Protein structural class prediction using physiochemical property based grouped weighted encoding indez
In this paper, a new index called grouped weighted coding was proposed for protein structural class prediction. The component coupled algorithm was adopted to compare the new index with other two traditional indices. We used the resubstitution and jack-knife test for evaluation. The result showed that the new index was 5-7％ higher than the amino acid composition index and was 1-3％ higher than the auto-correlation function index. The advantage of efficiency, biological significance and high accuracy made grouped weighted coding index more useful in protein structural class prediction.
Protein structural class prediction grouped weighted coding physiochemical property
Kai Jiang Shuming Ye Hang Chen Fei Gu
Department of Biomedical Engineering Zhejiang University Hangzhou, China Department of Biotechnology Zhejiang University Hangzhou, China
国际会议
上海
英文
275-278
2008-05-16（万方平台首次上网日期，不代表论文的发表时间）