DescFold: A web server for protein fold recognition

BMC Bioinformatics

Table 5 Comparison of receiver operator characteristics (< = 10 false positives) for different fold recognition methods based on all LiveBench-2008.2 targets.^a

	Receiver operator characteristics (< = 10 false positives)^b										Sensitivity^c
	1	2	3	4	5	6	7	8	9	10	All	Trivial	Easy	Hard
FFAS^d	121	174	205	218	228	263	267	269	278	278	302	15	218	69
Inub^d	29	34	126	149	183	195	209	210	211	228	257	14	189	54
Fugue^d	129	186	199	219	221	223	224	225	225	225	285	16	213	56
mGenThreader^d	179	197	205	211	215	215	216	222	232	232	290	16	215	59
3D-PSSM^d	25	75	83	97	127	140	175	176	178	179	220	12	181	27
DescFold^e	158	190	190	211	215	212	215	220	224	224	294	15	210	69

^a LiveBench-2008.2 has a total number of 513 targets, including 16 trivial, 246 easy and 256 hard targets. Please refer to the footnote of Table 4 for the definitions of trivial, easy and hard targets.
^b1-10: number of correct predictions with higher reliability than the 1-10 false prediction.
^c Number of correct predictions for all, trivial, easy and hard targets, respectively.
^d The results for FFAS, Inub, Fugue, mGenThreader, and 3D-PSSM were cited from http://meta.bioinfo.pl/results.pl?comp_name=livebench-2008.2
^eThe performance was evaluated based on the number of correctly assigned folds. We considered two hits as similar, provided that the Z-Score obtained by applying the CE structural alignment algorithm [51] was > = 4.0.

ISSN: 1471-2105