English  |  正體中文  |  简体中文  |  Items with full text/Total items : 26988/38789
Visitors : 2357549      Online Users : 36
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/987654321/30368

Title: 使用 CUDA 平行處理技術加速蛋白質結構表面之立體角性質比對分析
Comparison of solid angle property on protein surfaces with CUDA acceleration
Authors: Ying-Tsang Lo
羅英倉
Contributors: NTOU:Department of Computer Science and Engineering
國立臺灣海洋大學:資訊工程學系
Keywords: 立體角;幾何表面性質;CUDA;平行計算
solid angle;geometric surface feature;CUDA;parallel processing
Date: 2011
Issue Date: 2011-11-25T07:42:08Z
Abstract: 本研究使用蛋白質結構表面的幾何特徵及CUDA平行技術偵測蛋白質交互作用的結合區域位置。立體角的幾何特性對於兩個互補結合的物件是一項重要的參考依據,該特徵除了可以顯示蛋白質表面結構的幾何特性外,也可以做為定義表面區域上錨點胺基酸的有效標準。本研究除了使用立體角的特性外,並額外加上幾個具有影響力的物理特性,期能加強評估兩個蛋白質是否具有形狀互補特徵並可以進行交互作用的可能性。 本論文所提出的演算法主要是適用於酵素-酵素抑制蛋白質複合體及蛋白質受體-配體結合之資料庫,可以有效的預測蛋白質表面的結合位置,主要是因為該類型的結合區域大多都具有”突出”或者”凹陷”的互補特徵。除了藉由表面幾何特性定義出蛋白質的結合區域外,本論文另一個貢獻是加入CUDA平行處理架構,實現高效能蛋白質結構交互作用分析的計算方法。研究結果顯示本演算法對於尋找蛋白質的結合位置在蛋白質-蛋白質結合資料集及蛋白質-配體結合資料集分別具有62.86%及80.4%的正確性,在計算效率提升的貢獻上,本論文使用GPU加速的技術最高可以提升70倍的加速效果。
An improved approach based on geometric surface features and CUDA parallel technologies for detecting protein binding sites is proposed in this thesis. The feature of solid angle is suggested as one of the most important requirements for complementary matching of two objects. It reveals the geometric characteristics of protein surface structures and provides an efficient criterion for identifying the anchor amino acids on surface regions. In addition to the solid angle feature, several influential physical properties facilitate rapid evaluation on matching shape complementary from two interactive proteins. According to the performance of the proposed algorithms on enzyme-inhibitor protein complex and protein-ligand complex datasets, the binding sites of protein surface can be effectively determined. Most of identified binding regions on protein surfaces reveal either with concave or convex characteristics, and the results conform to regular conditions of binding phenomena with respect to enzyme-inhibitor complexes and protein-ligand complexes. More importantly, the main goal of this thesis not only identifies possible protein binding regions through geometric surface characteristics, but also provides a feasible and efficient computational approach by employing the CUDA parallel computing architecture. The experiments have shown that the proposed algorithms achieved an accuracy rate of 62.86% on enzyme-inhibitor complexes and 80.4% on protein-ligand complexes for the identification of binding regions, and the GPU implementation accelerated up to 70 times compared to the CPU implementation for solid angles calculation.
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0M98570006
http://ntour.ntou.edu.tw/handle/987654321/30368
Appears in Collections:[資訊工程學系] 博碩士論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML220View/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