%0 Journal Article
%T 基于生成对抗和卷积神经网络的蛋白质二级结构预测<br>Protein Secondary Structure Prediction Based on Generative Confrontation and Convolutional Neural Network
%A 赵亚武
%A 张华兰
%A 刘毅慧
%J Hans Journal of Computational Biology
%P 49-56
%@ 2164-5434
%D 2020
%I Hans Publishing
%R 10.12677/HJCB.2020.104006
%X 在生物信息学领域，对于蛋白质二级结构预测是一项具有挑战性的任务，对于确定蛋白质的结构和功能有着极其重要的意义。本文融合了生成对抗网络和卷积神经网络模型进行蛋白质二级结构预测，首先利用生成对抗网络提取蛋白质特征，其次将生成对抗网络提取的特征结合PSSM矩阵作为卷积神经网络的输入，得到预测结果。在测试集CASP9，CASP10，CASP11，CASP12，CB513和PDB25获得了87.06%，87.24%，87.31%，87.39%，88.13%和88.93%，比单独使用卷积神经网络提高了3.88%，4.6%，7.97%，5.85%，5.78%，4.25%。实验结果表明，生成对抗网络特征提取能力是非常显著的。<br />
In the field of bioinformatics, the prediction of protein secondary structure is a challenging task, and it is extremely important for determining the structure and function of proteins. In this paper, the generation of adversarial networks and convolutional neural network models are combined for protein secondary structure prediction. First, the anti-network is generated to extract protein fea-tures. Secondly, the extracted features of the anti-network are combined with the PSSM matrix as the input of the convolutional neural network to obtain the prediction results. In the test set CASP9, CASP10, CASP11, CASP12, CB513 and PDB25 obtained 87.06%, 87.24%, 87.31%, 87.39%, 88.13% and 88.93%, which is 3.88%, 4.6%, 7.97%, 5.85%, 5.78%, 4.25% higher than the convolutional neural network alone. The experimental results show that the feature extraction ability of generat-ing adversarial networks is very significant.
%K 生物信息学，生成对抗网络，卷积神经网络，蛋白质二级结构预测<br>Bioinformatics
%K Generative Adversarial Network
%K Convolutional Neural Network
%K Protein Secondary Structure Prediction
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=38966