| 引用本文: | 包文荣,赵巨东.用ID-SVM预测蛋白质的ATP结合位点[J].生物信息学,2013,11(3):181-185. |
| BAO Wen-rong,ZHAO Ju-dong.Use ID-SVM method to prediction ATP binding residues of a protein[J].Chinese Journal of Bioinformatics,2013,11(3):181-185. |
|
| 摘要: |
| 从蛋白质序列出发,对经Dr.G.P.S.Raghava整理和使用过的168条非冗余的ATP与蛋白质结合氨基酸序列进行分段,对ATP与蛋白质结合位点进行了统计分析。在此基础上,利用20种氨基酸的亲疏水性将20种氨基酸约化为6类。以氨基酸组分和6类亲疏水紧邻为参数,用多样性增量(ID)方法将氨基酸组分和6类亲疏水紧邻降维并将降维后的特征参数输入支持向量机中运算,本文运算结果显示用氨基酸组分ID值和6类亲疏水紧邻ID值共同作为特征参数结果最优,在七交叉检验下的预测总精度达到了99.67%,相关系数达到0.9934,好于前人的预测结果。 |
| 关键词: 多样性增量 支持向量机 6类亲疏水紧邻 三磷酸腺苷(ATP) |
| DOI：10.3969/j.issn.1672-5565.2013-03.20130304 |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(51068020)。 |
|
| Use ID-SVM method to prediction ATP binding residues of a protein |
|
BAO Wen-rong,ZHAO Ju-dong
|
|
(Faculty of Science,Inner Mongolia University of Technology, Hohhot 010051, China)
|
| Abstract: |
| starting from the protein sequence, Dr.GPSRaghava are analyzing the 168non-redundant ATP and protein segmentation which these are organized and used, statistical and analysis of ATP and protein-binding sites. On this basis, the use of 20kinds of amino acids hydrophobicity of the 20amino acids is reduced to 6. Close to the amino acid composition and 6Hydrophobicity parameter increment of diversity (ID) close to the amino acid composition and 6Hydrophobicity dimensionality reduction and dimensionality reduction characteristic parameter input support vector machine computing, this article the result of the operation is displayed next to the amino acid component ID value and 6Hydrophobicity ID value as a common characteristic parameters of the best results, seven cross-examination under the forecast total accuracy of 99.67%, correlation coefficient of 0.9934, better than the previous forecast results. |
| Key words: Increment of diversity Support Vector Machines Diad Seven Crosscheck Adenosine-triphosphate |
