--The benchmark dataset:
--The extracted features
|[DeepSVM-fold (CCM)]a||[DeepSVM-fold (PSFM)]b||[Learning to Rank]c|
|[MotifCNN-fold (CCM)]d||[MotifCNN-fold (PSFM)]e||[DeepFR]f|
a, b: DeepSVM-fold (CCM) and DeepSVM-fold (PSFM) are contained in ;
c: Learning to Rank (LTR) is a part of FoldRec-C2C ;
d,e: MotifCNN-fold (CCM) and MotifCNN-fold (PSFM) are contained in ;
f:DeepFR refers to ;
--The ranking result of ProtFold-DFG predictor
a: ProtFold-DFG is based on the DeepSVM-fold(CCM), DeepSVM-fold(PSFM), MotifCNN-fold(CCM), MotifCNN-fold(PSFM) and LTR, uses transitive closure and KL-divergence to construct the Directed Fusion Graph and gets the final ranking result through PageRank algorithm;
1. J.y. Shao, K. Yan and B. Liu. FoldRec-C2C: protein fold recognition by combining cluster-to-cluster model and protein similarity network. Brief Bioinform 2020.
2. Li, C.C. and Liu, B. MotifCNN-fold: protein fold recognition based on fold-specific features extracted by motif-based convolutional neural networks. Brief Bioinform 2019.
3. Liu, B., Li, C.C. and Yan, K. DeepSVM-fold: protein fold recognition by combining support vector machines and pairwise sequence similarity scores generated by deep learning networks. Brief Bioinform 2019.
4. Zhu J, Zhang H, Li SC et al. Improving protein fold recognition by extracting fold-specific features from predicted residue–residue contacts, Bioinformatics 2017;33:3749-3757.