摘要
本文采用神经网络方法对真核基因的PolyA信号进行识别．文中选择了人类、牛、猪、猫、山羊、兔、绵羊、大鼠、小鼠、小马、鸡、仓鼠共12种真核生物的270个基因组作为研究对象，训练组包括230个基因，预测组包括40个基因．结果表明，结合PolyA上游的碱基顺序特异性（紧挨PolyA的上游地GC），可以从神经网络识别的大量可能PolyA信号中得到正确PolyA信号．
In this paper,a neural network method was applied to predict the Poly A Signal in eukaryotic gene.270 genes of 12 kinds of organisms including human,bull,pig,cat,goat,rabbit,sheep,rat,mouse,house,chicken and hamster were selected as an object of study.In thses genes,230 genes were selected as the training set to construct a network model and other 40 genes were used as the samples for prediction. As a result,by using neural network method and gene theory,the position of Poly A Singal could be obtained by computer from numerous possible Poly A singals.
出处
《生物数学学报》
CSCD
北大核心
1994年第2期60-62,共3页
Journal of Biomathematics
基金
中科院院长基金
关键词
真核基因
PolyA信号
神经网络
Eukaryote gene,PolyA singal,PolyA site.Artificial neural network.
