Date Published: March 01, 2017
Publisher: International Union of Crystallography
Author(s): Paul Emsley.
The current tools of Coot for ligand validation and comparison in are presented. The user-selected ligand is assessed by ligand-distortion and map-correlation metrics, and compared with those of ligands of the wwPDB to create a percentile rank.
For many years, the validation of macromolecular structures has been a concern of practicing crystallographers and users of the PDB (Berman et al., 2000 ▸) (and more recently the wwPDB; Berman et al., 2003 ▸); see, for example, Brändén & Jones (1990 ▸), Dodson (1998 ▸) and Davis et al. (2007 ▸). Since 2007, crystallographic diffraction data deposition has been mandatory for structure depositions at the wwPDB sites. This, and the increase in the number of deposited structures, has enabled macromolecular model validation to be reconsidered (Read et al., 2011 ▸) and the recommendations that were made have been implemented by the wwPDB deposition sites to provide access to a concise summary of well established quality indicators.
The ligand-selection and scoring methods are described below.
The ligand-selection system often picks buffer molecules or cofactors. In the case of cofactors, these may not have been the main ligand of interest in the model.
Ligand-analysis tools have been integrated into the new version of Coot and have been used to assess the ligands of structures in the PDB. The resulting metrics can be used to assess the quality of any particular ligand under refinement or due for PDB data deposition. The combined ranks score gives a single number which combines all of the metrics for a quick assessment of a particular ligand.