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

Some Statistical Methods for Spatial...

Lin, Zhixiang.

FindBook

Google Book

Amazon

博客來

Some Statistical Methods for Spatial-Temporal Brain Gene Expression Data.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Some Statistical Methods for Spatial-Temporal Brain Gene Expression Data./
作者:	Lin, Zhixiang.
面頁冊數:	104 p.
附註:	Source: Dissertation Abstracts International, Volume: 76-11(E), Section: B.
Contained By:	Dissertation Abstracts International76-11B(E).
標題:	Bioinformatics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3663541
ISBN:	9781321945096

Some Statistical Methods for Spatial-Temporal Brain Gene Expression Data.
Lin, Zhixiang.

Some Statistical Methods for Spatial-Temporal Brain Gene Expression Data. - 104 p.

Source: Dissertation Abstracts International, Volume: 76-11(E), Section: B.

Thesis (Ph.D.)--Yale University, 2015.

This item is not available from ProQuest Dissertations & Theses.

The statistical methodologies developed in this thesis are motivated by our interest in studying neurodevelopment using the spatial-temporal brain gene expression data, including the human brain microarray data and the mouse brain RNA-Seq read count data. In Chapter 2, we first develop a Markov Random Field (MRF)-based approach to identify differentially expressed genes between adjacent time periods in the human brain microarray data. Utilizing the information embedded in the similarity between regions and time points, our method achieves gain in power in both simulation studies and real data analysis. The method is implemented by a Monte-Carlo Expectation-Maximization (MCEM) algorithm. In Chapter 3, we extend the model to identify differentially expressed genes in the mouse brain RNA-Seq data, where we implement a much faster EM algorithm with mean field-like approximation. In Chapter 4, we extend the MRF model to jointly estimate Gaussian Graphical Models (GGMs). We propose a Bayesian neighborhood selection procedure to estimate the graph structures for the gene-gene interaction networks. We have developed an efficient algorithm that is fast and enables parallel computing. We also demonstrate some theoretical properties of the Bayesian neighborhood selection procedure. Chapter 5 provides a summary of the thesis and some topics for future research.

ISBN: 9781321945096Subjects--Topical Terms:

553671
Bioinformatics.

Some Statistical Methods for Spatial-Temporal Brain Gene Expression Data.
LDR:02331nmm a2200301 4500 001 2069310
005 20160513093616.5
008 170521s2015 ||||||||||||||||| ||eng d
020 $a 9781321945096
035 $a (MiAaPQ)AAI3663541
035 $a AAI3663541
040 $a MiAaPQ $c MiAaPQ
100 1 $a Lin, Zhixiang. $3 3184314
245 1 0 $a Some Statistical Methods for Spatial-Temporal Brain Gene Expression Data.
300 $a 104 p.
500 $a Source: Dissertation Abstracts International, Volume: 76-11(E), Section: B.
500 $a Adviser: Hongyu Zhao.
502 $a Thesis (Ph.D.)--Yale University, 2015.
506 $a This item is not available from ProQuest Dissertations & Theses.
520 $a The statistical methodologies developed in this thesis are motivated by our interest in studying neurodevelopment using the spatial-temporal brain gene expression data, including the human brain microarray data and the mouse brain RNA-Seq read count data. In Chapter 2, we first develop a Markov Random Field (MRF)-based approach to identify differentially expressed genes between adjacent time periods in the human brain microarray data. Utilizing the information embedded in the similarity between regions and time points, our method achieves gain in power in both simulation studies and real data analysis. The method is implemented by a Monte-Carlo Expectation-Maximization (MCEM) algorithm. In Chapter 3, we extend the model to identify differentially expressed genes in the mouse brain RNA-Seq data, where we implement a much faster EM algorithm with mean field-like approximation. In Chapter 4, we extend the MRF model to jointly estimate Gaussian Graphical Models (GGMs). We propose a Bayesian neighborhood selection procedure to estimate the graph structures for the gene-gene interaction networks. We have developed an efficient algorithm that is fast and enables parallel computing. We also demonstrate some theoretical properties of the Bayesian neighborhood selection procedure. Chapter 5 provides a summary of the thesis and some topics for future research.
590 $a School code: 0265.
650 4 $a Bioinformatics. $3 553671
650 4 $a Biostatistics. $3 1002712
650 4 $a Neurosciences. $3 588700
690 $a 0715
690 $a 0308
690 $a 0317
710 2 $a Yale University. $3 515640
773 0 $t Dissertation Abstracts International $g 76-11B(E).
790 $a 0265
791 $a Ph.D.
792 $a 2015
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3663541