Towards automated Laue data processing: application to the choice of optimal X-ray spectrum

Laue data reduction has now reached a level of sophistication that allows nearly automated processing to be performed. The software described enables complete reduction of the data with essentially no user intervention, making Laue processing almost as straightforward as monochromatic data processing. Interactive work is limited to the indexing of only one Laue pattern. More importantly, it is shown that the data quality is substantially enhanced when soft-limited predictions are used. Further improvement obtained by taking advantage of the structure-factor amplitudes from a known closely related structure is described. To determine the most suitable type of insertion device to be used for time-resolved Laue crystallography, the technique described was applied to Laue data sets collected from photoactive yellow protein under identical conditions but with three different insertion devices: a wiggler, a broad-bandpass undulator and a single-line undulator. Although the optimal choice may ultimately be dictated by sample parameters (such as mosaic spread) and by the type of experiment (repeatable or non-repeatable reactions), the results here show that the use of single-line undulators will generally yield by far the best compromise between data quality, acquisition time and radiation damage.