3D ED/MicroED data from Proteinase K lamellae

This deposition contains six 3D ED/MicroED datasets in MRC file format. The data were collected as part of a workshop held at the New York Structural Biology Center in November 2025. There are sweeps of data from each of six different lamella of proteinase K crystals. Data collection Data were acquired using a TFS Glacios TEM operated at 200 keV on a Ceta-D detector, at cryogenic temperature (<100 K). The TEM was operated in microprobe mode, with spotsize 11, a 150 µM C2 aperture, and 100 µM SA aperture. The beam size was ~ 10 µm in diameter. This configuration reduced affects of charging on the lamella that can cause smearing of the diffraction spots. All data were acquired using a stage rotation rate of 0.25°/second with 1 second integrated exposures using TFS EPU-D software. The camera was binned by 2 with a nominal camera length of 1500 m, corresponding to a calibrated length of 1488 m. Data collection parameters: Data set name Tilt range Frames Total Dose (e-/Å2) lam01 80° 320 1.6 lam04 80° 320 1.6 lam07 60° 240 1.2 lam09 80° 320 1.6 lam11* 80° 320 1.6 lam12 80° 320 1.6 *lam11 contains multiple lattices. Data reduction We processed these data using DIALS. Data from some single crystals, or combinations of multiple, are sufficient for determining the structure of Tritirachium album Proteinase K to at least 1.7 Å resolution, phased using molecular replacement. The following tips may be helpful for DIALS data processing directly from the EPU-D generated .mrc files provided: The files include an incorrect rotation axis orientation. To correct this, include the “goniometer.axis=1,0,0” flag during import. Some pixels have negative values, which can negatively affect data integration. A flat value is added to remove negative pixel values using the “panel.pedestal=-17” flag during import. The pedestal value (here, -17) was determined by using the dials.show command with the “show_raw=True” flag and finding the smallest value in first quantile of pixel values for all frames. This should be found for each data set, but the –17 value works fine for all six rotation series here. We found the following import command allows for successful processing of the data, without other instrument specific modifications: dials.import panel.pedestal=-17 fast_slow_beam_centre=1015,1024 goniometer.axis=1,0,0 beam.probe=electron Sample preparation Proteinase K powder was purchased from Sigma (Cat. No. P2308). Crystals were obtained by dissolving 1 mg of Proteinase K powder with 200 µL of 1.5 M Ammonium Sulfate and vortexing for ~1 min at room temp. Grids were prepared using a Leica EM-GP2. Crystal slurry was applied to glow discharged Quantifoil Cu 200 mesh R2/2 grids, and back-blotted for 20 s using a chamber humidity of 90% and temperature of 4°C. Lamellae were milled on a TFS Arctis cryo-FIB-SEM using Xe plasma. Milling was performed by an automated workflow using the Arctis WebUI. A pre-mill coating was applied using a Pt sputter, 1 µm of GIS, followed by second Pt sputter. The final polishing step was performed at 30 pA. Lamella width was chosen based on the crystal size with a target thickness of 200 nm.