Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (5): 55-58.DOI: 10.3778/j.issn.1002-8331.2009.05.017

• 研究、探讨 • Previous Articles     Next Articles

Prediction of melting temperature of DNA duplex by least squares support vector machine

LI Jin-song,ZHANG Qiang,ZHOU Shi-hua   

  1. Liaoning Key Lab of Intelligent Information Processing,Dalian University,Dalian,Liaoning 116622,China
  • Received:2008-01-14 Revised:2008-08-29 Online:2009-02-11 Published:2009-02-11
  • Contact: LI Jin-song

双链DNA解链温度的最小二乘支持向量机预测方法

李金松,张 强,周士华   

  1. 大连大学 辽宁省智能信息处理重点实验室,辽宁 大连 116622
  • 通讯作者: 李金松

Abstract: In DNA computing,one requirement of the encoding DNA sequences is that they should keep similar thermodynamic stability in order to maintain the reliability of the computing process.Melting temperature is an important parameter used to evaluate the thermodynamic stability of DNA sequences.In this paper,the least squares support vector machine(LSSVM) is used to predict the melting temperature based on the Nearest-Neighbor(NN) mode and the precision of the prediction result is in ±5 ℃ error band.

Key words: Nearest-Neighbor(NN) mode, melting temperature, Least Squares Support Vector Machine(LSSVM), DNA computing

摘要: 在DNA计算中,为了确保计算结果的精度和可靠性,要求每个进行编码的DNA分子具有相同或者近似的热力学性质,解链温度Tm是评价DNA分子的热力学稳定性的一个重要的参数。以DNA序列的邻近法参数为基础,应用最小二乘支持向量机(LSSVM)的方法对解链温度进行预测。结果表明,DNA序列的解链温度误差可以达到±5 ℃的范围。

关键词: 邻近法模型, 解链温度, 最小二乘法支持向量机(LSSVM), DNA计算