福州大学学报（自然科学版）2017，Vol.45Issue(1)：16-24,9.DOI:10.7631/issn.1000-2243.2017.01.0016
以位置特异性得分矩阵和基因本体为特征的蛋白质亚细胞定位预测
Predicting subcellular localization of protein based on PSSM and GO features
摘要
Abstract
This paper puts forward a computational method for the problem of predicting protein subcellular localization.It uses features of the position specific scoring matrix (PSSM) and gene ontology (GO),and uses the support vector machine (SVM) to construct a classifier.This method considers the effect of evolutionary information to the subcellular location and also employs text mining method a logarithmic transfornation of CHI to determine weight for GO features to achieve higher precision.Furthermore,using SVM as base classifier to construct multi-label classification model,it is performed on two benchmark datasets.The application result shows that the proposed method effectively improves the accuracy of prediction of protein subcellular localization.关键词
定位预测/蛋白质亚细胞/位置特异性得分矩阵/基因本体/多标签分类Key words
location predict/protein subcellular/position specific scoring matrix/gene ontology/multi-label classification分类
信息技术与安全科学引用本文复制引用
刘冰静,郭红..以位置特异性得分矩阵和基因本体为特征的蛋白质亚细胞定位预测[J].福州大学学报（自然科学版）,2017,45(1):16-24,9.基金项目
福建省自然科学基金资助项目(2012J05114) （2012J05114）
福建省产学研重大专项基金资助项目(2012G106) （2012G106）
