Abstract

Raw diffraction intensities from an integration program are not usually on the same scale, due to a number of physical factors affecting the experiment. The process of data scaling, such as that done in the program AIMLESS or SCALA , attempts to remove the systematic differences from the intensities by modelling the experiment, putting them all on the same scale, or at least a self-consistent scale. Scaling relies on replicate measurements related by crystal symmetry, so it is important before scaling to determine the point-group symmetry of the diffraction pattern, and it is useful to choose a likely space group at this stage. The likely point-group symmetry (the Laue group plus any lattice centring) is determined by the program POINTLESS which compares reflections potentially related by symmetry to see if they agree, and ranks the possible Laue groups by scoring their agreement. It is often also possible to determine the space group by analysis of axial systematic absences. Scaling tries to make symmetry-related observations equal, but analysis of the residual differences after scaling provides a great deal of information about the quality of the data, both globally and locally. The intensities can be analysed to set the real resolution of the dataset, to detect bad regions (e.g. bad images), to analyse radiation damage and to assess the overall quality of the dataset. The signiï¬cance of any anomalous signal may be assessed by probability and correlation analysis. Analysis of distributions of the averaged intensities can detect crystal pathologies such as twinning and non-crystallographic translations

This lecture will present some examples of good and bad data and give some guidance on what to look out for and on the decisions that need to be made.