Indexing amyloid peptide diffraction from serial femtosecond crystallography: new algorithms for sparse patterns

Area of Science:

• Crystallography
• Structural Biology
• Biophysics

Background:

• Indexing still diffraction patterns from peptide nanocrystals with small unit cells is difficult due to limited spots and lack of orientation information.
• Conventional indexing methods struggle with sparse diffraction data, hindering structural determination of challenging biological samples.

Purpose of the Study:

• To develop and validate new indexing algorithms for analyzing sparse diffraction data from nanocrystals.
• To enable accurate crystal orientation determination from limited diffraction data, particularly for amyloid peptides.

Main Methods:

• Development of novel indexing algorithms within the Computational Crystallography Toolbox (cctbx).
• Utilizing aggregate data from thousands of diffraction patterns to derive unit-cell information.
• Applying algorithms to determine crystal orientation matrices for individual images using as few as five reflections.

Main Results:

• Accurate unit-cell information was derived from aggregate diffraction data.
• Crystal orientation matrices were successfully determined for individual images with minimal reflections.
• Demonstrated proof-of-concept by integrating X-ray free-electron laser (XFEL) data to 2.5 Å resolution for an amyloid peptide segment (GNNQQNY).

Conclusions:

• The new cctbx algorithms effectively overcome challenges in indexing sparse diffraction patterns from peptide nanocrystals.
• These methods are broadly applicable to various sparse diffraction datasets, including low-resolution virus structures and high-throughput screening.
• Successful application to amyloid peptide structure determination highlights the potential of these advanced crystallographic techniques.