蛋白质结构的主成分分析及氧链糖基化位点的人工神经网络预测

糖基化是蛋白质翻译后修饰的重要形式之一,氧链糖基化是糖基化的一种主要类型,对蛋白质氧链糖基化位点进行预测具有重要的意义.以窗口长度为41的蛋白质序列为研究对象,采用稀疏编码,利用主成分分析法研究了氧链糖基化蛋白质序列的结构特点;在提取主成分的基础上,设计了一个含单隐层的BP神经网络(256—8—4),对蛋白质氧链糖基化位点进行预测,把蛋白质序列分为4类;并同直接用BP神经网络分类的结果相比较,实验结果证明提出的方法省时,准确,预测的准确率达80~90%.

Principal Component Analysis on Structure of Protein Sequence and Prediction of O-Glycosylation Sites by Neural Network

Glycosylation is one of the most important post-translation modification of protein sequence, O-linked glycosylation is the main type of Glycosylation. It is very important to predict the O-Glycosylation sites. In this paper, a detailed analysis about the structure of O-glycosylated protein has been done by principal components analysis(PCA). After decreasing the dimension by PCA, we use a layered neural networks (256-8-4)to predict the O-Glycosylation sites. The experiment result shows that our method is faster and more accurate. The prediction accuracy is about 80%～90%.