The Research of Protein Secondary Structure Prediction System Based on KDTICM

Prediction, whereas the accuracies of them are no more than 80%. In this case, there is an urgent need to introduce a novel, high-accuracy prediction method. Based on the theory of KDTICM, we proposed a model, which is composed of four layers by intelligent interfaces, synthesizes several methods, such as SVM, KDD* and so on. From experiments, this model obtained Q3 accuracy 83.06%, 80.49% on RS126 and CB513, and for the proteins containing more alpha/beta structure; the Q3 accuracy obtained is 93.12%.