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

Pattern discovery in sequences and g...

Chudova, Darya.

FindBook

Google Book

Amazon

博客來

Pattern discovery in sequences and gene expression data.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Pattern discovery in sequences and gene expression data./
作者:	Chudova, Darya.
面頁冊數:	272 p.
附註:	Adviser: Padhraic Smith.
Contained By:	Dissertation Abstracts International68-12B.
標題:	Biology, Bioinformatics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3293054
ISBN:	9780549368854

Pattern discovery in sequences and gene expression data.
Chudova, Darya.

Pattern discovery in sequences and gene expression data. - 272 p.

Adviser: Padhraic Smith.

Thesis (Ph.D.)--University of California, Irvine, 2007.

In the first part of this dissertation, we consider the problem of pattern discovery in categorical sequences. Motif discovery in biological sequences is an important example of such problems. Many algorithms are available for this task, but little is known about performance bounds for such algorithms. Is a pattern with a given alphabet size, length, and frequency easy to find in a sea of background symbols, or is it indistinguishable? If an algorithm fails to identify a pattern, is it because the algorithm is weak or because the pattern is weak? We answer such questions by deriving the Bayes error rate for this problem under a Markov assumption. Analytical expressions highlight the relative roles of alphabet size, pattern length and frequency in the difficulty of the pattern discovery problem. We apply our analysis to known motifs in computational biology and analyze actual and best achievable performance of several discovery algorithms.

ISBN: 9780549368854Subjects--Topical Terms:

1018415
Biology, Bioinformatics.

Pattern discovery in sequences and gene expression data.
LDR:03133nam 2200301 a 45 001 940591
005 20110518
008 110518s2007 ||||||||||||||||| ||eng d
020 $a 9780549368854
035 $a (UMI)AAI3293054
035 $a AAI3293054
040 $a UMI $c UMI
100 1 $a Chudova, Darya. $3 1264723
245 1 0 $a Pattern discovery in sequences and gene expression data.
300 $a 272 p.
500 $a Adviser: Padhraic Smith.
500 $a Source: Dissertation Abstracts International, Volume: 68-12, Section: B, page: 8122.
502 $a Thesis (Ph.D.)--University of California, Irvine, 2007.
520 $a In the first part of this dissertation, we consider the problem of pattern discovery in categorical sequences. Motif discovery in biological sequences is an important example of such problems. Many algorithms are available for this task, but little is known about performance bounds for such algorithms. Is a pattern with a given alphabet size, length, and frequency easy to find in a sea of background symbols, or is it indistinguishable? If an algorithm fails to identify a pattern, is it because the algorithm is weak or because the pattern is weak? We answer such questions by deriving the Bayes error rate for this problem under a Markov assumption. Analytical expressions highlight the relative roles of alphabet size, pattern length and frequency in the difficulty of the pattern discovery problem. We apply our analysis to known motifs in computational biology and analyze actual and best achievable performance of several discovery algorithms.
520 $a In the second part of the dissertation, we introduce non-parametric Bayesian models for discovering patterns of differential and periodic expression in time course or multi-condition gene expression data. Conventional approaches treat the profiles of individual genes independently from each other. In reality, the expression is regulated by some common biological processes and genes share common expression patterns defined by these processes. The novelty of our approach is in the explicit modeling of common expression patterns with a Dirichlet process mixture model. We develop inference algorithms to simultaneously infer relevant expression patterns and identify genes exhibiting such patterns.
520 $a Our proposed Bayesian modeling results in (1) better enrichment of relevant biological categories within the differentially expressed set and (2) identification of previously unsuspected patterns of periodic expression. We extend the model to the analysis of a non-trivial collection of data from multiple experiments profiling hair cycle regulation. The cycles are masked by initial hair morphogenesis and injury response, but information-sharing allows us to uncover the hidden structure and identify which genes are likely to participate in the regulation of the hair cycle.
590 $a School code: 0030.
650 4 $a Biology, Bioinformatics. $3 1018415
650 4 $a Computer Science. $3 626642
690 $a 0715
690 $a 0984
710 2 $a University of California, Irvine. $3 705821
773 0 $t Dissertation Abstracts International $g 68-12B.
790 $a 0030
790 1 0 $a Smith, Padhraic, $e advisor
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3293054