Presenter: Aidan Daly Topic: Deconvolution of spatial transcriptomics data using clues from histology images
Biological tissue is best characterized as a patterned arrangement of differentiated cells working in concert to serve a high-level function. As such, understanding how gene expression varies in space is critical to understanding tissue function and dysfunction. Spatial transcriptomics (ST) technologies allow scientists to measure the expression of thousands of genes at once across thinly-sectioned tissues, but modern approaches are limited in their spatial resolution, making it difficult to attribute transcriptional changes to individual cells. In my talk, I will introduce two approaches to “deconvolution” of ST data, or separating transcriptional signals into weighted combinations of signals attributable to individual cell lines: “bottom-up” approaches, which use data from single-cell experiments to infer cell type proportions, and a novel “top-down” approach which observed cell type proportions to estimate cell-specific signals in situ. I will detail how the latter approach can be applied to understand the role of specific cell lines in age-related changes in the mouse colon.
