摘要
细胞凋亡蛋白的亚细胞位置与它的功能紧密相联.基于一个凋亡蛋白的亚细胞位置主要决定于它的氨基酸序列这一观点,提出了一种新的预测凋亡蛋白亚细胞位置的算法离散量方法.计算了蛋白质一级序列中紧邻残基对的出现个数,作为离散源中的参数,利用离散增量极小化对四类凋亡蛋白进行定位预测.采用Zhou和Doctor使用的数据库,通过Re-sub-stitution检验和Jack-knife检验方法,离散量方法比他们使用的协变判别式算法总体预测成功率分别高1.0%和12.2%;采用我们自己整理的扩大以后的数据库,通过Re-substitution检验和Jack-knife检验方法,总体预测成功率分别为88.1%和78.1%.
Subcellular location of apoptosis proteins is closely related to their function. Based on the concept that the subcellular location of an apoptosis protein is mainly determined by its amino acid sequence, a new algorithm for prediction of the subcellular location of an apoptosis protein is proposed. By use of adjacent residue-pair contents in the primary sequence of apoptosis proteins, the increment of diversity (ID), the sole prediction parameter, is calculated. The higher predictive success rates than the previous covariant discriminant algorithm is obtained by the re-substitution and jackknife tests using the same dataset. These results show that the algorithm is a useful for predicting the subcellular location of apoptosis proteins, and the adjacent residue-pair contents play a key role on the subcellular location of apoptosis proteins.
出处
《内蒙古大学学报（自然科学版）》
CAS
CSCD
北大核心
2004年第4期413-417,共5页
Journal of Inner Mongolia University：Natural Science Edition
基金
国家自然科学基金资助项目(30160025)
内蒙古大学青年科学基金资助项目(202045).
关键词
细胞凋亡蛋白
亚细胞位置
离散增量
紧邻残基对
apoptosis protein
subcellular location
increment of diversity algorithm
adjacent residue-pair contents
