UNI-FOLD has shown excellent performance in several benchmarks for protein structure alignment quality, including Lindahl (Lindahl and Elofsson, 2000) and SCOP (http://astral.berkeley.edu). The figure below illustrates the comparison in accuracies of UNI-FOLD, PSI-BLAST and HMMER on the Lindahl dataset. A more extensive benchmarking data that includes other methods' accuracies can be found in the UNI-FOLD paper (Poleksic and Fienup, 2008).
Lindahl,E. and Elofsson,A. (2000) Identification of related proteins on family, superfamily and fold level. J Mo. Biol. 295, 613-625.
Poleksic,A. and Fienup,M. (2008) Optimizing the size of the sequence profiles to increase the accuracy of protein sequence alignments generated by profile-profile algorithms. Bioinformatics, 24, 1145-53.