Spatially resolved omics technologies are revolutionising our understanding of biological tissues by introducing new ways to characterise tissue architectures and identify cell-cell interactions. However, handling spatial omics datasets remains challenging due to the large heterogeneity of profiling technologies, resulting in fragmented file formats, and the plethora of arbitrary ways to store analysis results further exacerbates this fragmentation. Additionally, the data volume often exceeds the memory capacity of standard computing environments. These challenges make it difficult to construct scalable, interoperable workflows and to share them reproducibly with the scientific community.

In this webinar, we will present a software solution—the SpatialData framework—that we developed in Python to address these data representation challenges. Specifically, we will demonstrate how our software infrastructure can be used to:

• Represent data from the most commonly available assays in a unified way
• Manipulate, query, and compute statistics on the data
• Visualise and annotate the data interactively
• Prepare the data for sharing to maximise interoperability