In the field of chemical crystallography, a lot of work with workflow automation is already done. Beginning with the diffractometer control, it is quite possible to have the cell refined, data integrated and corrected, the space group determined, the phase problem solved and the structure somewhat refined almost without the touch of the human hand nowadays. These changes, driven initially by the need of saving time while working on the non-complicated cases for use when something will go difficult. However, these advances often lead to false assumptions that computer will do anything & right (so why waste the time with even basic crystallographic training) and that chemical crystallography is „routine“ task.
During the work on the copper complexes [1], we came into several cases of pseudocentrosymmetricity caused by short side chain of α-amino acids used for the syntheses. We demonstrate, that the E statistics should mislead the software – and subsequently inadequately trained crystallographer – into assumption, that the structure is centrosymmetric [2], although it can be easily prevented by using basic knowledge from group theory and by fair use of the software [3].
Validation tool checkcif [4] was implemented to warn authors and/or reviewers about errors and inconsistencies. It was somewhat supposed, that in the situation when there is a shortage of trained crystallographers for refereeing vast numbers of the structures solved every day, such tool should also facilitated use of common chemists with some crystallographic knowledge as referees. However, it can also lead to rejection of the paper by less skilled referees based solely on presence of some A alerts or can mislead the authors. Although it is far more common to find the structure with overlooked symmetry, e.g. in the Dick Marsh’s works [5], in the pseudocentrosymmetric cases less skilled crystallographers can be easily confused by ADDSYM alerts with sometimes pretty high atom match percentage.
In such cases, the non-centrosymmetric structure solved in the centrosymmetric space group usually mimic a disordered structure. It can be still possible to detect the problem by careful inspection of bond lengths and angles and their respective s.u.’s or ADPs.
2. A. Tovar-Tovar, J.–C. García-Ramos, M. Flores-Alamo, L. Ruiz-Azuara, Acta Crystallogr. Sect. C Struct. Rep., 67, (2011), m1796–m1797.
3. Rigaku Oxford Diffraction, (2017), CrysAlisPro Software system, version 1.171.38.46, Rigaku Corporation, Oxford, UK.
4. A. L. Spek, Acta Crystallogr. Sect. D Biol. Crystallogr., 65, (2009), 148–155.
5. L. M. Henling, R. E. Marsh, Acta Crystallogr. Sect. C Struct. Chem., 70, (2014), 834–836.
This work has been supported by the Ministry of Education of Slovak Republic (Grant VEGA 1/1507/17).