Breaking the indexing ambiguity in serial crystallography

Area of Science:

• Structural Biology
• Crystallography
• Biophysics

Background:

• Serial crystallography yields incomplete data sets ('snapshots') from each diffraction experiment.
• Indexing ambiguity in certain space groups necessitates re-indexing snapshots against a reference dataset.
• Previous methods merged snapshots, creating twinned datasets unsuitable for structure solution.

Purpose of the Study:

• To develop algorithms for assembling complete serial crystallography datasets from incompletely indexed snapshots.
• To resolve indexing ambiguities in serial crystallography data.
• To improve data quality for structure solution and refinement.

Main Methods:

• Developed two algorithms for clustering serial crystallography snapshots based on indexing.
• Tested algorithms using 15,445 snapshots from photosystem I and noisy model data.
• Applied clustering to resolve indexing ambiguities and construct complete datasets.

Main Results:

• Algorithms successfully clustered snapshots, resolving indexing ambiguities.
• Complete datasets were constructed in the correct space group (P63).
• This overcomes the previous limitation of using twinned datasets (P6322).

Conclusions:

• The developed algorithms effectively assemble complete datasets in serial crystallography.
• Indexing ambiguity in serial crystallography can now be reliably resolved.
• These algorithms enhance the utility and scope of serial crystallography for structure determination.