SpicyR - Spatial analysis of in situ cytometry data

Nicolas Canete⁰, Ellis Patrick⁰
(0) University of Sydney

Find me on Wed Nov 25th, 1:30-2:50pm AEDT in Remo, table 116

Abstract
Highly multiplexed imaging techniques such as cyclic immunofluorescence (CycIF), as well as the mass cytometry based techniques
imaging mass cytometry (IMC) and multiplexed ion beam imaging (MIBI), has allowed many antibody parameters to be visualised
within the same image. These technologies enable a variety of distinct cell types to be analysed concurrently in their native
microenvironment. Indeed, there is increasing evidence that cell microenvironments are programmed not just by cell ontogeny, but
by signals from the microenvironment. Highly multiplexed imaging hence provides the necessary spatial data required to interrogate
the role of the microenvironment and identify any interdependencies between complex cell subsets in health and disease. Standard
image processing, cell segmentation, and cell classification, the phenotype of a single cell and its position within an image can be
identified. However, the analysis of the spatial relationships between these cells can be difficult. Here, we present the statistical use of
marked point process models to identify the spatial relationship between different cell types. Specifically, these models can be used
to identify specific localisations and avoidances between pairwise cell types. This approach can provide insights for immune responses
in cancer models, autoimmune disorders such as type 1 diabetes and multiple sclerosis, and in infectious disease such as HIV. Such models
can prove to be useful for applications in other high parameter disease models.