. Author manuscript; available in PMC: 2021 Sep 1.

Published in final edited form as: Structure. 2020 Jul 9;28(9):1071–1081.e3. doi: 10.1016/j.str.2020.06.006

Table 1.

ClusPro’s performance in top 10 and top 30 predictions for 227 targets in benchmark 5.0 (BM5)

Target type	Number of targets tested^a	Performance by target difficulty^b			“Good” models^d
Target type	Number of targets tested^a	Easy^c	Intermediate^c	Difficult^c	Number	%
Enzyme	88/61/13/14	44/2*/18	4/0*/2	3/0*/1	51	57.95%
Enzyme	88/61/13/14	50/2*/21	5/0*/2	4/0*/2	59	67.04%
Unbound antibody	26/20/5/1	11/0*/6	1/0*/0	0/0*/0	12	46.15%
Unbound antibody	26/20/5/1	12/0*/6	2/0*/0	0/0*/0	14	53.85%
Bound^e antibody	12/11/0/1	7/2*/4	N/A	0/0*/0	7	58.33%
Bound^e antibody	12/11/0/1	10/3*/6	N/A	0/0*/0	10	83.33%
“Others”	101/57/26/18	15/0*/5	9/0*/2	3/0*/1	27	26.73%
“Others”	101/57/26/18	22/0*/5	9/0*/2	5/0*/1	36	35.64%
Total^c	227/149/44/34	87/4*/33	14/0*/4	6/0*/2	97	42.73%
Total^c	227/149/44/34	94/5*/38	16/0*/4	9/0*/3	119	52.42%
% “Good” predictions		51.68%	31.82%	17.65%
% “Good” predictions		63.09%	36.36%	26.47%

Total/easy/medium/difficult targets

Number of targets with “good” (acceptable or better) predictions, among them predictions with high (***) and medium (**) accuracy.

Line 1: top 10 predictions, line 2: top 30 predictions

Targets with “good” (acceptable or better) predictions

Targets with the antibody structure from the complex