蛋白质结构预测是生物信息学研究的重要问题,而蛋白质二级结构预测是蛋白质结构预测的关键步骤。文中通过BLAST工具得到Identity小于等于35%的46个蛋白质复合物的单链作为数据集,分别采用5位编码和Profile编码,通过不同大小的滑动窗口,对蛋白质二级结构进行预测。实验结果显示,富含“生物进化信息”的Profile编码有着明显的优势,各种精确度均得到了较好的结果,尤其是精确度QE明显高于5位编码的QE。
Prediction of protein structure plays an important role in the research of bioinformatics, and prediction of secondary structure is the key step to protein structure predieition. Using BLAST, and get 46 protein's single chains who's identities not more than 35% as the data set. With 5 encoding and Profile encoding, to prediction protein' s secondary structure by different information windows. The experiment show that, Profile encoding method which is rich in "biological evolution information" gain the higher precision, the precision of QE is more higher than the precision QE of 5 encoding.