论文部分内容阅读
抗原表位预测是免疫信息学研究的重要方向之一,可以给实验提供重要的线索。B细胞表位或抗原决定簇是抗原中可被B细胞受体或抗体特异性识别并结合的部位。实际上,近90%的B细胞表位是构象性的。即使抗原蛋白质三级结构已知,B细胞表位预测仍然是一大挑战。该文结合实例阐述当今主要的构象性B细胞表位预测方法和算法:机器学习预测、非机器学习的计算预测、基于噬菌体展示数据的识别方法,以及一些也可用于构象性B细胞表位预测的通用蛋白质-蛋白质界面预测方法;介绍最新相关预测软件和Web服务资源,说明未来的研究趋势。
Epitope prediction is one of the important aspects of immunological informatics research, which can provide important clues to the experiment. B cell epitopes or epitopes are the sites in an antigen that are specifically recognized and bound by B-cell receptors or antibodies. In fact, nearly 90% of B cell epitopes are conformational. Even if the tertiary structure of the antigen protein is known, prediction of B cell epitopes remains a challenge. In this paper, we present the main prediction methods and algorithms of conformational B cell epitopes today: machine learning prediction, computational prediction of non-machine learning, identification methods based on phage display data, and some also can be used to predict conformation B cell epitopes General protein-protein interface prediction method; introduces the latest relevant prediction software and Web services resources, illustrating future research trends.