A comprehensive survey of scoring functions for protein docking models

BMC Bioinformatics

Table 1 AUC ROC of all twelve scoring functions on eight different datasets

Approach	Method	CAPRI Score v2022 Difficult	CAPRI Score v2022 Easy	CAPRI Score Refined	BM4	BM5	Dockground	PDB-2023	MaSIF-test
Deep Learning	DeepRank-GNN	69.54	74.75	76.01	55.26	95.97	59.76	62.65	69.24
	GNN-DOVE	55.13	58.30	70.99	59.25	60.76	83.87	53.23	55.71
	dMaSIF	71.84	76.65	78.53	62.53	87.89	79.94	91.83	89.90
	PIsToN	64.77	80.15	81.25	70.84	91.84	68.42	93.86	93.55
Classical	FireDock	59.47	75.06	72.79	56.73	95.41	56.08	87.98	77.82
	AP-PISA	58.08	79.41	84.87	56.94	91.62	79.19	89.19	76.43
	CP-PIE	59.91	78.14	81.29	62.52	97.16	72.26	88.81	76.98
	PyDock	62.68	75.17	75.28	64.10	88.98	77.17	87.64	74.73
	ZRANK2	63.83	72.32	74.09	54.31	90.37	64.18	73.22	74.66
	RosettaDock	55.33	70.13	63.95	57.02	72.26	61.62	65.04	71.17
	SIPPER	56.72	61.80	53.03	63.67	87.70	79.60	76.25	66.17
	HADDOCK	53.05	66.59	70.34	67.59	90.85	53.15	84.36	81.85

ISSN: 1471-2105