%0 Journal Article
%T Predicting Membrane Protein Types Using The Random Forests Algorithm<br>随机森林方法预测膜蛋白类型
%A 袁敏
%A 胡秀珍
%J 生物物理学报
%D 2009
%I 
%X There is a close relationship between the function and types of membrane proteins, so the prediction of membrane protein types is an important means in the research of protein function, and it has significance to predict the membrane protein types from the protein amino acid sequence. Based on the amino-acid sequence in protein, eight types of membrane proteins were predicted by using the algorithm of random forests with the increment of diversity and the pseudo amino acid composition information. The overall prediction accuracy was 93.8% for independent test and 86.3% for jackknife test respectively, better than results of previous prediction.
%K Biological membrane protein
%K Random forests algorithm
%K Increment of diversity
%K Discrete Fourier spectrum
%K Pseudo amino acid composition<br>生物膜蛋白
%K 随机森林法
%K 离散增量
%K 离散傅里叶谱
%K 伪氨基酸组分
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=90BA3D13E7F3BC869AC96FB3DA594E3FE34FBF7B8BC0E591&jid=E0C9D9BBED813D6674AC13E942EAC86D&aid=B37954BED0EBE65CA976120A1F133402&yid=DE12191FBD62783C&vid=C5154311167311FE&iid=94C357A881DFC066&sid=31125890FF093250&eid=26AEEED215BE97D0&journal_id=1000-6737&journal_name=生物物理学报&referenced_num=0&reference_num=0