Classification of protein fold is a critical step in protein structure prediction. TA-fold classifys the protein fold by combing the results from two algorithms, HH-fold and SVMfold. HH-fold is a template-based fold assignment algorithm using the HMM-HMM alignment program HHsearch. SVM-fold is a support vector machine-based ab-initio classification algorithm. When there are homoglogous templates to the query protein, HH-fold prediction is reported. Otherwise, the SVM-fold prediction is returned. Click here to read the workflow of TA-fold.
J. Xia, Z. Peng, D. Qi, H. Mu, J. Yang, An ensemble approach to protein fold classification by integration of template-based assignment and support vector machine classifier, Bioinformatics, 33: 863-887, 2017.