3DxN - Scalable Multiplexed 3D Tissue Imaging

The aim of the 3DxN project is to enable the analysis of cell morphology and tissue architecture in 3D and quantify protein and transcripts in their spatial context. The convergence of novel methods for imaging a broad range of biomolecules, spatially resolved single cell genomics, and AI applied to digitised tissue images has the potential change the way in which we diagnose disease, assess response to therapy, and manage patients. By correlating tissue samples with other clinical variables and biological analytes it will support the investigation of a wider range of hypotheses.  

We will leverage digitised H&E sections from prostate cancer cases with long-term clinical follow-up data post-diagnosis to identify image features not simply associated with Gleason grading and outcome. To advance our understanding of bone marrow diseases the 3D analysis of the complex features, as for example the structure of the bone marrow stem cell niche, will support the development of robust markers of disease modification in the context of clinical trials. Finally, we will investigate lupus nephritis (LN), the most severe complication of the autoimmune disease systemic lupus erythematosus. Due to demographics and access barriers these groups are often under-represented in research.

Work on this project will address major computational challenges that are associated with the analysis of 3D tissues. We aim to develop new approaches for phenotyping cells and tissue architecture components by integrating morphological features and the quantification of biomolecules imaged through novel machine learning and biomedical image analysis approaches.