SVM与主动学习方法相结合的蛋白质相互作用预测  

Protein-protein Interaction Prediction Combining Active Learning with SVM

作　　者：史文丽[1] 郭茂祖[1] 李晋[1,2] 刘晓燕[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001 [2]哈尔滨医科大学生物信息科学与技术学院,哈尔滨150081

出　　处：《计算机科学》2014年第2期82-86,共5页Computer Science

基　　金：国家自然科学基金(60932008;61172098;61271346);高等学校博士学科点专项科研基金(20112302110040)资助

摘　　要：提出了基于SVM的主动学习算法,用来解决蛋白质相互作用的预测问题。细胞中的生物过程是通过蛋白质相互作用实现的。但是通过实验验证蛋白质之间是否具有相互作用的代价非常大,而且数据很难获取。为了在有限的阳性样本情况下更加快速准确地预测蛋白质之间是否具有相互作用,引入了主动学习方法。主动学习算法可以用来构造有效训练集,其目标是通过迭代抽样,每次寻找最富有信息量的数据点,找到最有利于提升分类效果的样本,进而减小分类训练集的大小。比较了5种不同的主动学习算法,以寻找在有限资源前提下提高分类算法效率的最佳途径。实验表明,主动学习方法与SVM算法相结合,能够在保证SVM分类性能的前提下,有效减少学习所需的样本数量。An active learning method using SVM was introduced in this paper to solve the problem of protein-protein in- teraction prediction task. Biological processes in cells are carried out through protein-protein interactions. Since determi- ning whether a pair of genes interacts by wet-lab experiments is resource-intensive, we proposed a support vector ma- chine active learning algorithm for interaction prediction. Active machine learning can guide the selection of pairs of genes for future experimental characterization in order to accelerate accurate prediction of the human gene interactome. As a method of constructing an effective training set, the goal of active learning algorithm is to find informative sample which can enhance the classification results of the model during the iteration, thereby reducing the size of the training set and improving the efficiency of the model within limited time and resources. The experiment shows that compared with the general SVM, active learning with SVM can reduce the number of examples effectively on the premise of keep- ing correctness of the classifier.

关 键 词：支持向量机 主动学习 蛋白质相互作用 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]