蛋白質摺疊率的研究

Abstract

蛋白質序列是一條由20種胺基酸所組成的線性結構，而這每一條蛋白質序列都可以對應到其特定的三維結構。蛋白質由序列到三維結構的過程稱之為「蛋白質摺疊」。蛋白質是如何摺疊為其特定的三維結構？序列和結構之間又有著什麼樣的關係存在？在生物科學的領域裡，研究尋找這關係的現象與作用一直以來都是相當重要的議題。我們試著藉由研究蛋白質摺疊率與序列結構的關係，來了解蛋白質的摺疊。不同蛋白質的蛋白質有著相當不一樣的摺疊率。通常比較小的蛋白質其摺疊所需花的時間往往比較大的蛋白質所需花的時間來要少。在本研究中，我們利用向量支持回歸(Support Vector Regression)作為主要的研究工具。在只使用序列資訊的情況下，結果和蛋白質摺疊率的相關性達80％左右。

Understanding the principles of the relationship between a primary amino acid sequence and its unique three-dimensional structures is one of the most important issues in biology science. A related and challenging task is to understand the relationship between sequences and folding rates of proteins. Proteins have different rates of folding. Small proteins usually fold faster than larger ones. We currently use amino acid sequences (which predicts properties such as protein secondary structure) as feature vectors to predict protein folding rates, using support vector regression in machine learning tool. Preliminary results show 80% correlation between the predicted and experimental folding rates.