利用提升小波的蛋白质相互作用特征提取

Feature-mined of protein-protein interactions using lifting wavelet

利用提升小波从蛋白质序列中提取出它们相互作用的频谱特征,经支持向量机训练学习后,用于预测蛋白质间的相互作用.模拟计算结果表明,在阳性数据和阴性数据平衡的前提下,利用提升小波获取的低维蛋白质相互作用特征向量可以得到较高预测精度.进一步阐述了不同物种的蛋白质相互作用网络有着不同特征,为了得到更准确的预测结果,需要利用不同的方法提取蛋白质相互作用的特征.

The protein-protein interacting features only from the protein＇s sequence are calculated by using the lifting wavelet.These features then are learned by the support vector machine to train a model by which the protein-protein interactions are predicted.Numerical results report that,on the principle of balance between positive dataset and negative dataset,the low-dimensional vector of features has gained a better performance.Results also report that features are different among the local protein-protein interaction network of different species. For making a more accuracy prediction,it is essential to use several methods.