Table 3.

Summary of crystallographic analysis

Data collection statistics
Data set	Cell	Resolution (Å)	Measured, unique reflections	Rmergea	I/σ(I)b	Completeness
Native	63.2 Å, 67.6 Å, 81.2 Å, 106.5°	17–1.63	333,108, 75,658	0.067 (0.506)	17.2 (2.3)	0.926 (0.887)
HgCl2	62.8 Å, 67.7 Å, 81.0 Å, 106.6°	17–2.40	95,871, 25,017	0.067 (0.157)	20.3 (6.4)	0.985 (0.940)
Ethyl-Hg-phosphate	62.8 Å, 67.7 Å, 81.0 Å, 106.6°	17–2.30	105,225, 29,046	0.054 (0.145)	19.2 (7.4)	0.997 (0.999)
Gadolinium acetate	63.2 Å, 67.5 Å, 81.1 Å, 106.5°	17–2.20	99,151, 28,984	0.042 (0.164)	25.2 (7.2)	0.917 (0.917)
Uranyl acetate	62.9 Å, 67.7 Å, 80.8 Å, 107.0°	17–2.40	91,831, 25,405	0.064 (0.173)	20.1 (5.0)	0.995 (0.992)

Phasing statistics
	17–7.1 Å	− 4.5 Å	−3.6 Å	−3.1 Å	−2.7 Å	−2.5 Å	Total
FOM (after SOLVE)c	0.72	0.61	0.47	0.43	0.43	0.39	0.46
FOM (after RESOLVE)c	0.87	0.79	0.77	0.68	0.62	0.62	0.71

Refinement statistics
Data set	Resolution (Å)	Reflections (F > O)	Rfactor (Rfree)d	Number of atoms protein, water, ions	r.m.s.d. bonds (Å)	r.m.s.d. bond angles (°)
aRmerge = ∑hkl ∑i|Ii(hkl) − 〈I(hkl) 〉|/∑hkl ∑iIi(hkl); values in parentheses correspond to the highest resolution shell.
b As calculated with the program TRUNCATE (Bailey 1994).
c The figure of merit (FOM) is defined as the estimated cosine of the phase error.
dRfactor = ∑|Fobs − Fcalc|/∑Fobs; Rfree calculated with 9% of randomly selected data.
Native	17–1.7	63,641	0.149 (0.194)	5194, 753, 53	0.019	1.641