基于离散增量结合支持向量机方法的凋亡蛋白亚细胞位置预测

根据凋亡蛋白的亚细胞位置主要决定于它的氨基酸序列这一观点,基于局部氨基酸序列的n肽组分和序列的亲疏水性分布信息,采用离散增量结合支持向量机(ID_SVM)算法,对六类细胞凋亡蛋白的亚细胞位置进行预测。结果表明,在Re-substitution检验和Jackknife检验下,ID_SVM算法的总体预测成功率分别达到了94.6%和84.2%;在5-fold检验和10-fold检验下,其总体预测成功率也都达到了83%以上。通过比较ID和ID_SVM两种方法的预测能力发现,结合了支持向量机的离散增量算法能够改进预测成功率,结果表明ID_SVM是预测凋亡蛋白亚细胞位置的一种很有效的方法。

PREDICTING SUBCELLULAR LOCATION OF APOPTOSIS PROTEINS USING THE ALGORITHM OF THE INCREMENT OF DIVERSITY COMBINED WITH SUPPORT VECTOR MACHINES

According to the concept that the subcellular location of an apoptosis protein is mainly determined by its amino acid sequence,the six kind of subcellular locations of apoptosis proteins were predicted by using the algorithm of the increment of diversity(ID) combined with support vector machines(ID_SVM) based on the n-peptide components of local amino acid sequence and hydropathy and hydrophobicity.The results of Re-substitution and Jackknife tests showed that total predictive success rates for ID_SVM algorithm were 94.6% and 84.2%,respectively.The results of 5-fold cross-validation(5-CV) and 10-fold cross-validation(10-CV) tests showed that total predictive success rates were higher than 83%.By comparing the predicted ability of ID with ID_SVM,the authors found that the predictive success rate could be improved by combining ID-model with support vector machines.These results indicate that the ID_SVM algorithm is an effective method for predicting the subcellular location of apoptosis proteins.