Prediction of protein structural classes based on feature selection technique
The prediction of protein structural classes is beneficial to understanding folding patterns, functions and interactions of proteins. In this study, we proposed a feature selection-based method to accurately predict protein structural classes. Three datasets with sequence identity lower than 25% were used to test the prediction performance of the method. Through jackknife cross-validation, we have verified that the overall accuracies of these three datasets are 92.1%, 89.7% and 84.0%, respectively. The proposed method is more efficient and accurate than other existing methods. The present study will offer an excellent alternative to other methods for predicting protein structural classes.
Key wordsprotein structural class feature selection technique support vector machine tetrapeptide
Unable to display preview. Download preview PDF.
- Fan, R.E., Chen, P.H., Lin, C.J. 2005. Working set selection using the second order information for training SVM. J Mach Learn Res 6, 1889–1918.Google Scholar
- Qi, Y., Liang, H., Han, X., Lai, L. 2012. Sequence Preference of α-Helix N-Terminal Tetrapeptide. Protein Pept Lett 345–352.Google Scholar
- Shafiullah, G.M., Al-Mamun, H.A. 2010. Protein strucutral class prediction using support vector machine. 6th International Conference on Electrical and Computer Engineering 179–182.Google Scholar