论文部分内容阅读
蛋白质结构类的正确识别对于其三级结构预测具有十分重要的意义,有必要引入先进的算法提高预测精度。使用SIM-CA 法处理氨基酸组成、自相关系数提取的特征参数以及氨基酸对含量,进行了蛋白质结构类的预测。采用Miyazawa 和Jerni-gan 的疏水值时,All-α、All-β、αβ类的自检验的精度为89%、91%、89%,它检验的精度分别为74%、87%、91%;引入氨基酸对含量后,All-α、All-β、αβ类自检验精度为86%、89%、90%,它检验的精度为77%、88%、93%。SIMCA 的预测结果好于Bayes-ian 识别函数法,氨基酸对的引入可以提高预测精度。
Proper identification of protein structural classes is of great significance for the prediction of tertiary structure. It is necessary to introduce advanced algorithms to improve the prediction accuracy. Using SIM-CA method to process the amino acid composition, the characteristic parameters of the self-correlation coefficient extraction and amino acid content, we predicted the protein structure. The accuracy of the self-test of All-α, All-β and αβ was 89%, 91% and 89%, respectively. The accuracy of the tests was 74%, 87% and 91% The accuracy of self-test of All-α, All-β and αβ was 86%, 89% and 90% after the introduction of amino acid content. The accuracy of test was 77%, 88% and 93% respectively. The predictive result of SIMCA is better than that of Bayes-ian discriminant function. The introduction of amino acid pairs can improve the prediction accuracy.