Table 1. Crystallographic structures cited in this article.

PDB code; name; organism; reference	Crystallization details	Resolution (Å); space group; No. of protein molecules in AU†; R, Rfree (%)	The highest ± difference Fourier (Fo − Fc) electron-density peak (viewed in Coot) and any specific comments therefrom†	PDB validation assessment (clashscore; specific comments of interest based on the PDB report‡)
1dpg; G6PD; L. mesenteroides; Rowland et al. (1994 ▸)	Not in PDB file. See Rowland et al. (1994 ▸) for details; also Adams et al. (1983 ▸).	2.0 Å; P3221; chains A and B; 20.6, 25.7	No reflection data are available	8; Ramachandran outliers 0%; side-chain outliers 7.1%
1qki†§; G6PD (variant Canton R459L) complexed with structural NADP+; H. sapiens; Au et al. (2000 ▸)	Hanging-drop vapour diffusion. 1 + 1 µl drops in the well. 0.1 M sodium citrate, 0.05 M glycolic acid pH 5.8. Protein concentration 10 mg ml−1.	3 Å; P212121; 8 chains; 2 tetramers; 24.7, 29.4	49 peaks above 5σ; top peak 7.4σ; not clear how to model peaks 1–9 (maybe Fourier series-termination effects); of the top 10 peaks, peak 10 (6.1σ) is likely to be a glycerol	11; Ramachandran outliers 2.3%; side-chain outliers 7.4%; RSRZ 1.3%. 〈I/σ(I)〉 = 1.21 at 2.99 Å, i.e. these are relatively weak intensity data defining the ‘edge’ of the X-ray diffraction pattern; there are 20 amino acids at the N-termini which are not modelled; the eight NADP molecules fit their electron-density omit maps well. PDB-REDO, according to its bound water validation criteria, removed about 75% of the depositors’ bound waters.
2bh9; deletion variant of G6PD complexed with structural and coenzyme NADP+; H. sapiens; Kotaka et al. (2005 ▸)	Hanging-drop vapour diffusion. 1 + 1 µl drops in the well. 0.1 M Tris–HCl pH 7.5–8.2, 10–16% PEG 4000. Protein concentration 5 mg ml−1.	2.50 Å; F222; chain A; 19.6, 29.6	No reflection data are available	27; Ramachandran outliers 0.6%; side-chain outliers 14.5%. The Rfree/R factor gap of 10% is large and indicative of overfitting of the model.
2bhl†¶; G6PD (deletion variant) complexed with glucose 6-phosphate; H. sapiens; Kotaka et al. (2005 ▸)	0.1 M Tris pH 8.5, 0.2 M MgCl2, 12% PEG 4000, 5% glycerol pH 8.50	2.9 Å; C2221; chains A and B; 21.2, 26.1	30 peaks above 5σ; top peak 7.6σ; first ten peaks possibly bound waters; peak 11 (6.0σ) possibly a side-chain adjustment at Asn414	30; Ramachandran outliers 1.5%; side-chain outliers 13.7%; RSRZ 4.5%. Electron density on each G6P looks good.
7sei¶; G6PD (K403Q); H. sapiens; Garcia et al. (2022 ▸)††	PEG 8K, vapor diffusion, 295 K	3.65 Å; P41212; chain A; 19.1, 23.5	23 peaks above 5σ; top peak 12.4σ, which is a lengthy unmodelled ‘blob’. 28 unmodelled blobs were found in Coot.	12; Ramachandran outliers 3.1%; side-chain outliers 7.8%; RSRZ 0%. 17% of the residues were present in the sample but not in the PDB model.
6e08¶; G6PD in complex with structural NADP+; H. sapiens; Au et al. (2000 ▸)‡‡	20%(w/v) PEG 3350, 0.2 M potassium formate pH 7.3, vapor diffusion, sitting drop, 293 K	1.9 Å; F222; chain L; 17.2, 21.3	19 peaks above 5σ; top peak 7.45σ which looks like a split occupancy possibility on the Cys232 side chain; peak 2 at 6.05σ is possibly a bound water	3; Ramachandran outliers 0%; side-chain outliers 1.2%; RSRZ 1.9%
6d23†¶; G6PD (apo form); T. cruzi; Ortíz et al. (2019 ▸)	4 µl 20 mg ml−1 protein (in 20 mM Tris pH 8.0, 50 mM NaCl, 0.5 mM MgCl2) + 4 µl 6% PEG 400, 1.6 M ammonium sulfate, 0.1 M HEPES pH 7.5, vapor diffusion, hanging drop, 293 K	2.85 Å; P21; chains A, B, C and D; 20.4, 24.9	17 peaks above 5σ; top peak 7σ; not clear how to model top three peaks above 6σ	5; Ramachandran outliers 0.3%; side-chain outliers 6.4%; RSRZ 4.8%
6d24†¶; G6PD in complex with G6P; T. cruzi; Ortíz et al. (2019 ▸)	4 µl 33 mg ml−1 protein (in 20 mM Tris pH 8.0, 50 mM NaCl, 0.5 mM MgCl2) + 4 µl 4% PEG 400, 1.8 M ammonium sulfate, 0.1 M HEPES buffer, 5 mM G6P pH 7.5, vapor diffusion, hanging drop, 293 K	3.35 Å; P21; chains A, B, C and D; 20.5, 25.4	29 peaks above 5σ; top 12 peaks all negative with top two peaks being −11.4σ and −10.2σ; all 12 peaks appear to be misassigned as sulfates or incorrect occupancy estimates	6; Ramachandran outliers 0.9%; side-chain outliers 8.9%; RSRZ 2.6%
