English  |  正體中文  |  简体中文  |  Items with full text/Total items : 27287/39131
Visitors : 2443758      Online Users : 38
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/17912

Title: 近似特徵比對技術應用於唯一樣式之偵測
Approximate Feature Matching Techniques for Unique Pattern Detection
Authors: Pei-Chih Wu
吳佩芷
Contributors: NTOU:Department of Computer Science and Engineering
國立臺灣海洋大學:資訊工程學系
Keywords: 專一蛋白質片段;近似特徵比對;強化合併技術;位元分類方法
Unique Peptide Motif;Approximate Feature Matching;Reinforced Merging Algorithm;Bitwise Clustering Method
Date: 2004
Issue Date: 2011-07-04
Abstract: 蛋白質家族序列係由具備相似序列或相同生物功能的多個蛋白質組成,通常同一家族的蛋白質他們也具備相似的三度空間立體結構,但是有的研究報告顯示某些高度相似性的蛋白質酵素,除了具有共同的催化功用外,還可以經由其特異的序列與其他細胞蛋白質之交互作用,進而衍化出新的功能,因此在一個蛋白質家族中需對每一條序列進行分析,各別擷取出具有代表性之專一蛋白質片段,再依據其所能提供重要的訊息及實驗證明以了解其功能的特異性。本論文提出一套基於強化合併技術之近似擷取演算法以預測專一蛋白質片段的完整系統,透過系統的定位預測來確認專一蛋白質片段的位置並進行後續抗體的研究開發。本系統的設計分為三個子系統,分別為分類、蒐尋及合併子系統,每一個子系統都含有本論文所研提的新演算法技術。分類模組執行傳統分類技術的系統設計,將20種胺基酸依使用者設定之BLOSUM/PAM相似分數矩陣進行分組,並作為後續可容忍字串比對的依據。除了傳統階層結構式的分類技術外,本論文也提出位元編碼及基本位元演算法以進行胺基酸分群運算,並會證明所提出的位元比對演算法的確具有近似特徵分類及正確比對的功能。在蒐尋模組運算中,本論文根據前分類模組之技術分別設計所需之字串蒐尋比對演算法,依完全/近似字串比對演算對可容忍字串的需求進行開發設計,論文中所提出之位元比對技術可以解決在分組過程因交叉組別所造成的誤差,我們也概念性舉例說明如何透過本論文所提之演算法來克服該問題。在最後的合併模組設計中,本論文提出新的字串合併概念,區域性的基本片段將依據鄰近關係進行重疊檢測及合併,由下而上逐漸組成更具代表性的專一蛋白質片段,在論文中我們也將推導鬆散型及嚴密型合併之相對關係,這些合併成果都將反應專一蛋白質片段唯一性之強度。
A protein family is composed of several members with highly homologous sequences and/or similar biological functions. In general, members of a protein family possess similar three-dimensional structures. However, previous experimental results revealed that enzymes with high sequence homology may acquire differential function other than the common catalytic ability, probably due to additional interactions among the variable regions and other cellular proteins. It is thus important to identify and localize the unique peptide motifs in each member of a protein family for functional analysis. In this thesis, we have suggested reinforced merging algorithms to identify the unique peptide motifs present in the highly conservative protein families. This algorithm could efficiently identify the unique peptide motifs from a set of family sequences. The commendable advantages of the proposed algorithms are able to perform approximate matching functions with tolerant characteristics, which will provide more suitable prediction results for bio-related experiments. The proposed systems contain three main phases: clustering, searching, and merging phases. In clustering phase, the module classifies 20 amino acids into different groups based on specified BLOSUM/PAM series of matrices. Traditional and novel clustering methodologies are analyzed and compared in this thesis. Searching phase performs exact/approximate string matching procedures. We have shown examples that our proposed algorithms can provide better results with respect to grouped tolerant characteristics. In the last phase, merging algorithms initiate a novel idea to extract unique peptide motif by bottom-up merging processes. This developed system will be implemented and compared with existing algorithms, and we believe that the developed tools are efficient and effective for biologists to analyze protein sequences prior to their practical laboratory experiments such as peptide antibody design.
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0M92570001
http://ntour.ntou.edu.tw/ir/handle/987654321/17912
Appears in Collections:[資訊工程學系] 博碩士論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML140View/Open


All items in NTOUR are protected by copyright, with all rights reserved.

 


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback