Collaborative Computational Project for Electron cryo-Microscopy (CCP-EM): 2021 - 2026

The behaviour of living systems ultimately comes down to the interactions of biological molecules inside cells, and understanding these is vital to numerous human efforts, including controlling disease and improving food production. While experimental techniques such as macromolecular crystallography have for many years given detailed information on important molecules in the cell, many classes of molecules are not amenable to this technique. Moreover, as our understanding of pathways in the cell grows, there is increasing interest in the context in which these molecules operate. In other words, where in the cell do these molecules do their job, and which other cellular components are necessary for their function. Electron cryo-Microscopy (cryoEM) provides very useful information here, and bridges the gap between individual molecules and the whole cell. In the most favourable cases, detailed images of assemblies of molecules can be obtained, while at lower resolutions electron tomograms can show internal molecular details from within intact parts of cells or tissues.

Advances in instrumentation and data processing led to a significant increase in the quality of cryoEM data, which was characterised in 2014 as the "Resolution Revolution", and recognised by the 2017 Nobel Prize in Chemistry. Consequently there has been a surge in interest in the technique from structural and cellular biologists trying to understand a wide range of biological systems. There has been significant investment in the research infrastructure supporting cryoEM, most notably the establishment of several electron microscope facilities around the country. In the last few years, pharmaceutical companies and biotechnology companies have recognised the importance of cryoEM to their discovery pipelines, and have also begun investing in the area. A key component of this research infrastructure is the computational support to manage the data, process the micrographs, and interpret the data in terms of molecular volumes and/or atomic structures. The Collaborative Computational Project for Electron cryo-Microscopy (CCP-EM) was established during the period 2012 - 2016 to provide this part of the research infrastructure.

The proposed project is intended to provide continued support to the cryoEM community. One of the major products of the CCP-EM partnership is a software suite for processing cryoEM data collected at microscope facilities. Individual computer programs in this suite are developed independently, either by members of CCP-EM or collaborators. The role of CCP-EM is to collate these programs into a single suite, develop workflows through the suite, and distribute the suite to practising scientists. When done well, this is a win-win arrangement in which scientists get access to a comprehensive set of software in one place, and methods developers get access to a large user base. It is well known, however, that software rapidly becomes unusable if not actively maintained and it is the responsibility of the core team of CCP-EM to ensure the longevity of software in the suite.

We will also expand the scope of the suite. We will improve the tools for validating the structural information obtained, and facilitate the deposition of data in international archives. We will help to drive FAIR principles - that data from cryoEM experiments are accessible and usable to the wider community. We will increase our support for sub-tomogram averaging, a particular technique for obtaining in situ structural information of molecules. Finally, we will make more use of machine learning i.e. advanced algorithms that can learn from the data.

All these advances will be tightly coupled with our on-going user training programme, and support for individual methods developers. We will also continue our very popular annual Spring Symposium, which now provides a forum for 300 researchers to share experiences and to develop the cryoEM community.

The CCP-EM partnership exists to further strengthen and expand the cryo-EM community. Activities can be broadly divided into three areas: (1) development of the CCP-EM software suite, (2) training programme and community events, and (3) strategic initiatives and outreach. We request funding for a core team of computational scientists to coordinate and develop these activities. Since software development, training and outreach are closely linked, we expect the team to work together in all areas.

The software suite is designed to integrate individual programs from diverse sources into a convenient package for scientists. It provides a framework for collaborating with other developers and helping to make novel methods available to the community. Collaboration is required, and the suite is not intended to be a container for all available software. The core team is developing a set of Python libraries covering a data model, job control, scheduling and workflow definition, which will merge the RELION pipeliner and the existing CCP-EM project manager. We propose to continue this development, integrate it into the production version of RELION, and expand its use to the rest of CCP-EM, facilitating closer integration between reconstruction, model building and data validation.

The extended framework will be used to integrate subtomogram averaging functionality, and develop the link to tomographic reconstruction. We will also develop on-the-fly processing for eBIC (and other facilities), tools for validation and simplified EMDB deposition, automation of data processing workflows and a new combined GUI for CCP-EM and RELION, as well as allowing third-party plugins to be easily incorporated into the pipeline. Machine-learning methods for data-driven analysis will be embedded at all levels.

We will continue to organise the annual Spring Symposium, expand the training programme, contribute to international meetings, and organise hackathons for developers.

There is widespread interest in the international research community in using cryoEM to tackle big scientific challenges, such as understanding molecular machines in action and membrane complexes in situ. The technological advances enabling this were recognised in the award of the 2017 Nobel Prize in Chemistry. The field has strong links to atomic structure methods, especially crystallography, while electron tomography has strong connections to cell biology. The spatial dimension is needed to understand biological networks and machines, and complements traditional systems biology approaches. By supporting cryoEM research groups in the UK and encouraging a collaborative effort, CCP-EM advances the usage of cryoEM, and has an impact on many structural biology projects.

CCP-EM has an impact on individual researchers through its training and knowledge exchange aspects. The expansion of cryoEM as an important component in the toolkit for understanding cellular and sub-cellular biology relies on the availability of researchers competent in the computational techniques required to interpret the data. Existing researchers will of course benefit through improved software tools and environment. However, we specifically wish to lower the barriers to entry into the field of cryoEM. This applies not only to students and young postdocs, but also to researchers moving from other fields or wishing to use cryoEM as an additional technique.

The problems being addressed by cryoEM are of major importance in biomedical science, e.g. dynamic systems involving protein folding/refolding/misfolding, important in neurodegeneration and other misfolding diseases, virus-host interactions, and drug binding studies. Although our primary focus in the Partnership is on the fundamental science, advances here ultimately have an impact on translational and medical research. We are focussing on a technique rather than a particular scientific area, so the impact is likely to be very broad, leading eventually to improved medicine and health for the nation.

The insight that cryoEM provides to disease mechanisms is attracting the attention of a number of pharmaceutical companies. Around a dozen companies have purchased licences to use the CCP-EM software suite, and this number is growing. Company R&D staff also regularly attend our events. As well as helping to elucidate the underlying mechanisms of disease, cryoEM allows scientists to visualise the effect of drug molecules on proteins and complexes, in a near-native environment. As the technique matures, it is likely to become part of the drug development pipeline. Other industries (biotechnology, agribusiness) will also potentially benefit from the extra insight into biological processes provided by cryoEM.

By uniting the UK cryoEM community, the Partnership will have an impact on the strategy for future developments. By acting together, the UK cryoEM community will have a stronger voice. This will be used to influence software developers, instrument manufacturers and data standards development. They will also be able to provide input for national and international policy makers and funders. Finally, CCP-EM will take part in public engagement events to help strengthen the public understanding and appreciation of biomedical research.