Identifying the Mesophilic And Thermophilic Proteins From Their Amino Acid Composition With V-Support Vector Machines

摘要：

Many researchers had proved that both single amino acid composition and dipeptide composition can influence protein thermostability.We use v-support vector machines approach to predict hyperthermophilic protein,thermophilic protein and mesophilic protein from single amino acid composition, dipeptide composition and the combination of the two factors.For the prediction accuracies,we conclude that,single amino acid composition is suitable for prediction of mesophilic protein;dipeptide composition is suitable for prediction of hyperthermophilic protein and thermophilie protein;when considering the combination of the two factors, the prediction accuracy of hyperthermophilic protein is 84.1%, thermophilic protein is 83.4%,mesophilic protein is 84.4%,average accuracy is 84.0%. It shows that the protein thermostability can be predicted properly based on the combination of single amino acid composition and dipeptide composition. Obviously, dipeptide composition is correlative significantly to protein thermostability based on the prediction accuracies.

关键词： protein thermostability amino acid composition v-support vector machines

作者: Yanrui Ding Yujie Cai Jun Sun Wenbo Xu

作者单位: School of Information Technology,Jiang Nan University,Wuxi,214122,China Key Laboratory of Industrial Key Laboratory of Industrial Biotechnology,Jiang Nan University,Wuxi,214122,China School of Biotechn School of Information Technology,Jiang Nan University,Wuxi,214122,China

会议类型: 国际会议

会议名称: 2008年国际电子商务、工程及科学领域的分布式计算和应用学术研讨会(2008 International Symposium on Distributed Computing and Applications for Business Engineering and Science)

会议地点: 大连

会议语种:英文

页码: 1222-1227

在线出版日期: 2008-07-27（万方平台首次上网日期，不代表论文的发表时间）