The Comparative Pathology Workbench for image and atlas data including scRNA-Seq and pathology analysis. The Helmsley GCA and the broader HCA resource will deliver single cell transcriptome data spatially organised in terms of gross anatomy, tissue location and with images of cellular histology. This will enable the application of bioinformatics analysis, machine learning and data-mining revealing an atlas of cell types and sub-types and ultimately disease conditions. Here we develop a visual analysis paradigm and associated web-based tools we believe are needed for the clinical domain expert to retrieve, analyse, visualise, and annotate the data and in particular sharing for expert-led collaboration, annotation and analysis. This is particularly important in the context of the histology of normal and diseased tissues within and across species to fully understand human normal and diseased condition with cross-comparison and integration with animal models of the disease.To this end we are extending a “workbench” concept initiated as part of the Centre for Comparative Pathology at Edinburgh University and further developed under CZI funding. For the Helmsley GCA programme we will deliver a working prototype interface for trial within the Helmsley programme and wider HCA as well as applications within comparative pathology.For this we are implementing a focussed data visual-analytics “workbench” to provide the capability for clinical pathologists to analyse, annotate and compare complex histopathological cell-level imaging in conjunction with sequencing data analysis of the same or related tissues. This will provide access to the atlas data in tabular (spread-sheet style) views enabling comparison between individuals, time-course data and cross-species coupled with management of configuration and “state” for tracking, saving and sharing the analysis process.For specific focus, the workbench test context will be normal small intestine and colon and Crohn’s Disease pathological abnormalities affecting the terminal ileum and ascending colon. We will extend our OMERO-based prototype to provide a configurable tabular viewer of multiple “view-cells”. The workbench enables a structured visual-analysis that utilises both human histopathological scan data and other data to interrogate and understand normal physiological and pathological mechanisms. The data-samples and image-data will be annotated in spatial terms using the conceptual atlas framework and ontologies to develop the spatial-description standards needed to annotate both sample-based and image-based data and enable intra and inter-species query and analysis.Important to the design of the workbench is that is web-browser based so that the interface will function on virtually all operating systems and architectures and uses cloud or other systems for easy access from anywhere on the internet. This in turn enables easy sharing between collaborators with the possibility of real-time discussion and editing between remotely located groups. Each “bench” can be dynamically configured and edited both by the owner and by any user of the system given editing permissions, others can be provided with “viewing only” permissions in analogous fashion to the sharing of a Google-document. In addition, each bench has the option of coarse bench-wide and fine cell-specific discussion threads for the purposes of detailed collaborative analysis of data and pathology assignment and annotation.All development of the workbench is fully open-source (managed using GitHub) and where possible building on other supported open-source tools and development environments. Image The Comparative Pathology Workbench Home Page Image A typical "Bench" within the Comparative Pathology Workbench Document Comparative Pathology Workbench - an integrated tool for spatial data annotation and analysis with interfaces for visualisation and querying the data (886.9 KB / PDF) This article was published on 2024-08-27