The NIH Common Fund’s Transformative High Resolution Cryo-Electron Microscopy (cryo-EM) program is requesting input from the community in identifying challenges in screening samples for high resolution cryo-EM, an essential step required for data collection at the National Centers for Cryo-EM. We welcome responses from researchers interested in cryo-EM, including those who have not previously worked in the field.
For more details, see the Request for Information (RFI): Screening for High Resolution Cryo-Electron Microscopy (NOT-RM-21-012). Please respond to the email address in the RFI by February 26.
Hi Paula,
In my opinion, a screening cryo-TEM is not needed at a national cryo-EM center. The modern cryo-TEMs take 12 or more samples, which should be enough. Before coming to a national center, investigators must have done screening “at home”, so at least some of the samples should be OK. I think a cryo-TEM specifically for screening is not cost-effective at a national center. However, some TEMs intended for screening are capable of good results by themselves, and in some cases time-pressure could be taken off the high-end cryo-TEM(s).
I would imagine that data collection at national centers should follow what has been done for X-ray crystallography synchrotrons. There are very few home source X-ray generators for screening crystals compared to 20 years ago. The current scheme in crystallography is to prepare lots of crystals, don’t screen, send them to the synchrotron, screen for diffracting crystals, collect data, and solve structure. The data collection and structure solution for cryo-EM is now routine business, and I don’t see why the earlier steps can not mirror what has been done for crystallography. Improve the current autoloader to accept 50-100 grids rather than a dozen. User prepares lots of grids, sends them to national center for clipping, loading, screening, and imaging. The data collection software can be improved to automatically screen through the grids, search for good ice, identify particles, and collect data. I realize that loading a grid and imaging an atlas may take more time than shooting photons at a crystal for a diffraction pattern, but 20 years ago we used to expose crystals to x-rays for 10 min at a home source to attain diffraction. Now days it’s less than 1 sec at the synchrotron, so technology can expedite the process. In my opinion, resources should be spent on expanding the autoloader to handle more grids, and developing the software to make the microscope operate more intelligently.