基于分组重量编码的蛋白质功能预测

A method of encoding based on grouped weight for protein function prediction

从蛋白质序列出发,采用分组重量编码(Encoding Based on Grouped Weight,简记EBGW),并结合最近邻居算法对蛋白质功能进行预测。对酵母(Saccharomyces cerevisiae)蛋白质的1826条序列进行预测,整体预测准确率与其他基于序列信息的蛋白质功能预测方法相当。实验结果表明基于EBGW编码方案的新方法可有效地应用于蛋白质功能预测。

From protein sequences,the encoding method of EBGW（Encoding Based on Grouped Weight）is applied to protein function prediction associated with the nearest neighbor algorithm.By analyzing the 1 826 Sacchammyces cerevisiae proteins,the average speciflcity precision is 83%.While the dataset is the same,this average speciflcity precision of this method is 11% higher than the Global optimization method.The experiment results show that the method of this paper is efficient to assign function to the unknown proteins.