5aq1†¶; G6PD–G6P–NADPH ternary complex; T. cruzi; Mercaldi et al. (2016 ▸)	2 µl 10 mg ml−1 protein, 5 mM G6P, 2 mM NADPH in 20 mM Tris–HCl pH 8.0, 0.2 M NaCl, 5 mM ME. Precipitant: 1 µl 30% Jeffamine ED-2003 pH 7.0, 0.1 M HEPES pH 7.0; vapor diffusion, sitting drop, 293 K.	2.65 Å; I4122; chains A, B and C; 20.0, 22.6	58 peaks above 5σ; top three peaks occur at Tyr443A/B/C (10.1σ, 8.6σ and 8.6σ) each accompanied by a planar difference electron density at 5.3σ down to 3.8σ. Tyr443 is remote from the NADP or G6P binding sites (∼20 Å) and is not mentioned in the publication.	1; Ramachandran outliers 0%; side-chain outliers 2.8%; RSRZ 2.1%. The NADP and G6P molecules fit their electron-density omit maps well. PDB report: ‘The analyses of the Patterson function revealed a significant off-origin peak that is 61.85% of the origin peak, indicating pseudo-translational symmetry’.
7zhv†¶; G6PD complexed with glucose 6-phosphate; L. donovani; Berneburg, Rahlfs et al. (2022 ▸)§§	12% PEG 3000, 200 mM ammonium sulfate, vapor diffusion, sitting drop, 283 K	3.3 Å; C2; chains A and B; 25.0, 30.9	7 peaks above 5σ; top peak −6.8σ but the problem in the model that it is signifying is unclear	11; Ramachandran outliers 0%; side-chain outliers 1.4%. 12% of the residues are present in the sample but not in the PDB model.
7zht†¶; G6PD apo form; L. donovani; Berneburg, Rahlfs et al. (2022 ▸)	18% PEG 3000, 200 mM ammonium chloride, vapor diffusion, sitting drop, 283 K	2.8 Å; P2; chains A, B, C and D; 30.4, 24.4	35 peaks above 5σ; top peak −9.0σ; not clear how to model the top peaks 1–3 (may be Fourier series-termination effects)	17; Ramachandran outliers 0.2%; side-chain outliers 6.4%; RSRZ 4.7%. 6%, 7%, 14% and 14% of chains A, B, C and D, respectively, are unmodelled. 〈I/σ(I)〉 = 1.25 at 2.81 Å, i.e. these are relatively weak intensity data defining the ‘edge’ of the X-ray diffraction pattern. PDB report: ‘The analyses of the Patterson function reveals a significant off-origin peak that is 29.63% of the origin peak, indicating pseudo-translational symmetry’.
3b4y†¶; FGD complexed with F420 and citrate; M. tuberculosis; Bashiri et al. (2008 ▸)	1.4 M trisodium citrate pH 6.5, vapor diffusion, sitting drop, 291 K	1.95 Å; P21212; chains A and B; 23.4, 19.2	32 peaks above 5σ; the top two peaks of 10.3σ and 8.1σ are interestingly shaped unmodelled ‘blobs’ as defined by Coot	4; Ramachandran outliers 0; side-chain outliers 1.8%; RSRZ 6.0%; 〈I/σ(I)〉 = 1.4 at 1.95 Å, i.e. these are relatively weak intensity data defining the ‘edge’ of the X-ray diffraction pattern
5lxe†¶; F420-dependent glucose-6-phosphate dehydrogenase from R. jostii RHA1; Nguyen et al. (2017 ▸)	Ammonium sulfate, sodium acetate pH 4.6, PEG 4000, vapor diffusion, sitting drop, 293.15 K	1.45 Å; P212121; chains A and B; 18.5, 15.6	48 peaks above 5σ; top two peaks 14.7σ and 13.8σ; these 48 peaks mainly look like waters but some are unmodelled ‘blobs’ and a few side chains could be remodelled	2; Ramachandran outliers 0; side-chain outliers 1.8%; RSRZ 1.9%

^†Entries with more than one subunit in the asymmetric unit (AU) had their space groups checked and confirmed using Zanuda (Lebedev & Isupov, 2014 ▸).

^‡The validation report from the PDB concerns the derived model and not unmodelled peaks. The F_o − F_c map was inspected in the Coot visualization system to describe the unmodelled peaks (Emsley et al., 2010 ▸).

^§The difference-map evaluation in this table used the PDB-REDO entry for PDB entry 1qki as the deposited cif file of structure factors did not have map coefficients. The PDB-REDO file has the eight chains displayed in the Draw Sequence View tool in Coot (Emsley et al., 2010 ▸), whereas PDB entry 1qki shows chains A, B, C and D even though the PDB file contains all eight chains and Coot displays them, i.e. chains A–H. This is presumably a format-interconversion problem.

^¶The difference-map evaluation in this table used the PDB-REDO entry.

^††Primary citation of related structure 7seh.

^‡‡Primary citation of related structure 6e07.

^§§Primary citation of related structures 7zht, 7zhu, 7zhw, 7zhx, 7zhy and 7zhz.