Automatic crystal centring procedure at the SSRF macromolecular crystallography beamline

Area of Science:

• Crystallography
• Structural Biology
• Biophysics

Background:

• X-ray diffraction is crucial for determining protein crystal structures.
• High-throughput protein crystallography requires automated beamline procedures.
• Accurate crystal centring is essential for efficient automated data collection.

Purpose of the Study:

• To develop and report a rapid, automated crystal centring procedure for synchrotron beamlines.
• To address challenges in crystal recognition caused by illumination, size, and viewing angles.
• To enhance the efficiency of high-throughput protein crystallography.

Main Methods:

• Utilized a multi-camera system with region segment thresholding for crystal centring.
• Implemented an illumination-invariant loop recognition and classification model using image segmentation.
• Employed an edge detection algorithm for crystal sample identification in high-magnification images.

Main Results:

• The crystal centring method demonstrated high success rates under fluctuating light conditions.
• The procedure proved effective for poorly frozen and opaque crystal samples.
• Successfully integrated into the Blu-Ice data collection system at the Shanghai Synchrotron Radiation Facility.

Conclusions:

• The developed multi-camera, region-segmentation-based method provides robust and efficient automated crystal centring.
• This advancement significantly contributes to the automation of high-throughput protein crystallography.
• The method enhances the reliability and speed of crystal screening at synchrotron facilities.