Structural biology training platforms for experimental and predictive approaches

Structural biology provides mechanistic insights and understanding of the molecular machines that underpin all biological processes in living cells. From the 3D structure of macromolecules we can unravel the mechanistic details of how drugs bind their targets, enzymes catalyse important reactions, proteins replicate the very DNA that encodes all of life. There are fantastic structural biology training resources online, found in detailed online forums, delivered in specialised workshops by established community initiatives and through in-person knowledge sharing. These learning pathways have been successful in training generations of researchers, particularly in the core experimental structural biology techniques that have been centre stage for over 40 years. It is now required that the scale and specialism of available training be developed in line with new technologies in experimental and predictive structural biology. In experimental structural biology, the landmark increase in the utility of cryoEM and continued maturation of X-ray crystallography methods further places these techniques as tools rather than specialisms that researchers use to understand biology. The scale and quality of structural data in the archives can be used to improve structural biology processes themselves. In particular though, in predictive structural biology the development of AlphaFold was made possible by referencing the decades worth of archived macromolecular structures in the PDB. Prediction methods now provide the opportunity for all scientists in all fields to utilise macromolecular structures in asking and answering biological questions. Thus experimental and predictive structural biology work together, supporting the scaling of data generation and helping to drive the next generation of predictive structural biology tools. We are at an important moment to encourage best practices in the use of structural biology experimental and predictive data, and to raise awareness to correctly archive data that will inform the next generation of predictive structural biology tools transformative to science.
 
Diamond Light Source is a key driver for throughput and technologies in experimental structural biology, while CCP4 and CCP-EM develop essential software management tools and analysis pipelines for x-ray crystallography and cryoEM data. EMBL-EBI hosts the Protein Databank in Europe (PDBe) and Electron Microscopy Databank (EMDB), and contributes to the major worldwide effort of the wwPDB to curate structural biology data deposited into the databases. All are active in delivering training to their user communities, but there is an opportunity to unite training resources on data generation, analysis, deposition and curation with a view to highlight the opportunities for data reuse that can drive technical innovation and prediction tools in structural biology. This project will achieve this by knowledge and instructor exchange between existing specialised workshops. We will then design learning pathways on best practices in experimental and predictive structural biology. Finally, we will create an online learning series which presents case studies in X-ray, cryoEM and predictive structural biology, to highlight the practices of data handling and reuse. These case studies will exemplify individual scientific innovations but also present these on a learning pathway to highlight data reuse opportunity. By drawing attention to the current aspects of archived data that have driven predictive and experimental structural biology advances, we will raise awareness of best practices leading to reusable data while further highlighting the future opportunities for data reuse.