Prediction of protein structural classes based on predicted secondary structure

Abstract

Prediction of protein structural classes is an important area in bioinformatics, it is beneficial to research protein function, regulation and interactions. In this paper, a 20-dimensional feature vector is extracted based on the predicted secondary structure sequence and the corresponding E-H sequence. Hierarchical classification model based on flexible neural tree (FNT) which is a special kind of artificial neural network with flexible tree structures is used to complete the experiment. 640 dataset and 25 pdb dataset with low homology are chosen as the test dataset. The 10-fold cross validation test is used to test and compare this method with other existing methods. The overall accuracies of our method are 2.7 % and 3.2 % higher for the two datasets respectively.