A protein secondary structure prediction framework based on the support vector machine

Our framework for predicting protein secondary structures differs from existing prediction methods since we consider physiochemical information and context information of secondary structure segments. We have employed Support Vector Machine (SVM) for training the CB513 and RS126 data sets, which are collections of protein secondary structure sequences, through sevenfold cross validation to uncover the structural differences of protein secondary structures. We apply the sliding window technique to test a set of protein sequences based on the group classification learned from the training data set. Our prediction approach achieves 77.8% segment overlap accuracy (SOV) and 75.2% three-state overall per-residue accuracy (Q₃) on CB513 set, which outperform existing protein secondary structure prediction methods.