| 用神经网络方法由蛋白质一级序列预测其二级结构含量 |
| |
| 引用本文： | 秦红珊,杨新岐,王克起.用神经网络方法由蛋白质一级序列预测其二级结构含量[J].天津大学学报(英文版),2002,8(4):303-307. |
| |
| 作者姓名： | 秦红珊 杨新岐 王克起 |
| |
| 作者单位： | 1. 天津大学理学院,天津,300072
2. 天津大学材料科学与工程学院,天津,300072 |
| |
| 摘 要： | 基于氨基酸组成和有偏自相关函数的特征参量 ,利用BP神经网络 ,提出了一种预测蛋白质二级结构中α螺旋和 β折叠含量的计算方法 .采用相互独立的非同源蛋白质数据库对该方法的准确性进行检验 ,对蛋白质二级结构α螺旋和 β折叠含量的预测的结果为 :自检验的平均绝对误差分别为 0 .0 70和 0 .0 6 8,相应的标准偏差分别为 0 .0 49和 0 .0 47;他检验的平均绝对误差分别为 0 .0 75和 0 .0 70 ,相应的标准偏差分别为 0 .0 5 0和 0 .0 49.与常用方法相比 ,利用此方法预测蛋白质二级结构含量可有效提高预测精度 .
|
| 关 键 词： | 二级结构α和β含量 一级序列 BP神经网络 氨基酸组成 有偏自相关函数 |
Prediction of the Helix/Sheet Content of Proteins from Their Primary Sequences by Neural Network Method |
| |
| QIN Hong-shan,YANG Xin-qi,WANG Ke-qi.Prediction of the Helix/Sheet Content of Proteins from Their Primary Sequences by Neural Network Method[J].Transactions of Tianjin University,2002,8(4):303-307. |
| |
| Authors: | QIN Hong-shan YANG Xin-qi WANG Ke-qi |
| |
| Abstract: | The amino acid composition and the biased auto-correlation function are considered as features, BPneural network algorithm is used to synthesize these features. The prediction accuracy of this method is verifiedby using the independent non-homologous protein database. It is shown that the average absolute errors for re-substitution test are 0. 070 and 0. 068 with the standard deviations 0. 049 and 0. 047 for the prediction of thecontent of α-helix andβ-sheet respectively. For cross-validation test, the average absolute errors are 0.075 and0.070 with the standard deviations 0.050 and 0.049 for the prediction of the content of α-helix and β-sheet re-spectively. Compared with the other methods currently available, the BP neural network method combined withthe amino acid composition and the biased auto-correlation function features can effectively improve the predic-tion accuracy. |
| |
| Keywords: | primary sequence BP neural network amino acid com-position biased auto-correlation function |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
