人工智慧應用於計算生物學之研究(I)

Tunghai University Institutional Repository > 工學院 > 工業工程與經營資訊學系所 > 國科會研究報告 > Item 310901/9356

請使用永久網址來引用或連結此文件: http://140.128.103.80:8080/handle/310901/9356

題名:	人工智慧應用於計算生物學之研究(I)
其他題名:	Research Artificial Applying Intelligence in Computational Molecular Biology (I)
作者:	張炳騰 Chang, Ping-Teng
貢獻者:	行政院國家科學委員會東海大學工業工程與經營資訊系
關鍵詞:	蛋白質序列排比;基因演算法;模糊理論;仿射性間格懲罰函數 Protein sequence alignment;Genetic algorithm;Fuzzy theory;Affine gap cost
日期:	2005
上傳時間:	2011-06-17T06:20:32Z (UTC)
摘要:	本研究提出一個新的方法，用來建立蛋白質成對序列排比。目前在蛋白質序列排比上一直存在一個致命的問題，那就是當使用明確?資料去進行分析時，太多的不確定因子與敏感性資訊的遺失，導致序列排比問題出現瓶頸。由於使用不同的軟體和演算法會造成不同的結果，對於正在研究生物基因的科學家們，不同演算法亦很難廣泛地運用於基因序列上。因此基於這個最重要的前提下，本研究提出模糊的概念，將250單點突變矩陣(point accepted mutations, PAMs)與62區塊突變矩陣(blocks substitution matrix,BLOSUM)利用基因演算法(genetic algorithm, GA)於序列排比上。最主要的目的是用來減少不確定因子的影響，避免利用明確?或權重的方式，造成重要資訊的遺失，以及增加解的正確性與適用性。實驗果顯示，不論是利用PAM250還是BLOSUM62矩陣，利用GA演算法皆能找到更長且配對的蛋白質序列，並在不同矩陣的運用上，利用模糊矩陣所產生解的變動性要比明確?小，也就是說，將模糊概念運用於序列排比，確實能夠減少不確定性的影響並且增加解在區域相似上的利用性。 In this paper a novel way to construct pairwise alignment of protein sequence is proposed. Currently in protein sequence alignment the vital problem is having too many uncertain factors and causes significant data loss while using crisp data. Due to using different software and algorithms that will bring about different results, for scientists researching in protein, different algorithms will be difficult to use widespread. Therefore, for this important premise, fuzzy concept is introduced and fuzziness is implemented in the matrix for 250 point accepted mutations (PAMs) and matrix for 62 blocks substitution matrix (BLOSUM) in sequence aligning, and integrated with the Genetic algorithm (GA). The purpose for this implementation is to reduce the effects of uncertain factor, avoid making use of crisp values or weights resulting in significant data loss, and increase solution accuracy and method suitability. Results of experiment shows that application of fuzzy matrix to sequence alignment could find more continuous and identical protein sequence. Furthermore, this research used different matrix to sequence alignment. The result shows that variation of fuzzy matrix is smaller than crisp matrix. Therefore, application of fuzzy matrix certainly can reduce the effects of uncertain factor and increase solution accuracy. Hence, these results of experiment evidenced fuzzy logic useful to dealing with the uncertainties problem, and applied to protein sequence alignment successfully. The new method can provide different viewpoint for related research.
關聯:	研究編號：NSC94-2213-E029-008 研究期間：2005-08 ~ 2006-07
顯示於類別:	[工業工程與經營資訊學系所] 國科會研究報告

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
942213E029-6.pdf		363Kb	Adobe PDF	917	檢視/開啟

在THUIR中所有的資料項目都受到原著作權保護.

資料載入中.....