Automated identification of crystallographic ligands using sparse-density representations

Abstract

A novel procedure for the automatic identification of ligands in macromolecular crystallographic electron-density maps is introduced. It is based on the sparse parameterization of density clusters and the matching of the pseudo-atomic grids thus created to conformationally variant ligands using mathematical descriptors of molecular shape, size and topology. In large-scale tests on experimental data derived from the Protein Data Bank, the procedure could quickly identify the deposited ligand within the top-ranked compounds from a database of candidates. This indicates the suitability of the method for the identification of binding entities in fragment-based drug screening and in model completion in macromolecular structure determination.

Models of protein–ligand crystal structures: trust, but verify

“…In addition to automated atom typing and restraint file generation for ligands (see “Stereochemistry-based validation of protein-ligand models”), several methods are available to enable accurate ligand identification [79, 80], ligand building [81, 82], ligand placement and ligand refinement [37, 83, 84]. Such automated methods help enforce correct atom typing and correctly applied restraints and, in turn, result in better quality protein–ligand models as assessed by stereochemical validation methods.…”

The solvent component of macromolecular crystals

“…The application and limitations of X-ray crystal structure models for structureguided lead discovery (Blundell et al, 2002;Tickle et al, 2004), fragment screening (Burley, 2004;Hartshorn et al, 2005;Fischer & Hubbard, 2009) and drug design (Goodwill, 2001) have been reviewed elsewhere (Davis et al, 2008). As in the case of ordered solvent modelling outlined in x3.1.2, automated programs for ligand placement have been developed and implemented in most crystallographic structuredetermination (Oldfield, 2001;Zwart et al, 2004;Wlodek et al, 2006;Terwilliger et al, 2006Terwilliger et al, , 2007Binkowski et al, 2010;Klei et al, 2014;Echols, Moriarty et al, 2014;Carolan & Lamzin, 2014) and validation (Kleywegt, 2007;Kleywegt & Harris, 2007;Smart et al, 2011;Pozharski et al, 2013;Weichenberger et al, 2013) packages.…”

Carbohydrate 3D structure validation

“…Although general automated tools for detecting ligands are available [39,40], currently the only carbohydrate-specific one is the CCP4 Sails program (J Agirre and K Cowtan, unpublished; URL: https://fg.oisin.rc-harwell.ac.uk/projects/sails). This software relies on deposited data for generating fingerprints of sugars which are then matched to the experimental map in a fast six-dimensional search, similarly to how the NAUTILUS program builds nucleic acid [41].…”