Collaborative Computational Project for cryo electron microscopy (CCP-EM): expanding the reach of cryoEM

The Collaborative Computational Project for cryo electron microscopy (CCP-EM) is a wide and diverse partnership of researchers in biological cryoEM supporting both software developers and end users to accelerate the visualisation of the molecules of life at near atomic resolution. It collates efforts to standardise and improve computational processing of experimental data, covering both the provision of software and the development of appropriate skills in a growing community.
Since 2013 there has been an exponential increase in the rate of structures solved by cryoEM, and it will soon become the dominant experimental technique for solving biomolecule structures. The reach of applications within the field is rapidly widening with high resolution studies of ligand binding becoming more common, as well as lower resolution tomography now connecting with optical microscopy and volumeEM of cells. There is a pressing need for CCP-EM to evolve, and in this proposal we lay out a programme of work to achieve this. We organise this programme around four major themes:
(1) in situ cryo electron tomography, for the determination of macromolecular structures in their native environment and their role in medically important pathways;
(2) high resolution cryoEM with ligand identification and validation, contributing to rational drug design;
(3) heterogeneous reconstruction to identify conformational and/or compositional variability;
(4) data/metadata management to handle the increasing data volumes, and ensure that database entries are well annotated and FAIR.
Through these themes, we begin to address more complex biomedical challenges. It should be noted that AI-based structure prediction techniques are maturing rapidly, providing valuable input, but they are not yet able to tackle the systems studied by cryoEM. Indeed, future AI methods will require well-annotated training data, necessitating more experimental data and improvements in metadata management. At a practical level, the themes will be supported by the CCP-EM software suite which brings together a set of community-developed codes within a unifying framework. An associated training programme of workshops and online resources, as well as other community events, will help disseminate advances across the community.
We have recently released the CCP-EM 2.0 software suite currently comprising 17 diverse software packages and tools developed by ourselves and by our collaborators. These are integrated within our Doppio web-based interface with end-to-end workflow provided by our ccpem-pipeliner. Doppio provides a modern web-based user interface for local and remote processing, enabling researchers to efficiently process and interpret their data. CryoEM has illuminated a wealth of exquisite structures which have increased our understanding of viral pathogens, vaccine development, neurodegenerative fibrils (ex vivo and in situ) and radically expanded the understanding of membrane protein pharmaceutical targets. The themes outlined above will enable structural and cellular biologists to further decipher the molecular action of biomolecules, through ligand interaction, cellular localisation and structural and compositional dynamics. Thus we expect the exponential increase in cryoEM experimental studies to continue. This poses the final challenge we will meet: managing data and metadata effectively for the community. At present ~400Pb of cryoEM data is produced per annum worldwide. To deliver FAIR (findability, accessibility, interoperability, and reusability) data practices, aid individual sites to manage and deposit their data and to enable the next generation of AI-based tools we will build on our experience and network to provide scalable solutions for the community.