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

Posterior decoding methods for optim...

Schwartz, Ariel Shaul.

FindBook

Google Book

Amazon

博客來

Posterior decoding methods for optimization and accuracy control of multiple alignments.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Posterior decoding methods for optimization and accuracy control of multiple alignments./
作者:	Schwartz, Ariel Shaul.
面頁冊數:	118 p.
附註:	Adviser: Lior Pachter.
Contained By:	Dissertation Abstracts International68-08B.
標題:	Biology, Bioinformatics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3275596
ISBN:	9780549169482

Posterior decoding methods for optimization and accuracy control of multiple alignments.
Schwartz, Ariel Shaul.

Posterior decoding methods for optimization and accuracy control of multiple alignments. - 118 p.

Adviser: Lior Pachter.

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

Comparative genomics analyses, such as genome annotation, phylogenetic inference, protein structure prediction and protein function prediction, rely heavily on multiple sequence alignment (MSA), which is essentially a homology prediction problem at the character level over multiple sequences. Alignments are also crucial in other settings, including natural language processing, as seen in word alignment in machine translation, or in the problem of citation entity alignment that is introduced in this dissertation. Fundamental as it is, MSA is a hard computational problem, and producing accurate MSAs remains challenging.

ISBN: 9780549169482Subjects--Topical Terms:

1018415
Biology, Bioinformatics.

Posterior decoding methods for optimization and accuracy control of multiple alignments.
LDR:03016nam 2200289 a 45 001 963506
005 20110831
008 110831s2007 ||||||||||||||||| ||eng d
020 $a 9780549169482
035 $a (UMI)AAI3275596
035 $a AAI3275596
040 $a UMI $c UMI
100 1 $a Schwartz, Ariel Shaul. $3 1286563
245 1 0 $a Posterior decoding methods for optimization and accuracy control of multiple alignments.
300 $a 118 p.
500 $a Adviser: Lior Pachter.
500 $a Source: Dissertation Abstracts International, Volume: 68-08, Section: B, page: 5368.
502 $a Thesis (Ph.D.)--University of California, Berkeley, 2007.
520 $a Comparative genomics analyses, such as genome annotation, phylogenetic inference, protein structure prediction and protein function prediction, rely heavily on multiple sequence alignment (MSA), which is essentially a homology prediction problem at the character level over multiple sequences. Alignments are also crucial in other settings, including natural language processing, as seen in word alignment in machine translation, or in the problem of citation entity alignment that is introduced in this dissertation. Fundamental as it is, MSA is a hard computational problem, and producing accurate MSAs remains challenging.
520 $a In this thesis we develop new posterior decoding based methods for multiple alignments of discrete characters. We begin by discussing biological sequence alignment, and describe a new alignment accuracy measure (AMA), which is based on a metric for the space of alignments. In the case of pairwise sequence alignment, it is possible to find an alignment that maximizes the expected AMA value using posterior decoding. We then describe an extension of the pairwise alignment method to MSA using a novel approach that is based on a poset representation of multiple alignments, and an efficient online topological ordering algorithm. This leads to a sequence annealing algorithm, which is an incremental method for building MSAs one match at a time. The sequence annealing algorithm outperforms all existing algorithms on benchmark test sets of protein sequence alignments. It simultaneously achieves high recall and high precision, dramatically reducing the number of incorrectly aligned residues in comparison to other programs. The method can adjust the recall/precision trade off, and can reliably identify homologous regions among protein sequences. Finally, we demonstrate that similar posterior decoding methods can be applied to the problem of multiple word alignment of citation sentences (citances ) in the bioscience literature. We extend the definition of AMA to many-to-many word alignments, and develop a posterior decoding algorithm for multiple citance alignment.
590 $a School code: 0028.
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, Berkeley. $3 687832
773 0 $t Dissertation Abstracts International $g 68-08B.
790 $a 0028
790 1 0 $a Pachter, Lior, $e advisor
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3275596