The structural details of the essential entity of prion disease, fibril prion protein (PrPSc), are still elusive despite the large body of evidence supporting the prion hypothesis. Five major working models of PrP Sc structure, which are not compatible with each other, have been proposed. However, no systematic evaluation has been performed on those models. We devised a method that combined systematic point mutation with threading on knowledge-based amino acid potentials. A comprehensive mutation experiment was performed on mouse prion protein, and the PrPSc conversion efficiency of each mutant was examined. The models were evaluated based on the mutation data by using the threading method. Although the data turned out to be rather more consistent with the models that assumed a conversion of the N-terminal region of core PrP into a β helix than with others, substantial modifications were also required to further improve the current model based on recent experimental results.