. 2020 Apr 4;36(12):3739–3748. doi: 10.1093/bioinformatics/btaa207

Table 4.

Comparison of PIZSA, CIPS, GraphRank and iScore on the CAPRI Score_set

Target	PIZSA	CIPS	GraphRank^a	iScore^a	CAPRI best^a
T29	5/3**	5/2**	4	4	9/5**
T30	0	0	0	0	0
T32	0	0	4/1**	4/1**	2
T35	0	0	0	0	1
T37	4/1*/1	4/1*/3	2/1**	4/2**	6/1***
T39	0	0	0	0	0
T40	8/2*/5	6/4**	4/3**	4/1***	10/10***
T41	5	4/1**	8	10/2**	10/2***
T46	0	0	3	4	4
T47	8/3*/5	7/1*/6	8/5*/3	10/6*/4	10/10***
T50	2/1**	2/1**	0	4/3**	7/6**
T53	1	0	5/1**	5/1**	8/3**
T54	0	0	0	0	0
Total	7/3*/5	6/2*/6	8/1*/4	9/2*/5	10/4*/3

Note: Summary of the number of acceptable-, medium- (**) or high-quality (***) model identified by different scoring methods on various CAPRI targets. CAPRI best indicates the best result for each target obtained by any of the CAPRI participants.

Performance of methods directly adapted from Geng et al. (2020).