一种预测蛋白质结构型的新方法

The Prediction of the Structural Class of Protein: Application of the Measure of Diversity

基于蛋白质的结构类型决定于它的二级结构序列的概念 ,将蛋白质的二级结构含量和二级结构序列参数 Nα,Nβ,Nβαβ结合起来构成离散源 ,分别计算四种结构类型的标准离散量 D( Xα) ,D( Xβ) ,D( Xα/β) ,D( Xα+β) ,利用离散增量的概念 ,蛋白质的结构类型是由这个蛋白质的离散量 D( X)与四个标准离散量之间离散增量的最小值所决定的 .因此 ,对标准集中 35 9个蛋白的结构型进行检测并对检验集中 1 1 7个蛋白质进行结构预测 ,标准集的准确率为 87% ,检验集的预测准确率为 88%

According to the concept that structural class of a protein is mainly determined by its secondary structure sequence.The structural class of a protein can be predicted by using of the increment of diversity between the protein and a set of standard set of proteins.The standard sources of diversity are respectively determined by the percentage of α -helices and β -sheets and the secondary structure parameters N α,N β,N βαβ. The four increments of diversity between the measure of diversity D(X α),D(X β),D(X α/β),D(X α+β) and a measure of diversity D(X) of a new protein are respectively calculated.The structural class of a protein is determined by the lowest increment of diversity.The average rate of correct recognition is 87% for standard set of 359 proteins and that of correct prediction is 88% for test set of 117 proteins.