We share the CB6133, CB5926, CB513 and CASP10, CASP11 datasets with the corrected pi-helical assignments as discussed in the PiPred paper (submitted).
The structure of the files is identical with the originals.
Please note that CB dataset matrices are saved in the reshaped form (N, 700, 56), where N is the number of entries in the dataset.
cb513+profile_split1_updated.npy.gz
cullpdb+profile_5926_filtered_updated.npy.gz
cullpdb+profile_5926_updated.npy.gz
cullpdb+profile_6133_filtered_updated.npy.gz
cullpdb+profile_6133_updated.npy.gz
casp10_updated.h5
casp11_updated.h5
Results of scanning 7700 PFAM families for which no homologs of known structure are available:
PiPred_PFAM_scan.xls