A0874
Title: Spatially regularized Gaussian mixtures for clustering spatial transcriptomic data
Authors: Andrea Sottosanti - University of Padova (Italy) [presenting]
Davide Risso - University of Padua (Italy)
Francesco Denti - University of Padua (Italy)
Abstract: Spatial transcriptomics measures the expression of thousands of genes in a tissue sample while preserving its spatial structure. These technologies have enabled the investigation of the spatial variation of gene expressions and their impact on specific biological processes. Identifying genes with similar expression profiles is of utmost importance, thus motivating the development of flexible methods leveraging spatial data structure to cluster genes. A modeling framework is proposed for clustering observations measured over numerous spatial locations via Gaussian processes. Rather than specifying their covariance kernels as a function of the spatial structure, it is used to inform a generalized Cholesky decomposition of their precision matrices. This approach prevents issues with kernel misspecification and facilitates the estimation of a non-stationarity spatial covariance structure. Applied to spatial transcriptomic data, the model identifies gene clusters with distinctive spatial correlation patterns across tissue areas comprising different cell types, like tumor and stromal areas.
