Lin Hao, Wang Hao, Ding Hui, Chen Ying-Li, Li Qian-Zhong
Center for Bioinformatics, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China.
Acta Biotheor. 2009 Sep;57(3):321-30. doi: 10.1007/s10441-008-9067-4. Epub 2009 Jan 24.
Apoptosis proteins play an essential role in regulating a balance between cell proliferation and death. The successful prediction of subcellular localization of apoptosis proteins directly from primary sequence is much benefited to understand programmed cell death and drug discovery. In this paper, by use of Chou's pseudo amino acid composition (PseAAC), a total of 317 apoptosis proteins are predicted by support vector machine (SVM). The jackknife cross-validation is applied to test predictive capability of proposed method. The predictive results show that overall prediction accuracy is 91.1% which is higher than previous methods. Furthermore, another dataset containing 98 apoptosis proteins is examined by proposed method. The overall predicted successful rate is 92.9%.
凋亡蛋白在调节细胞增殖与死亡之间的平衡中起着至关重要的作用。直接从一级序列成功预测凋亡蛋白的亚细胞定位，对于理解程序性细胞死亡和药物发现大有裨益。本文利用周的伪氨基酸组成（PseAAC），通过支持向量机（SVM）对总共317个凋亡蛋白进行了预测。采用留一法交叉验证来检验所提方法的预测能力。预测结果表明，总体预测准确率为91.1%，高于先前的方法。此外，用所提方法对另一个包含98个凋亡蛋白的数据集进行了检验。总体预测成功率为92.9%。
