LONDON, UK, August 20 2026: Cancer Research Horizons, the innovation engine of Cancer Research UK, announced today a partnership with Bioptimus, a global AI tech company pioneering the first world model for biology, to advance AI-powered precision medicine. The agreement allows Bioptimus to access S:CORT (Stratification in COloRectal cancer Trial), one of the most comprehensive multimodal colorectal cancer datasets, including over 1,750 patient profiles.
The partnership expands access to Cancer Research Horizons’ growing portfolio of deep, multi-modal, and clinically-linked cancer datasets — spanning multi-omics, imaging, pathology, and beyond. Cancer Research Horizons responsibly makes this data accessible through partnerships designed to help accelerate the development of precision medicine tools and approaches and support the identification of novel targets, with the potential to improve patient outcomes. The collaboration will also increase the number of datasets to STELA (Spatial Tissue Embedding Learning Atlas), Bioptimus' global initiative to build the world's largest clinically linked, multi-modal tissue atlas.
The S:CORT dataset will serve as a key evaluation benchmark, enabling Bioptimus to rigorously assess the performance of its AI models across a range of clinically relevant tasks during iterative model development.
Joe Day, Senior Business Development Executive for Data & AI at Cancer Research Horizons, said: “Cancer Research Horizons is committed to increasing access to world-class cancer research. Through partnerships with companies specialised in advanced analytics and computational biology, we seek to maximise the impact of data for patient benefit. Combining Cancer Research Horizons’ high-quality datasets with AI-driven approaches, such as those employed by Bioptimus, offers the potential to develop new technologies, extract insights at speed, and accelerate research.”
Jean-Philippe Vert, CEO of Bioptimus, said: "Building AI models that genuinely move the needle for cancer patients requires data of exceptional depth and quality. The S:CORT dataset, with its rich combination of imaging, genomic, transcriptomic, and clinical data, is exactly the kind of resource that allows us to rigorously evaluate and continue the development of M-Optimus to deliver real-world clinical advancements. This is an important advancement in our efforts to build the world’s most advanced AI foundation model for biology. We're grateful to Cancer Research Horizons for their commitment to making this data accessible responsibly, and we look forward to demonstrating how the STELA initiative can help break down the data silos that have long slowed the path from scientific discovery to clinical impact.”
Assembled from multiple UK patient cohorts, the S:CORT dataset integrates as many as six distinct data modalities per patient. This multi-modal resource pairs deep, multi-layered molecular and cellular profiling with comprehensive longitudinal clinical outcomes, capturing both the biological complexity and the long-term therapeutic response of the cohort.