利用蛋白質序列與結構關係預測酵素種類

Abstract

蛋白質結構上的active site和binding site是蛋白質功能的主要作用區域。因此定性或定量分析active site和binding site的構形和周邊胺基酸的位置分布，一直是以生物計算法預測蛋白質功能的主要方式。目前的預測蛋白質功能預測方法大多是利用已知的protein-ligand complex進行人工或自動結構分析來取得binding site周邊原子的特殊樣式(Motif)，進而預測蛋白質的可能功能。本論文則提供另一個不同的觀點，本研究團隊中認為要決定蛋白質功能雖然是由active site或是binding site的位置而決定，但是在相同功能的蛋白質之間應該會存在一些local conserved region，而這些local conserved region會是在演化過程或是產生作用的過程中變動幅度較小的區塊。換言之，這些local conserved region雖然可能不是binding site周邊的胺基酸，卻影響了active site或是binding site的結構與作用方式。為了描述這些區塊的架構，無法只單就整個單一序列或結構上的分析而決定，而是需要進一步瞭解區塊內序列與結構兩者之間的關係一併考慮。為此，本論文利用了針對每個胺基酸周圍固定範圍圈出一個球體來代表一個局部區塊的資訊，針對每個球體進行序列與結構的分析，進而利用蛋白質分類資料庫enzyme data bank，透過相同分類的蛋白質找出屬於該分類的local conserved region，並討論這些local conserved region與功能間的關係。

The active sites and binding sites of the protein structure are the main working areas of the functional proteins. Based on this assessment, the qualitative and quantitative analyses of the variant characteristics of the active site and binding site are the first steps towards predicting protein functionalities. These characteristics include conformation, bio-chemical property etc. Present methods to predict protein function mostly exploit known protein-ligand complex to conduct factitious or automatic structure analysis to obtain the Motif of surrounded atoms of binding site. However, this thesis is providing a different point of view. Our assumption is though the protein function is decided by the location of active site or binding site, there should be some local conserved regions between the proteins that possess same function. In other words, these local conserved regions might not be the surrounded amino acids of binding site, but it indeed affects the structure and the way of working for active site or binding site. To describe the framework of these areas, it can’t be decided only by a single sequence or the structure analysis. What it needs is a further understanding of the relationship between sequence and structure, and takes it into consideration. To reach this goal, we developed the following steps. First, circle each residue around 10Å as a spheroid to represent a local region. Second, use each spheroid to conduct sequence and structure analysis. Third, with enzyme data bank, we can identify the belonging local conserved region of protein through same category of protein. In the future, further discussion about the relationships between theses local conserved region and the functions should be investigated.