. 2002 Nov;11(11):2714–2726. doi: 10.1110/ps.0217002

Table 1.

Number of correctly identified decoys from single decoy sets by different potentials

	RAPDF^a	Atomic KBP^b	DFIRE-A^c	DFIRE-A_1.45^d	DFIRE-A_1.70^e	DFIRE-B^f
Misfold	25/25^g	25/25	25/25	25/25	25/25	23/25
Asilomar^h	33/33	33/33	33/33	33/33	33/33	32/33
Pdberr & sgpa	5/5	5/5	5/5	5/5	5/5	5/5
Ifu	31/44	33/44	34/44	32/44	34/44	18/44

^a All atom potential from Samudrala and Moult (1998)).

^b All atom potential from Lu and Skolnick (2001).

^c All-atom DFIRE-based potential (α = 1.57).

^d DFIRE-based potential (α = 1.45)

^e DFIRE-based potential (α = 1.70)

^f DFIRE-based potential (α = 1.57) for backbone and C_β atoms.

^g The first number and the second number in each cell are the number of correctly identified decoys and the total number of decoys, respectively.

^h As in (Petrey and Honig, 2000), the native structure of protein NDK is replaced by the structure of PDB code 1nue; the following eight decoys were excluded from the original set because of mismatched sequences: crabpi_vriend, edn_biosym, edn_weber, mchpr_vihinen, ndk_abagyan, ndk_vihenen, p450_abagyan, p450_weber.