计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (2): 133-134.

• 数据库、信号与信息处理 • 上一篇    下一篇

结合中心氨基酸组成成分预测固有不规则蛋白质

贺 波,王科俊,冯伟兴,许诺琳   

  1. 哈尔滨工程大学 自动化学院,哈尔滨 150001
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-01-11 发布日期:2012-01-11

Predicting intrinsically disordered proteins by using central amino acid compositions

HE Bo, WANG Kejun, FENG Weixing, XU Nuolin   

  1. College of Automation, Harbin Engineering University, Harbin 150001, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-01-11 Published:2012-01-11

摘要: 在固有不规则蛋白质结构预测过程中,针对短的不规则结构区域特征提取困难,提出一种结合中心氨基酸组成成分进行预测的方法。利用滑窗技术,计算20种氨基酸在窗口内出现的频率,构建一个子预测器;计算窗口中心氨基酸形成不规则结构的统计概率,以此作为新的特征参数;对子预测器的结果与新的特征参数分别赋予一个系数,进行加权组合,建立基于组合模型的固有不规则蛋白质结构预测器。实验结果表明,该预测器在保持对长的不规则结构区域预测精度较高的前提下,能够显著提高短的不规则结构区域的预测精度。

关键词: 蛋白质, 固有不规则结构, 预测

Abstract: Due to the difficulties with feature extraction in the short disordered regions, a method based on the central amino acid compositions is proposed to predict intrinsically disordered regions. The frequency of amino acids in the windows is computed to build a sub-predictor by using sliding windows. The statistics probability that the central amino acids form disordered structure is computed as new feature. The results of sub-predictor and new features are combined by weighting coefficients and build a predictor for intrinsically disordered proteins based on the combined model. The results show that this predictor not only obtains slightly higher accuracy than the most predictors for long disordered regions, but also improves significantly the predicting accuracy of short disordered regions.

Key words: protein, intrinsically disorder, prediction