SMX

Serial Millisecond Crystallography

Content:
Introduction
General Methodology
Instruction/Manual
Instruments
Data Collection
Cleaning
Data Interpretation
Reference

Introduction:

Room-temperature structure determination is of increasing importance in the elucidation of protein dynamics. As opposed to cryo-crystallography, room-temperature X-ray structures reveal realistic conformational flexibility that is crucial for protein function.

In addition, room-temperature crystallography enables time-resolved studies 9 that reveal protein dynamics and enzyme catalysis with atomic detail.

A promising solution is to adapt the methodology developed for macro-molecular crystallography at XFELs to synchrotron sources, where radiation damage cannot be outrun but where the radiation dose per crystal can be reduced by using many crystals.

The most direct approach is serial millisecond crystallography (SMX), which utilizes the same high-viscosity injectors successful at XFELs to distribute the radiation dose over thousands of crystals to determine room-temperature structures with minimum radiation-damage.

In some runs, we varied the longer side of the beam to adjust hit rate and signal to noise depending on the specific characteristics of the sample. The wider beam perpendicular to
the flow of the LCP extrusion allowed scanning a larger volume of the crystal-carrying medium. Generally, this led to increased hit rates and thus balanced sample consumption, data collection speed and signal-to-noise ratio.
The speed of extrusion, the average crystal size, the beam shape and the applied flux determine the average radiation dose per crystal.

Data were collected continuously using the EIGER 16 M detector at 100 Hz.

Comparison of data sets revealed that
scanning a crystal with higher frame rate data collection for SMX is superior to lower frame rate data collection. This was supported not only by anomalous peak heights (see Fig. 2) but also by other data statistics such as signal-to-noise ratio in the higher resolution
shells (see Supplementary Table 2). Using the anomalous signal to assess data quality, we show that the improved statistics are not owing to multiple patterns collected from a single crystal, which may bias statistics based on CC* (see Fig. 2). Thus, there are clear advantages of the latest detector generation with high frame rate and negligible dead time over slower detectors for SMX data Acquisition.

Sample optimisation for SMX is important before starting the experiment

In Serial Macromolecular Crystallography (SMX), data convergence refers to the point in data processing where adding more diffraction images (hits) no longer significantly improves the quality or completeness of the merged dataset. It’s a key concept in determining when enough data has been collected for successful structure determination.