Figure 3
Crystallography data collection and processing at FMX. Raster scans of (a) a bovine trypsin rod-shaped crystal and (b) 5–10 µm-sized proteinase K crystals loaded on MiTeGen loops. The raster-scan areas are shown as red-dashed rectangles; the corresponding heat maps acquired after the raster scans are shown in the insets. The grid coloring is the diffraction pattern spot count. (c) Clustering of the serial crystallography partial datasets based on unit-cell size. The data shown were acquired from raster scans of proteinase K microcrystals with a 200 Hz detector frame rate. Of these 279 partial datasets, 234 (blue dots) were indexed to c = 106.58 ± 0.41 Å, with 99.6% completeness (Table 1). The remaining 45 partial datasets (orange dots) with c = 110.06 ± 0.84 Å did not yield a complete dataset and were excluded from refinement. The unit-cell size was converted to Cartesian coordinates for better visualization. The data were clustered into two groups using a K-mean clustering algorithm. (d) Hierarchical cluster-analysis dendrogram for the proteinase K partial datasets acquired at a 750 Hz detector frame rate. The cutoff limit was set to 0.6 (dashed line). |