A multi-crystal method for extracting obscured crystallographic states from conventionally uninterpretable electron density

• DOI: 10.1038/ncomms15123
• Publication Date: 2017-04-24T09:56:38Z
• Authors: Nicholas M. Pearce; Tobias Krojer; Anthony R. Bradley; Patrick Collins; Radosław P. Nowak; Romain Talon; Brian D. Marsden; Sebastian Kelm; Jiye Shi; Charlotte M. Deane; Frank von Delft
• Keywords: Crystal (programming language)

Abstract In macromolecular crystallography, the rigorous detection of changed states (for example, ligand binding) is difficult unless signal strong. Ambiguous (‘weak’ or ‘noisy’) density experimentally common, since molecular are generally only fractionally present in crystal. Existing methodologies focus on generating maximally accurate maps whereby minor become discernible; practice, such map interpretation disappointingly subjective, time-consuming and methodologically unsound. Here we report PanDDA method, which automatically reveals clear electron for state—even from inaccurate maps—by subtracting a proportion confounding ‘ground state’; objectively identified statistical analysis distributions. The method completely general, implying new best practice all changed-state studies, including routine collection multiple ground-state crystals. More generally, these results demonstrate: incompleteness atomic models; that single data sets contain insufficient information to model them fully; accuracy requires further map-deconvolution approaches.