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

A collection of novel algorithms for...

Keibler, Evan M.

Linked to FindBook

Google Book

Amazon

博客來

A collection of novel algorithms for low memory HMM decoding allowing for parallelization and variable sequences.

Record Type:	Electronic resources : Monograph/item
Title/Author:	A collection of novel algorithms for low memory HMM decoding allowing for parallelization and variable sequences./
Author:	Keibler, Evan M.
Description:	91 p.
Notes:	Source: Dissertation Abstracts International, Volume: 68-06, Section: B, page: 3898.
Contained By:	Dissertation Abstracts International68-06B.
Subject:	Biology, Bioinformatics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3268043
ISBN:	9780549070177

A collection of novel algorithms for low memory HMM decoding allowing for parallelization and variable sequences.
Keibler, Evan M.

A collection of novel algorithms for low memory HMM decoding allowing for parallelization and variable sequences. - 91 p.

Source: Dissertation Abstracts International, Volume: 68-06, Section: B, page: 3898.

Thesis (Ph.D.)--Washington University in St. Louis, 2007.

The hidden Markov model (HMM) has proven to be a valuable tool for probabilistic modeling in fields from computer vision to speech recognition. Recently the HMM and its variants have been successfully applied to many problems in computational biology. Memory usage and running time requirements of standard algorithms make applying these models to large sequences (such as full human chromosomes) infeasible. Here we present a novel algorithm for optimal decoding of large sequences in low, near constant memory for any length sequence. We then extend this approach to allow for optimal decoding of a single sequence using multiple processors in parallel. Using concepts derived from our work on low memory and parallel HMM decoding, we then present an algorithm for decoding sequences containing variable character (e.g. SNPs). This algorithm produces the optimal decoding for each possible variant of the sequence being considered. We implemented these algorithms in gene prediction and sequence alignment software and present results of genome scale experiments.

ISBN: 9780549070177Subjects--Topical Terms:

1018415
Biology, Bioinformatics.

A collection of novel algorithms for low memory HMM decoding allowing for parallelization and variable sequences.
LDR:01997nmm 2200277 4500 001 1835902
005 20080107105547.5
008 130610s2007 eng d
020 $a 9780549070177
035 $a (UMI)AAI3268043
035 $a AAI3268043
040 $a UMI $c UMI
100 1 $a Keibler, Evan M. $3 1924522
245 1 2 $a A collection of novel algorithms for low memory HMM decoding allowing for parallelization and variable sequences.
300 $a 91 p.
500 $a Source: Dissertation Abstracts International, Volume: 68-06, Section: B, page: 3898.
500 $a Adviser: Michael Brent.
502 $a Thesis (Ph.D.)--Washington University in St. Louis, 2007.
520 $a The hidden Markov model (HMM) has proven to be a valuable tool for probabilistic modeling in fields from computer vision to speech recognition. Recently the HMM and its variants have been successfully applied to many problems in computational biology. Memory usage and running time requirements of standard algorithms make applying these models to large sequences (such as full human chromosomes) infeasible. Here we present a novel algorithm for optimal decoding of large sequences in low, near constant memory for any length sequence. We then extend this approach to allow for optimal decoding of a single sequence using multiple processors in parallel. Using concepts derived from our work on low memory and parallel HMM decoding, we then present an algorithm for decoding sequences containing variable character (e.g. SNPs). This algorithm produces the optimal decoding for each possible variant of the sequence being considered. We implemented these algorithms in gene prediction and sequence alignment software and present results of genome scale experiments.
590 $a School code: 0252.
650 4 $a Biology, Bioinformatics. $3 1018415
650 4 $a Computer Science. $3 626642
690 $a 0715
690 $a 0984
710 2 $a Washington University in St. Louis. $3 1017519
773 0 $t Dissertation Abstracts International $g 68-06B.
790 1 0 $a Brent, Michael, $e advisor
790 $a 0252
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3268043