ResolVI - addressing noise and bias in spatial transcriptomics

Abstract Technologies for estimating RNA expression at high throughput, in intact tissue slices, and with high spatial resolution (spatial transcriptomics; ST) shed new light on how cells communicate and tissues function. A fundamental step common to all ST protocols is quantification, namely segmenting the plane into regions, each approximating a cell, and then collating the molecules inside each region to estimate the cellular expression profile. Despite many advances in this area, a persisting problem is that of wrong assignment of molecules to cells, which limits most current applications to the level of a priori defined cell subsets and complicates the discovery of novel cell states. Here, we develop resolVI, a model that operates downstream of any segmentation algorithm to generate a probabilistic representation, correcting for misassignment of molecules, as well as for batch effects and other nuisance factors. We demonstrate that resolVI improves our ability to distinguish between cell states, to identify subtle expression changes in space, and to perform integrated analysis across datasets. ResolVI is available as open source software within scvi-tools.