Table 7.

Structural prediction accuracy statistics for the methods benchmarked over 2000 random 5S RNA selections.

		Percent sequence similarity
		20–40	40–60	60–80	80–100	0–100
Dynalign new constraint	Sens	0.895	0.903	0.914	0.909	0.907
	PPV	0.838	0.824	0.821	0.785	0.821
Dynalign M constraint	Sens	0.892	0.901	0.912	0.909	0.905
	PPV	0.837	0.821	0.815	0.786	0.817
FOLDALIGN	Sens	0.726	0.753	0.787	0.510	0.749
	PPV	0.772	0.784	0.803	0.572	0.778
StemLoc	Sens	0.272	0.673	0.899	0.776	0.740
	PPV	0.652	0.805	0.901	0.826	0.840
Consan	Sens	0.666	0.799	0.931	0.865	0.842
	PPV	0.671	0.761	0.817	0.722	0.775
Single Prediction	Sens	0.680	0.722	0.766	0.784	0.739
	PPV	0.603	0.632	0.672	0.673	0.647

Results are summarized for sequence similarity values ranging from 20% through 100% in steps of 20% and for the overall data set (0 – 100). Dynalign new constraint refers to Dynalign with probabilistic alignment constraints proposed here. Software version numbers and parameters for the algorithms are described in Section 7.6.