基于数据挖掘与机器学习的蛋白质疏水性分析的研究  

Study on Protein Hydrophobicity Analysis Based on Data Mining and Machine Learning

作　　者：周斯涵 刘月兰[1] 

机构地区：[1]哈尔滨师范大学

出　　处：《哈尔滨师范大学自然科学学报》2017年第3期34-38,共5页Natural Science Journal of Harbin Normal University

摘　　要：蛋白质的疏水性对蛋白质的稳定性、构象和蛋白质功能具有重要意义,通过数据挖掘中的机器学习算法实现了将一个数据集中已知疏水性的多个蛋白质样本数据,分配给具有特征值的各个目标类.将这些已知其特定类归属的数据作为KNN,LR,决策树,SVM四类分类器的训练集,利用这些已知数据训练后的分类器来处理未知疏水性的蛋白质数据,最终判断该数据的分类.该算法对蛋白质疏水性的预测,其准确率可达90%以上.The hydrophobicity of proteins is important for protein stability,conformation and protein function.Through the machine learning algorithm in data mining,we can allocate a plurality of protein sample data with known hydrophobicity in one data set to each target class with eigenvalues. The data belonging to the KNN,LR,decision tree and SVM classifier are used as the training set of these classifications. Using the known data after training the classifier to deal with unknown hydrophobic protein data,and ultimately to determine the classification. The algorithm can predict the hydrophobicity of protein,and its accuracy rate is over 90%.

关 键 词：机器学习 蛋白质 数据挖掘 疏水性 SVM 

分 类 号：R857.3[医药卫生—航空、航天与航海医学]