Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (27): 164-167.

• 数据库与信息处理 • Previous Articles     Next Articles

Improved dynamic model-based clustering for time-course gene expression data

LIU Yu-hong,WANG Shi-tong,XU Hong-lin   

  1. School of Information Engineering,Southern Yangtze University,Wuxi,Jiangsu 214122,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-09-21 Published:2007-09-21
  • Contact: LIU Yu-hong

改进的时序基因表达数据动态聚类算法

刘宇宏,王士同,徐红林   

  1. 江南大学 信息工程学院,江苏 无锡 214122
  • 通讯作者: 刘宇宏

Abstract: This paper refers to a dynamic model-based clustering algorithm in [1],which can analyze a time-course gene expression data as a set of time series dynamically,such that better clustering results can be produced.Some reasonable improvements are used in the initialization hereinafter.And the joint probability distribution for the time-course gene expression dataset is also reanalyzed using Bayes theory.Experimented results demonstrate that the results obtained by the improved clustering algorithm are better than those obtained by the dynamic model-based clustering algorithm.

Key words: time-course gene expression, autoregressive equation, dynamic model, Bayes theory

摘要: 文[1]采用了一种基于动态模型的聚类算法,将时序基因表达数据作为一组时间序列进行动态的聚类分析,得到了较为理想的聚类结果。对上述算法在数据初始化方面进行了合理改进,并利用贝叶斯理论对数据的联合概率分布进行了重新分析。实验表明,提出的改进算法所得聚类结果明显优于原算法所得结果。

关键词: 时序基因表达数据, 自回归模型, 动态模型, 贝叶斯理论