A new method for predicting protein fold-classes and protein domains f
rom sequence data is constructed and used for generating a data base o
f protein fold-class assignments. Any given sequence of amino acids is
assigned a specific prediction of one out of 45 typical protein fold-
classes, a prediction of one out of 4 super fold-classes for the conte
nt of secondary structures and a profile of fold-class predictions alo
ng the sequence. The prediction accuracy for the super fold-classes is
around 91% correct and 82% correct for the specific fold-classes. Thi
s accuracy is maintained down to a few percent of sequence identity.