CrystalCMP: an easy-to-use tool for fast comparison of molecular packing

Abstract

A new approach is introduced for the comparison of molecular packing and the identification of identical crystal structure motifs. It has been tested on data sets for the solid forms of benzamide, cabergoline and trospium. In this approach, the packing similarity is calculated using a simple formula involving the distances between molecular centres and the relative orientations of molecular entities inside a finite molecular cluster. The approach is independent of the atomic labelling, the unit‐cell parameters, the space group setting and the number of molecules in the asymmetric part of the unit cell. Owing to its low sensitivity to volume changes, this approach allows the comparison of various solid forms (such as polymorphs, hydrates, solvates, co‐crystals or salts) of identical or similar molecular compounds. The method is also suitable for identifying similar results from direct space methods, which are often used in powder diffraction.

Dihalomethanes as Bent Bifunctional XB/XB‐Donating Building Blocks for Construction of Metal‐involving Halogen Bonded Hexagons

“…The measure of the extent with whicht wo crystal structures deviate from the perfecto verlapping is defineda st he "packing similarity" (PS). [12] The value of PS, in general,i sn ot limited, but in accord with Ref. [12] the crystalsa re almost identicali fP S < 5.…”

CrystalCMP: automatic comparison of molecular structures

“…A consequence of such a weighting is the possibility to identify similar molecular packing in two crystal structures, one of which is expanded compared with the other (either because they have been measured at different temperatures or because they crystallize with different solvent molecules of various sizes). This was confirmed in previous work on several selected data sets [see Rohlíček et al (2016)]. However, this method required user interaction and, as a result, the software could not be used to automatically compare large quantities of data.…”

“…Detection and evaluation of the precision of the molecular symmetry and using it for calculation of the angle of rotation, as was described previously (Rohlíček et al, 2016), was found to be problematic and unsuitable for automation. For that reason, the angle of rotation was replaced by the calculation of root-mean-square deviation (RMSD) of atomic coordinates after both concerned molecules are shifted to the same origin represented by their centroids (see Fig.…”

“…In the case of molecular crystals, the similarity of crystal structures can be studied via the packing similarity of molecules in a molecular cluster. There are already three approaches represented by the COMPACK (Chisholm & Motherwell, 2005), xPac (Gelbrich & Hursthouse, 2005;Gelbrich et al, 2012) and CrystalCMP (Rohlíček et al, 2016) codes. Working with a molecular cluster instead of atoms in the unit cell brings major benefits for organic crystals.…”

“…CrystalCMP's method (Rohlíček et al, 2016) is based on the comparison of representative molecular clusters, in which only one type of molecule is included. During the comparison process, the generated molecular clusters of the individual crystal structures are overlapped according to their central molecules, and the similarity is calculated as the deviation of the centroids of the overlapping molecules and the angles of rotation of the molecules from each other.…”