xia2.ssx

xia2.ssx is an autoprocessing pipeline that uses DIALS to process still-shot data.

The xia2.ssx pipeline is triggered automatically on still-shot data collections on I24 and on VMXi grid-scans. To fully process still-shot data, a unit cell and space group need to be provided. For a VMXi grid scan, this information is taken from the sample information page if available. For I24 SSX data collections, data collection is usually performed straight after sample preparation, which means that sample information is not available via ISPyB.

Instead, to provide the unit cell and space group, a small plain text .yml file must be created for each sample in the visit directory, which is picked up by the autoprocessing infrastructure. The filename must match the sample name and be placed in the "processing" folder at:

For example, if the sample name used for the data colletion was "lyso", then the file needs to be created at:

and contain the following items:

The success of indexing still-shot data is heavily dependent on having a good detector geometry - therefore either a reference geometry can be provided as an option in the yml file, or the pipeline will refine the detector geometry as its first step, before indexing and integrating the full collection.

One workflow is to determine a good reference geometry near the start of the experiment through a xia2.ssx job on a well-diffracting sample (it is the file geometry_refinement/refined.expt in a xia2.ssx processing job) and then this can be used as a reference geometry for future processing.

For this reference geometry to be used in autoprocessing, the file must be copied to:

and renamed to:

During indexing, up to 3 crystal lattices are searched for on each image. During integration, an enhanced algorithm for still-shot data is used to model anisotropic spot shapes and account for offsets between the reflections and the diffracting condition that occur for still-shot data. After integration, the dials.cosym program is used to resolve indexing ambiguities, if these are allowed by the space group symmetry or cell parameters. Finally the data are scaled and merged.

A summary view of the main processing statistics is displayed in ISPyB. The "Logs & Files" button displays a list of processing logfiles and datafiles for the individual processing steps. The main xia2.ssx log is the "xia2.ssx.log" file. A merging HTML report, named dials.merge.html is generated, which contains a more extensive set of data quality statistics and plots. The main data output datafile from xia2.ssx is the "merged.mtz" file, which is cut to the approximate resolution limit. A separate merged MTZ file, "merged_full.mtz", contains data to the full resolution limit of the detector.

If your publication uses data processed with the xia2.ssx pipeline, please cite the following reference:

• Beilsten-Edmands, James, et al. "Processing serial synchrotron crystallography diffraction data with DIALS." Methods in Enzymology. Vol. 709. Academic Press, 2024. 207-244.

If you deposit a structure in the PDB which is determined from data processed with the xia2.ssx pipeline, please make sure both xia2 and DIALS are included in the software category.

The documentation for xia2.ssx can be found here: https://xia2.github.io/serial_crystallography.html