Diamond Annual Review 2025-26

Over the past year, substantial progress has been made in building the core platforms and infrastructure to ensure the facility is ready. Software development has matured, with useful capabilities already emerging ahead of the dark period. Workshops and away days have helped science and software teams explore opportunities and challenges in adopting these new systems. Following deployment to the ViSR beamline, components are now being integrated into I15-1 to enhance and automate their mail-in programme. Imaging and macromolecular crystallography have also started to benefit from improved pipelines and web- based tools. The department is organised into five specialised groups: Scientific Computing, Data Analysis, Data Acquisition, Beamline Controls, and Scientific Information Management Systems. This structure is carefully designed to emphasise the critical role of each area in delivering and supporting robust software, computing, and control solutions, ensuring the highest standards of service for the facility. This reflects that software and computing services are an essential driver of the facility’s operations and its scientific productivity. Diamond-II core software Diamond-II requires a step change in software, controls and computing to support future scientific delivery. The Scientific Software, Controls and Computation (SSCC) department is responsible for overseeing all software, computing, and control systems that underpin and advance Diamond’s scientific programme. Scientific Software, Controls and Computation 29 Anti-collision modelling for beamlines We have developed tools for modelling and visualising robotic systems with a focus on collision awareness and spatial constraints. The software is written in Python and Java Script to represent instrument geometry, using imported CAD derived STL models, alongside kinematic descriptions and visualisation tools. An interactive web- based environment with virtual and augmented reality is included. The software enables offline exploration of robot and instrument motion, workspace limits, and potential collision scenarios, supporting safer deployment and informed system design. By combining geometry, simulation, and visualisation in a lightweight framework, this approach provides a practical foundation for integrating anti-collision modelling into Diamond’s control and acquisition system paving the way for physical AI and the automation of complex experiments. New ULIMS services Universal LIMS (Laboratory Information Management System) is a new set of web services being developed as part of the Diamond-II upgrade. It will provide users and beamline scientists with tools to manage the logistics and scientific metadata for their experiments. Over the past year development of these services has continued, with minimum viable products of the Data Catalogue and Sample Information services now complete, and Experiment Definition and Container Logistics in progress. The Universal LIMS team has been closely involved in the I15-1 mail-in project to allow users to provide the information on how to run their samples on the beamline, and automatic data ingest has been set-up in the Data Catalogue for B18. AI opportunities AI offers immense opportunities from the orchestration of complex experiments to the efficient management of datasets, and sophisticated analysis of scientific results. Recognising the transformative potential of AI, Diamond is proactively developing newAI-powered capabilities. Over the past year, Diamond secured £2.5 million in funding to address future computational and storage requirements for AI, alongside £8 million dedicated to pioneering work in protein crystallography – actively showcasing our expertise and achievements in this field. In the coming year, we will deliver these advantages to the UK user community and beyond, through the provision of sovereign capabilities, open and curated datasets, and a concerted effort to enhance AI integration across Diamond as a whole. Visualising a robot model alongside the physical robot in a beamline experiment hutch 30 Annual review 2025/26 Scientific Software, Controls and Computation