Table 2. Molecular replacement with Phaser.

Target	EdaFold
	Best TFZ model				BestCARMSD		Best CG-CARMSD
	TFZ	v- CARMSD	CARMSD
1bq9	8.0	1.15	1.08		1.04		1.30
1scjB	18.1	2.33	2.17		1.9		2.38
1hz5	6.0	23.3	2.63		1.8		2.01
1ctf	5.6	28.07	3.3		2.76		2.35
1ig5	7.0	1.85	1.7		1.52		1.84
1dtj	6.1	23.8	1.8		1.65		1.76
1a19	4.7	18.05	2.1		1.86		2.55
1bm8	5.8	17.0	3.4		2.5		2.81
1m6t	8.9	1.21	1.17		1.06		1.35
3chy	8.8	2.19	2.13		2.12		2.7
Target	Rosetta
	Best TFZ model			Best CARMSD		Best CG- CARMSD
	TFZ	v- CARMSD	CARMSD
1bq9	5.3	16.73	3.33	2.55		2.97
1scjB	10.4	4.19	3.5	2.78		3.1
1hz5	6.2	23.6	2.44	1.62		1.9
1ctf	5.6	19.0	3.06	2.3		2.39
1ig5	5.9	20.22	2.58	1.84		1.85
1dtj	5.8	21.8	1.67	1.4		1.82
1a19	5.0	19.9	2.84	1.96		2.58
1bm8	5.8	22.9	4.6	1.98		2.61
1m6t	8.0	1.33	1.23	0.96		1.54
3chy	6.1	2.53	2.38	2.31		2.78

Columns labelled Best TFZ model show measures on the predicted model with the best Phaser TFZ score. TFZ stands for Translation Function Z score, typically used to determine the probability of success of a molecular replacement test. v- CARMSD for molecular replacement verification CARMSD, represents the CARMSD of the molecular replacement model to the native structure after optimal unit cell translation and symmetric expansion of atoms in the space group. If this value is less than Å away from the value of the CARMSD to native after optimal superimposition by rigid body alignment (column CARMSD ) then the molecular replacement solution is considered successful. Best CARMSD shows the lowest CARMSD to native of the all-atom refined models used in molecular replacement. Best CG- CARMSD shows the lowest CARMSD found in the coarse-grained models.