Cryo-EM performance testing of hardware and data acquisition strategies

Microscopy (Oxf). 2021 Nov 24;70(6):487-497. doi: 10.1093/jmicro/dfab016.


The increasing popularity and adoption rate of cryo-electron microscopy (cryo-EM) is evidenced by a growing number of new microscope installations around the world. The quality and reliability of the instruments improved dramatically in recent years, but site-specific issues or unnoticed problems during installation could undermine productivity. Newcomers to the field may also have limited experience and/or low confidence in the capabilities of the equipment or their own skills. Therefore, it is recommended to perform an initial test of the complete cryo-EM workflow with an 'easy' test sample, such as apoferritin, before starting work with real and challenging samples. Analogous test experiments are also recommended for the quantification of new data acquisition approaches or imaging hardware. Here, we present the results from our initial tests of a recently installed Krios G4 electron microscope equipped with two latest generation direct electron detector cameras-Gatan K3 and Falcon 4. Three beam-image shift-based data acquisition strategies were also tested. We detail the methodology and discuss the critical parameters and steps for performance testing. The two cameras performed equally, and the single- and multi-shot per-hole acquisition schemes produced comparable results. We also evaluated the effects of environmental factors and optical flaws on data quality. Our results reaffirmed the exceptional performance of the software aberration correction in Relion in dealing with severe coma aberration. We hope that this work will help cryo-EM teams in their testing and troubleshooting of hardware and data collection approaches.

Keywords: apoferritin; cryo-EM; performance; resolution; single particle analysis.

MeSH terms

  • Cryoelectron Microscopy* / instrumentation
  • Cryoelectron Microscopy* / methods
  • Data Collection
  • Reproducibility of Results
  • Workflow