The democratization of cryo-EM

To the Editor:

The cryo-electron microscopy (cryo-EM) community is in the throes of a ‘resolution revolution’¹. Years of technical development have led to hardware and software that can provide refined atomic structures with resolutions adequate for obtaining novel biological insights and for structure-based drug design^2,3. However, these advances come at the cost of a multi-million-dollar investment that is required to establish a cutting-edge microscope, a detector and computational support. We believe that there is now the opportunity, and need, to provide democratic access to maximize the impact of cryo-EM on basic and applied science.

Possible models for such access are already emerging, following lessons learned from the experience accumulated over the past two decades, especially at synchrotron facilities across the world⁴. Crystallography beamlines are excellent examples of how the organization of large facilities, pioneered in physics, can shape the landscape in biological and biomedical research. Cryo-EM can learn from this experience. Several good initial exemplars, including the eBIC center in the United Kingdom (http://www.diamond.ac.uk/Science/Integrated-facilities/eBIC.html), the NeCEN center in the Netherlands (http://www.necen.nl), the Janelia Farm campus of the Howard Hughes Medical Institute (https://www.janelia.org/support-team/cryo-electron-microscopy) and the newly created national cryo-EM facility at the US National Cancer Institute (http://www.cancer.gov/research/resources/cryoem), are enabling users to access high-end cryo-EM infrastructure.

Such centralized facilities allow a broad spectrum of users to focus on fundamental biological questions with the knowledge that excellent infrastructure will be available. Interactions between the technical staff at the facilities and users with academic and commercial backgrounds can help optimize pipelines and establish a ‘team culture’, as at synchrotrons where industry groups bring a keen eye for efficiency. Other models for access are those driven by commercial consortia, such as the Cryo-EM Consortium in Cambridge, UK (http://www2.mrc-lmb.cam.ac.uk/cambridge-pharmaceuticalcryo-em-consortium/). Another model enabling the collectivization of research and partnering with industry is found in cooperative research centers⁵, which have helped foster scientific excellence and shaped regional competitiveness in, for example, the Basque region (https://issuu.com/innobasque/docs/oecd_assessment__recommendations).

Cryo-EM is rapidly maturing from a technique limited to a relatively small circle of expert users to one of very broad interest. Waiting times to gain access to equipment are long in most places, and the establishment of additional cryo-EM hubs would drive scientific synergy, training and technological development with academic and industrial sectors, stimulating both science and the economy.

Open access to synchrotron facilities enabled a step change that led X-ray crystallography to permeate the global life science community and become an indispensable part of the drug discovery process. Establishing a free market between synchrotrons drove technical developments and increased efficiency, leading to improvement by orders of magnitude in data-acquisition speed, data quality, automation and remote access. Such changes also bring challenges, such as the emergence of a generation of scientists who are less experienced users, but we suggest that this is natural in the evolution of any new technology. The evolution of the crystallography field taught the importance of validation, and the cryo-EM field is building on these lessons to provide robust validation tools and metrics⁶.

To what extent should cryo-EM be centralized in dedicated synchrotron-like facilities? Would the field be better served if resources were distributed to individual institutions? We suggest that it is not necessary to choose between these options, as both are important for the progress of the field. Investing in central facilities that are efficient, are responsive and drive new technology development will help large sectors of the user community, whereas local institutional investments could inspire interdisciplinary collaborations within and across neighboring institutions. Synchrotrons are far more expensive than individual electron microscopes, but the overall scale of investment in cryo-EM might need to be similar in order to enable the recent rapid advances to have a lasting impact. Is this affordable? National funding agencies in Europe and in the United States are indeed beginning to step up, and transnational access for various biophysical methods is now being supported by programs such as Instruct (http://www.structuralbiology.eu) and iNEXT (http://www.inext-eu.org/), building on a tradition of support for synchrotron access (e.g., http://www.biostruct-x.eu/).

More uniform distribution of cryo-EM technology in areas where there is a dearth of microscopes such as southern Europe, the central United States and many Asian countries will also be necessary to drive the field forward. Deciding where new centers should be based, how they should be funded and how open access can be ensured will undoubtedly require careful scientific and economic assessment, considering criteria such as existing electron microscopy service culture and support from in-house research groups. New paradigms may also be possible—a connected international network of cryo-EM centers might lower barriers to broad adoption of the technology. Whichever portfolio of models is pursued, it will be necessary to democratize the technology in order to release the enormous potential of the cryo-EM revolution.