4 datasets found

The CASP dataset was used for testing. The dataset contains 96 template-free proteins and 90 template-based proteins.

The SCOP 2.06 dataset was used for testing. The dataset contains 4,188 domains, covering 550 folds.

The SCOP 1.75 dataset was used for training and validation. The dataset contains 16,712 proteins covering 7 major structural classes with total 1,195 identified folds.

Protein fold recognition is an important problem in structural bioinformatics. Almost all traditional fold recognition methods use sequence (homology) comparison to indirectly...