| 基于KL散度和BP神经网络的人类基因启动子识别 |
| |
| 引用本文： | 李文举,梅丽,信润海,韦丽华.基于KL散度和BP神经网络的人类基因启动子识别[J].辽宁师范大学学报(自然科学版),2010,33(1). |
| |
| 作者姓名： | 李文举 梅丽 信润海 韦丽华 |
| |
| 作者单位： | 辽宁师范大学,计算机与信息技术学院,辽宁,大连,116081 |
| |
| 基金项目： | 辽宁省博士科研启动基金项目(20061052) |
| |
| 摘 要： | 人类基因启动子预测和识别是DNA序列分析中的一项重要任务.提出了一个基于KL散度和BP神经网络的人类基因启动子识别算法.利用KL散度提取分辨力最强的六联体来区分启动子和非启动子区域,将这些六联体的出现频率作为识别启动子的组成成分特征,结合CpG岛特征,应用BP神经网络技术建立人类启动子识别系统.该系统有3个分类器,即启动子-外显子分类器,启动子-内含子分类器和启动子-3’UTR分类器,每个分类器都是一个BP神经网络,通过3个分类器的结果来综合预测启动子序列.对测试集的实验结果为:敏感性达到51.4%,特异性达到52.9%.
|
| 关 键 词： | 启动子识别 组成成分特征 CpG岛 KL散度 BP神经网络 |
Promoter recognition in human genome based on KL divergence and BP neural network |
| |
| LI Wen-ju,MEI Li,XIN Yun-hai,WEI Li-hua.Promoter recognition in human genome based on KL divergence and BP neural network[J].Journal of Liaoning Normal University(Natural Science Edition),2010,33(1). |
| |
| Authors: | LI Wen-ju MEI Li XIN Yun-hai WEI Li-hua |
| |
| Institution: | School of Computer and Information Technology/a>;Liaoning Normal University/a>;Dalian 116081/a>;China |
| |
| Abstract: | Promoter prediction and recognition in human genome is an important task in DNA sequence analysis.We present a novel human promoter recognition algorithm based on KL divergence and BP neural network.We extract the most effective 6-mers that distinguish promoter sequenec regions from other DNA sequences regions by KL divergence,and choose frequencies of the 6-mers as the component features.We combine the component features and CpG island features,and then apply BP neural network to construct a human promoter... |
| |
| Keywords: | promoter recognition component feature CpG islands KL divergence BP neural network |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
