Top: the mean and standard deviation of the scores obtained when 10-folds cross-validating Disorder723 through the relevant feature functions. Bottom: baselines using Disorder723 to assess their model
isorder723, Casp10 and Pdb30 datasets. All datasets have roughly the same proportion of disordered residues ( 6%). Pdb30 contains times more proteins and times more residues than Casp10. Number of pro
Scripts used to enumerate Disordered Regions in Proteins as downloaded from Uniprot. Uniprot file not included (>120 GB). Following columns needed: Entry, Entry NAme, Protein names, Gene Names, Organi
