A New Coding Scheme for Protein Secondary Structure Prediction Using Support Vector Machines

摘要：

A new coding scheme for protein secondary structure prediction is presented in this paper. The data set is first encoded by the pair-coupled amino acid composition. And then the conformation propensity factors are added. The re-encoded data set is trained by support vector machine (SVM). In test phase, the samples are encoded three times respectively, and a Max Wins algorithm is used to determine the final result. The experimental results on the CB513 data set show the proposed method is one of excellent methods for protein secondary structure prediction.

关键词： protein secondary structure prediction conformation propensity factors support vector machine coding scheme

作者: Jun Guo Fuming Lin Su Wang Xiaoping Liu Youguang Chen

作者单位: Computer Center East China Normal University 3663 Zhong Shan Rd. N., Shanghai, China

会议类型: 国际会议

会议名称: 2010 Second Asia-Pacific Conference on Information Processing(2010年第二届亚太地区信息处理国际会议 APCIP 2010)

会议地点: 南昌

会议语种:英文

页码: 109-112

在线出版日期: 2010-09-17（万方平台首次上网日期，不代表论文的发表时间）