Structure-based drug design heavily depends on the availability of high quality protein structures mostly derived by X-ray crystallography. The task of the crystallographer usually ends with the creation of a protein model fitted to the experimental electron density; the task of the modeler starts with using this model in various kinds of calculations like homology modeling, lead identification by molecular docking and scoring techniques. Besides some general quality parameters like the structure resolution, the detailed information from electron density remains mostly unconsidered.<br/>In this talk, we present a novel approach to estimate the electron density support for individual atoms in protein crystal structures named EDIA. For each atom, the quality of fit is calculated under consideration of the molecular structure the atom belongs to. EDIA values can then be visualized to get a quick glance at the structural quality on an atomistic level. Compared to other existing methods like RSR or RSCC, EDIA results in an absolute, easy to interpret value per atom. This data is highly valuable in the molecular modeling process. First of all, reliable and less reliable atom locations can be easily separated. For ligands, alternative binding modes might get into sight. For side-chains, potentially flexible regions can be identified. Both can then be taken into account upon docking, pharmacophore modeling or lead optimization to name just a few applications.<br/>EDIA values are furthermore of interest to crystallographers and developers of novel computational methods. For crystallographers, they allow to easily spot regions in the structure demanding special attention. For method developers, they allow to create large benchmark data collections of high quality in a fully automated fashion.

Dealing with x-ray data in structure-based design: Estimating and visualizing electron density support

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.