Advances in long-wavelength native phasing at X-ray free-electron lasers

Abstract

Long-wavelength pulses from the Swiss X-ray free-electron laser (XFEL) have been used for de novo protein structure determination by native single-wavelength anomalous diffraction (native-SAD) phasing of serial femtosecond crystallography (SFX) data. In this work, sensitive anomalous data-quality indicators and model proteins were used to quantify improvements in native-SAD at XFELs such as utilization of longer wavelengths, careful experimental geometry optimization, and better post-refinement and partiality correction. Compared with studies using shorter wavelengths at other XFELs and older software versions, up to one order of magnitude reduction in the required number of indexed images for native-SAD was achieved, hence lowering sample consumption and beam-time requirements significantly. Improved data quality and higher anomalous signal facilitate so-far underutilized de novo structure determination of challenging proteins at XFELs. Improvements presented in this work can be used in other types of SFX experiments that require accurate measurements of weak signals, for example time-resolved studies.

Keywords: X-ray free-electron lasers; anomalous data-quality indicators; de novo protein structure determination; serial femtosecond crystallography; single-wavelength anomalous diffraction.

Figure 1

Anomalous data-quality indicators. (a) Anomalous signal strength (S _ano) of thaumatin 4.57 and 6.06 keV data sets for different numbers of indexed diffraction images. The horizontal line indicates an S _ano value of 10, below which SAD phasing is difficult. The two vertical lines indicate the minimal numbers of indexed images required for structure determination using native-SAD in this study. (b) Correlation coefficient between the measured and calculated anomalous difference structure-factor amplitudes (CC_ano ^{model versus data}) for thaumatin data sets with minimal amount of data necessary for successful structure solution using native-SAD at the SwissFEL and the Swiss Light Source (SLS) (Leonarski et al., 2018 ▸).

Figure 2

Detector-distance optimization using thaumatin crystals. (a) The smallest standard deviation of the distribution of the unit-cell parameters (green triangles) indicates the optimal sample-to-detector distance 94.9 mm. The indexing rate (blue squares) is highest at the optimal detector distance. (b) Average peak height of the anomalous Fourier difference map (S _ano) from thaumatin at 4.57 keV (pink circles) calculated at different detector distances. S _ano is highest at the optimal detector distance which corresponds to the smallest standard deviation of c axis length (green triangles).

Figure 3

Comparison of the anomalous signal strength (S _ano) and anomalous correlation coefficient (CC_ano) for thaumatin at 4.57 keV. (a) Data sets with different numbers of images were processed using old and new versions of partialator from the CrystFEL software suite, with and without post-refinement and partiality correction. S _ano is significantly larger when the newer version of post-refinement and partiality correction is used (green circles), as compared with the older version (purple triangles) and the newer version with only scaling (orange squares). (b) Similarly, CC_ano from 50 000 randomly selected thaumatin 4.57 keV diffraction images is higher when the newer version of CrystFEL with post-refinement and partiality correction is used on the same data set (green circles).

Figure 4

Phasing and automatic model building. (a) The experimental electron-density map (contoured at 1.0σ) after phasing and density modification obtained from the A_2A data set with a minimal number of indexed images (50 000) at 4.57 keV superposed with the A_2A molecule after automatic model building and refinement. (b) The 2mF _o − DF _c electron-density map after automatic model building and refinement. The MapCC values indicate the correlation coefficient between the map shown in the figure and the final refined map.