The success rate for recognizing proteins within the same family, superfamily, or fold in the Lindahl benchmark.

aThe percentage in each cell is the fraction of correctly recognized match of proteins in the same fold, super family, and family as first rank or within top 5 rank of the template .
bFrom Ref. [10].
cFrom Ref. [48].
dFrom Ref. [11].
eFrom Ref. [12].
fThis work.
gThis work (The 43 proteins with >30% sequence similarity to PREFAB training set are removed).
hThe standard error was estimated by bootstrap simulation on 10,000 re-sampling of the data set.