Abstract
Spatial transcriptomics is a groundbreaking technology that allows the measurement of the expression of thousands of genes in a tissue sample preserving gene spatial locations. This technology has enabled the study of the spatial variation of the genes across the tissue and is becoming an important tool in neurobiology and cancer research. In this talk, I will introduce SpaRTaCo, a statistical model that clusters the spatial expression profiles of the genes according to a partition of the tissue. This is accomplished by performing a co-clustering, that is, inferring the latent block structure of the data and inducing two types of clustering: of the genes, using their expression across the tissue, and of the image areas, using the gene expression in the spots where the RNA is collected. While informative, this co-clustering approach is computationally intensive, undermining its use as an explorative, iterative tool; I will discuss some approximation strategies, relying on nearest-neighbor Gaussian processes and a varying degree of supervision, to improve the computational burden of the approach.
Collegamento Microsoft Teams
Organizzazione
Monica Chiogna