博客來

Mixed-model methods for genome-wide association analysis with binary traits.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Mixed-model methods for genome-wide association analysis with binary traits./
作者:	Zhong, Sheng.
面頁冊數:	77 p.
附註:	Source: Dissertation Abstracts International, Volume: 76-02(E), Section: B.
Contained By:	Dissertation Abstracts International76-02B(E).
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3638717
ISBN:	9781321225433

Mixed-model methods for genome-wide association analysis with binary traits.
Zhong, Sheng.

Mixed-model methods for genome-wide association analysis with binary traits. - 77 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 analysis has been widely applied for the last decade, as a major tool to identify the risk variants contributing to complex diseases. For association mapping of quantitative traits, linear mixed models (LMMs) have been used to account for relatedness and population structure as well as covariates. Such methods can be applied to binary traits as well. However, the use of a linear mean structure can result in a loss of power in this case. Use of a generalized linear mixed model (GLMM) can avoid that problem, but the resulting methods can be very computationally challenging in the context of genome-wide association studies in samples with related individuals.

ISBN: 9781321225433Subjects--Topical Terms:

517247
Statistics.

Mixed-model methods for genome-wide association analysis with binary traits.
LDR:03376nmm a2200337 4500 001 1988028
005 20150716112159.5
008 150803s2014 ||||||||||||||||| ||eng d
020 $a 9781321225433
035 $a (MiAaPQ)AAI3638717
035 $a AAI3638717
040 $a MiAaPQ $c MiAaPQ
100 1 $a Zhong, Sheng. $3 2122879
245 1 0 $a Mixed-model methods for genome-wide association analysis with binary traits.
300 $a 77 p.
500 $a Source: Dissertation Abstracts International, Volume: 76-02(E), Section: B.
500 $a Adviser: Mary S. McPeek.
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 analysis has been widely applied for the last decade, as a major tool to identify the risk variants contributing to complex diseases. For association mapping of quantitative traits, linear mixed models (LMMs) have been used to account for relatedness and population structure as well as covariates. Such methods can be applied to binary traits as well. However, the use of a linear mean structure can result in a loss of power in this case. Use of a generalized linear mixed model (GLMM) can avoid that problem, but the resulting methods can be very computationally challenging in the context of genome-wide association studies in samples with related individuals.
520 $a In this thesis, we develop a new case-control association testing method, BATMAN, which is applicable to samples with related individuals and is scalable for genome-wide analysis. BATMAN is based on a novel estimating equation approach that can be viewed as a hybrid of logistic regression and LMM approaches. In terms of statistical power, BATMAN outperforms or performs as well as approaches based on either logistic regression or LMMs. Our method results in an even greater power increase over the other methods when the sample includes missing data (genotype, phenotype or covariates) and also has related individuals. In that case, we make full use of the relationship information to incorporate partially missing data in the analysis while correcting for dependence. Unlike available GLMM-based methods, our estimating equation approach is computationally efficient and stable for genome-wide analysis. Furthermore, BATMAN performs retrospective analysis, and so is robust to the misspecification of the phenotype model, because the correct calibration (i.e. type 1 error) of BATMAN only relies on the null variance of the genotype data. BATMAN is applicable to samples with arbitrarily-related individuals, including samples that combine small families and unrelated individuals as well as individuals related by large, complex inbred pedigrees, assuming that the genealogy is known.
520 $a We conduct simulation studies to confirm the correct calibration of BATMAN and its superiority in terms of power over other methods, and we apply BATMAN to analyze data on type 2 diabetes from the Framingham Heart Study.
590 $a School code: 0330.
650 4 $a Statistics. $3 517247
650 4 $a Genetics. $3 530508
650 4 $a Biostatistics. $3 1002712
690 $a 0463
690 $a 0369
690 $a 0308
710 2 $a The University of Chicago. $b Statistics. $3 1673632
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=3638717